diff --git a/docs6/docs/file-formats.md b/docs6/docs/file-formats.md
index df33e12db..551b3449e 100644
--- a/docs6/docs/file-formats.md
+++ b/docs6/docs/file-formats.md
@@ -547,14 +547,41 @@ While you can do format conversion using `mlr --icsv --ojson cat myfile.csv`, th
 <b>mlr help format-conversion</b>
 </pre>
 <pre class="pre-non-highlight-in-pair">
-TO DO: brief list of formats w/ xref to m6 webdocs.
-
-Examples: --csv for CSV-formatted input and output; --icsv --opprint for
-CSV-formatted input and pretty-printed output.
-
-Please use --iformat1 --oformat2 rather than --format1 --oformat2.
-The latter sets up input and output flags for format1, not all of which
-are overridden in all cases by setting output format to format2.
+Type 'mlr help {topic}' for any of the following:
+Essentials:
+  mlr help topics
+  mlr help basic-examples
+  mlr help data-formats
+Flags:
+  mlr help flags
+Verbs:
+  mlr help list-verbs
+  mlr help usage-verbs
+  mlr help verb
+Functions:
+  mlr help list-functions
+  mlr help list-function-classes
+  mlr help list-functions-in-class
+  mlr help usage-functions
+  mlr help usage-functions-by-class
+  mlr help function
+Keywords:
+  mlr help list-keywords
+  mlr help usage-keywords
+  mlr help keyword
+Other:
+  mlr help auxents
+  mlr help mlrrc
+  mlr help output-colorization
+  mlr help type-arithmetic-info
+Shorthands:
+  mlr -g = mlr help flags
+  mlr -l = mlr help list-verbs
+  mlr -L = mlr help usage-verbs
+  mlr -f = mlr help list-functions
+  mlr -F = mlr help usage-functions
+  mlr -k = mlr help list-keywords
+  mlr -K = mlr help usage-keywords
 </pre>
 
 <!---
@@ -583,21 +610,41 @@ You can include comments within your data files, and either have them ignored, o
 <b>mlr help comments-in-data</b>
 </pre>
 <pre class="pre-non-highlight-in-pair">
-Miller lets you put comments in your data, such as
-
-    # This is a comment for a CSV file
-    a,b,c
-    1,2,3
-    4,5,6
-
-Notes:
-* Comments are only honored at the start of a line.
-* In the absence of any of the below four options, comments are data like
-  any other text. (The comments-in-data feature is opt-in.)
-* When `--pass-comments` is used, comment lines are written to standard output
-  immediately upon being read; they are not part of the record stream.  Results
-  may be counterintuitive. A suggestion is to place comments at the start of
-  data files.
+Type 'mlr help {topic}' for any of the following:
+Essentials:
+  mlr help topics
+  mlr help basic-examples
+  mlr help data-formats
+Flags:
+  mlr help flags
+Verbs:
+  mlr help list-verbs
+  mlr help usage-verbs
+  mlr help verb
+Functions:
+  mlr help list-functions
+  mlr help list-function-classes
+  mlr help list-functions-in-class
+  mlr help usage-functions
+  mlr help usage-functions-by-class
+  mlr help function
+Keywords:
+  mlr help list-keywords
+  mlr help usage-keywords
+  mlr help keyword
+Other:
+  mlr help auxents
+  mlr help mlrrc
+  mlr help output-colorization
+  mlr help type-arithmetic-info
+Shorthands:
+  mlr -g = mlr help flags
+  mlr -l = mlr help list-verbs
+  mlr -L = mlr help usage-verbs
+  mlr -f = mlr help list-functions
+  mlr -F = mlr help usage-functions
+  mlr -k = mlr help list-keywords
+  mlr -K = mlr help usage-keywords
 </pre>
 
 Examples:
diff --git a/docs6/docs/manpage.md b/docs6/docs/manpage.md
index 6516533cb..3dc77a951 100644
--- a/docs6/docs/manpage.md
+++ b/docs6/docs/manpage.md
@@ -28,9 +28,9 @@ NAME
        as CSV and tabular JSON.
 
 SYNOPSIS
-       Usage: mlr [I/O options] {verb} [verb-dependent options ...] {zero or
-       more file names} Output of one verb may be chained as input to another
-       using "then", e.g.
+       Usage: mlr [flags] {verb} [verb-dependent options ...] {zero or more
+       file names} Output of one verb may be chained as input to another using
+       "then", e.g.
          mlr stats1 -a min,mean,max -f flag,u,v -g color then sort -f color
        Please see 'mlr help topics' for more information.  Please also see
        https://johnkerl.org/miller6
@@ -116,6 +116,43 @@ DATA FORMATS
        | fox jumped          | Record 2: "1":"fox", "2":"jumped"
        +---------------------+
 
+HELP OPTIONS
+       Type 'mlr help {topic}' for any of the following:
+       Essentials:
+         mlr help topics
+         mlr help basic-examples
+         mlr help data-formats
+       Flags:
+         mlr help flags
+       Verbs:
+         mlr help list-verbs
+         mlr help usage-verbs
+         mlr help verb
+       Functions:
+         mlr help list-functions
+         mlr help list-function-classes
+         mlr help list-functions-in-class
+         mlr help usage-functions
+         mlr help usage-functions-by-class
+         mlr help function
+       Keywords:
+         mlr help list-keywords
+         mlr help usage-keywords
+         mlr help keyword
+       Other:
+         mlr help auxents
+         mlr help mlrrc
+         mlr help output-colorization
+         mlr help type-arithmetic-info
+       Shorthands:
+         mlr -g = mlr help flags
+         mlr -l = mlr help list-verbs
+         mlr -L = mlr help usage-verbs
+         mlr -f = mlr help list-functions
+         mlr -F = mlr help usage-functions
+         mlr -k = mlr help list-keywords
+         mlr -K = mlr help usage-keywords
+
 VERB LIST
        altkv bar bootstrap cat check clean-whitespace count-distinct count
        count-similar cut decimate fill-down fill-empty filter flatten format-values
@@ -149,117 +186,171 @@ FUNCTION LIST
        version ! != !=~ % & && * ** + - . .* .+ .- ./ / // &lt; &lt;&lt; &lt;= == =~ &gt; &gt;= &gt;&gt; &gt;&gt;&gt;
        ?: ?? ??? ^ ^^ | || ~
 
-HELP OPTIONS
-       Type 'mlr help {topic}' for any of the following:
-         mlr help topics
-         mlr help auxents
-         mlr help basic-examples
-         mlr help comments-in-data
-         mlr help compressed-data
-         mlr help csv-options
-         mlr help data-format-options
-         mlr help data-formats
-         mlr help double-quoting
-         mlr help format-conversion
-         mlr help function
-         mlr help keyword
-         mlr help list-functions
-         mlr help list-function-classes
-         mlr help list-functions-in-class
-         mlr help list-functions-as-paragraph
-         mlr help list-functions-as-table
-         mlr help list-keywords
-         mlr help list-keywords-as-paragraph
-         mlr help list-verbs
-         mlr help list-verbs-as-paragraph
-         mlr help misc
-         mlr help mlrrc
-         mlr help number-formatting
-         mlr help output-colorization
-         mlr help separator-options
-         mlr help type-arithmetic-info
-         mlr help usage-functions
-         mlr help usage-functions-by-class
-         mlr help usage-keywords
-         mlr help usage-verbs
-         mlr help verb
-       Shorthands:
-         mlr -l = mlr help list-verbs
-         mlr -L = mlr help usage-verbs
-         mlr -f = mlr help list-functions
-         mlr -F = mlr help usage-functions
-         mlr -k = mlr help list-keywords
-         mlr -K = mlr help usage-keywords
+COMMENTS-IN-DATA FLAGS
+       Miller lets you put comments in your data, such as
 
-OPTIONS
-       In the following option flags, the version with "i" designates the
-       input stream, "o" the output stream, and the version without prefix
-       sets the option for both input and output stream. For example: --irs
-       sets the input record separator, --ors the output record separator, and
-       --rs sets both the input and output separator to the given value.
+           # This is a comment for a CSV file
+           a,b,c
+           1,2,3
+           4,5,6
 
-   DATA-FORMAT OPTIONS
-       --idkvp   --odkvp   --dkvp      Delimited key-value pairs, e.g "a=1,b=2"
-                                        (Miller's default format).
+       Notes:
 
-       --inidx   --onidx   --nidx      Implicitly-integer-indexed fields (Unix-toolkit style).
-       -T                              Synonymous with "--nidx --fs tab".
+       * Comments are only honored at the start of a line.
+       * In the absence of any of the below four options, comments are data like
+         any other text. (The comments-in-data feature is opt-in.)
+       * When `--pass-comments` is used, comment lines are written to standard output
+         immediately upon being read; they are not part of the record stream.  Results
+         may be counterintuitive. A suggestion is to place comments at the start of
+         data files.
 
-       --icsv    --ocsv    --csv       Comma-separated value (or tab-separated with --fs tab, etc.)
+       --pass-comments          Immediately print commented lines (prefixed by `#`)
+                                within the input.
+       --pass-comments-with {string}
+                                Immediately print commented lines within input, with
+                                specified prefix.
+       --skip-comments          Ignore commented lines (prefixed by `#`) within the
+                                input.
+       --skip-comments-with {string}
+                                Ignore commented lines within input, with specified
+                                prefix.
 
-       --itsv    --otsv    --tsv       Keystroke-savers for "--icsv --ifs tab",
-                                       "--ocsv --ofs tab", "--csv --fs tab".
-       --iasv    --oasv    --asv       Similar but using ASCII FS 0x1f and RS 0x1e\n",
-       --iusv    --ousv    --usv       Similar but using Unicode FS U+241F (UTF-8 0xe2909f)\n",
-                                       and RS U+241E (UTF-8 0xe2909e)\n",
+COMPRESSED-DATA FLAGS
+       Miller offers a few different ways to handle reading data files which have been compressed.
 
-       --icsvlite --ocsvlite --csvlite Comma-separated value (or tab-separated with --fs tab, etc.).
-                                              The 'lite' CSV does not handle RFC-CSV double-quoting rules; is
-                                              slightly faster and handles heterogeneity in the input stream via
-                                              empty newline followed by new header line. See also
-                                               https://johnkerl.org/miller6/file-formats.html#csv-tsv-asv-usv-etc
+       * Decompression done within the Miller process itself: `--bz2in` `--gzin` `--zin`
+       * Decompression done outside the Miller process: `--prepipe` `--prepipex`
 
-       --itsvlite --otsvlite --tsvlite Keystroke-savers for "--icsvlite --ifs tab",
-                                       "--ocsvlite --ofs tab", "--csvlite --fs tab".
-       -t                              Synonymous with --tsvlite.
-       --iasvlite --oasvlite --asvlite Similar to --itsvlite et al. but using ASCII FS 0x1f and RS 0x1e\n",
-       --iusvlite --ousvlite --usvlite Similar to --itsvlite et al. but using Unicode FS U+241F (UTF-8 0xe2909f)\n",
-                                       and RS U+241E (UTF-8 0xe2909e)\n",
+       Using `--prepipe` and `--prepipex` you can specify an action to be
+       taken on each input file.  The prepipe command must be able to read from
+       standard input; it will be invoked with `{command} &lt; {filename}`.  The
+       prepipex command must take a filename as argument; it will be invoked with
+       `{command} {filename}`.
 
-       --ipprint --opprint --pprint    Pretty-printed tabular (produces no
-                                       output until all input is in).
-                           --right     Right-justifies all fields for PPRINT output.
-                           --barred    Prints a border around PPRINT output
-                                       (only available for output).
+       Examples:
 
-                 --omd                 Markdown-tabular (only available for output).
+           mlr --prepipe gunzip
+           mlr --prepipe zcat -cf
+           mlr --prepipe xz -cd
+           mlr --prepipe cat
 
-       --ixtab   --oxtab   --xtab      Pretty-printed vertical-tabular.
-                           --xvright   Right-justifies values for XTAB format.
+       Note that this feature is quite general and is not limited to decompression
+       utilities. You can use it to apply per-file filters of your choice.  For output
+       compression (or other) utilities, simply pipe the output:
+       `mlr ... | {your compression command} &gt; outputfilenamegoeshere`
 
-       --ijson   --ojson   --json      JSON tabular: sequence or list of one-level
-                                       maps: {...}{...} or [{...},{...}].
-                           --jvstack   Put one key-value pair per line for JSON output.
-                        --no-jvstack   Put objects/arrays all on one line for JSON output.
-                     --jsonx --ojsonx  Keystroke-savers for --json --jvstack
-                     --jsonx --ojsonx  and --ojson --jvstack, respectively.
-                           --jlistwrap Wrap JSON output in outermost [ ].
-                   --flatsep {string} Separator for flattening multi-level JSON keys,
-                                       e.g. '{"a":{"b":3}}' becomes a:b =&gt; 3 for
-                                       non-JSON formats. Defaults to ..\n",
+       Lastly, note that if `--prepipe` or `--prepipex` is specified, it replaces any
+       decisions that might have been made based on the file suffix. Likewise,
+       `--gzin`/`--bz2in`/`--zin` are ignored if `--prepipe` is also specified.
 
-       -p is a keystroke-saver for --nidx --fs space --repifs
+       --bz2in                  Uncompress bzip2 within the Miller process. Done by
+                                default if file ends in `.bz2`.
+       --gzin                   Uncompress gzip within the Miller process. Done by
+                                default if file ends in `.gz`.
+       --prepipe {decompression command}
+                                You can, of course, already do without this for
+                                single input files, e.g. `gunzip &lt; myfile.csv.gz |
+                                mlr ...`. Allowed at the command line, but not in
+                                `.mlrrc` to avoid unexpected code execution.
+       --prepipe-bz2            Same as `--prepipe bz2`, except this is allowed in
+                                `.mlrrc`.
+       --prepipe-gunzip         Same as `--prepipe gunzip`, except this is allowed in
+                                `.mlrrc`.
+       --prepipe-zcat           Same as `--prepipe zcat`, except this is allowed in
+                                `.mlrrc`.
+       --prepipex {decompression command}
+                                Like `--prepipe` with one exception: doesn't insert
+                                `&lt;` between command and filename at runtime. Useful
+                                for some commands like `unzip -qc` which don't read
+                                standard input. Allowed at the command line, but not
+                                in `.mlrrc` to avoid unexpected code execution.
+       --zin                    Uncompress zlib within the Miller process. Done by
+                                default if file ends in `.z`.
 
-       Examples: --csv for CSV-formatted input and output; --icsv --opprint for
+CSV-ONLY FLAGS
+       --allow-ragged-csv-input or --ragged
+                                If a data line has fewer fields than the header line,
+                                fill remaining keys with empty string. If a data line
+                                has more fields than the header line, use integer
+                                field labels as in the implicit-header case.
+       --headerless-csv-output  Print only CSV data lines; do not print CSV header
+                                lines.
+       --implicit-csv-header    Use 1,2,3,... as field labels, rather than from line
+                                1 of input files. Tip: combine with `label` to
+                                recreate missing headers.
+       --no-implicit-csv-header Opposite of `--implicit-csv-header`. This is the
+                                default anyway -- the main use is for the flags to
+                                `mlr join` if you have main file(s) which are
+                                headerless but you want to join in on a file which
+                                does have a CSV header. Then you could use `mlr --csv
+                                --implicit-csv-header join --no-implicit-csv-header
+                                -l your-join-in-with-header.csv ...
+                                your-headerless.csv`.
+       -N                       Keystroke-saver for `--implicit-csv-header
+                                --headerless-csv-output`.
+
+FILE-FORMAT FLAGS
+       TO DO: brief list of formats w/ xref to m6 webdocs.
+
+       Examples: `--csv` for CSV-formatted input and output; `--icsv --opprint` for
        CSV-formatted input and pretty-printed output.
 
-       Please use --iformat1 --oformat2 rather than --format1 --oformat2.
-       The latter sets up input and output flags for format1, not all of which
-       are overridden in all cases by setting output format to format2.
+       Please use `--iformat1 --oformat2` rather than `--format1 --oformat2`.
+       The latter sets up input and output flags for `format1`, not all of which
+       are overridden in all cases by setting output format to `format2`.
 
-   FORMAT-CONVERSION KEYSTROKE-SAVERS
+       --asv or --asvlite       Use ASV format for input and output data.
+       --csv or -c              Use CSV format for input and output data.
+       --csvlite                Use CSV-lite format for input and output data.
+       --dkvp                   Use DKVP format for input and output data.
+       --iasv or --iasvlite     Use ASV format for input data.
+       --icsv                   Use CSV format for input data.
+       --icsvlite               Use CSV-lite format for input data.
+       --idkvp                  Use DKVP format for input data.
+       --ijson                  Use JSON format for input data.
+       --inidx                  Use NIDX format for input data.
+       --io {format name}       Use format name for input and output data. For
+                                example: `--io csv` is the same as `--csv`.
+       --ipprint                Use PPRINT format for input data.
+       --itsv                   Use TSV format for input data.
+       --itsvlite               Use TSV-lite format for input data.
+       --iusv or --iusvlite     Use USV format for input data.
+       --ixtab                  Use XTAB format for input data.
+       --json or -j             Use JSON format for input and output data.
+       --nidx                   Use NIDX format for input and output data.
+       --oasv or --oasvlite     Use ASV format for output data.
+       --ocsv                   Use CSV format for output data.
+       --ocsvlite               Use CSV-lite format for output data.
+       --odkvp                  Use DKVP format for output data.
+       --ojson                  Use JSON format for output data.
+       --omd                    Use markdown-tabular format for output data.
+       --onidx                  Use NIDX format for output data.
+       --opprint                Use PPRINT format for output data.
+       --otsv                   Use TSV format for output data.
+       --otsvlite               Use TSV-lite format for output data.
+       --ousv or --ousvlite     Use USV format for output data.
+       --oxtab                  Use XTAB format for output data.
+       --pprint                 Use PPRINT format for input and output data.
+       --tsv                    Use TSV format for input and output data.
+       --tsvlite or -t          Use TSV-lite format for input and output data.
+       --usv or --usvlite       Use USV format for input and output data.
+       --xtab                   Use XTAB format for input and output data.
+       -i {format name}         Use format name for input data. For example: `-i csv`
+                                is the same as `--icsv`.
+       -o {format name}         Use format name for output data. For example: `-o
+                                csv` is the same as `--ocsv`.
+
+FLATTEN-UNFLATTEN FLAGS
+       --flatsep or --jflatsep or --oflatsep {string}
+                                Separator for flattening multi-level JSON keys, e.g.
+                                `{"a":{"b":3}}` becomes `a:b =&gt; 3` for non-JSON
+                                formats. Defaults to `.`.
+       --no-auto-flatten
+       --no-auto-unflatten
+
+FORMAT-CONVERSION KEYSTROKE-SAVER FLAGS
        As keystroke-savers for format-conversion you may use the following:
-       --c2t --c2d --c2n --c2j --c2x --c2p --c2m
+             --c2t --c2d --c2n --c2j --c2x --c2p --c2m
        --t2c       --t2d --t2n --t2j --t2x --t2p --t2m
        --d2c --d2t       --d2n --d2j --d2x --d2p --d2m
        --n2c --n2t --n2d       --n2j --n2x --n2p --n2m
@@ -270,130 +361,270 @@ OPTIONS
        PPRINT, and markdown, respectively. Note that markdown format is available for
        output only.
 
-   SEPARATORS
-       THIS IS STILL TBD FOR MILLER 6
+       --c2b                    Use CSV for input, PPRINT with `--barred` for output.
+       --c2d                    Use CSV for input, DKVP for output.
+       --c2j                    Use CSV for input, JSON for output.
+       --c2m                    Use CSV for input, markdown-tabular for output.
+       --c2n                    Use CSV for input, NIDX for output.
+       --c2p                    Use CSV for input, PPRINT for output.
+       --c2t                    Use CSV for input, TSV for output.
+       --c2x                    Use CSV for input, XTAB for output.
+       --d2b                    Use DKVP for input, PPRINT with `--barred` for
+                                output.
+       --d2c                    Use DKVP for input, CSV for output.
+       --d2j                    Use DKVP for input, JSON for output.
+       --d2m                    Use DKVP for input, markdown-tabular for output.
+       --d2n                    Use DKVP for input, NIDX for output.
+       --d2p                    Use DKVP for input, PPRINT for output.
+       --d2t                    Use DKVP for input, TSV for output.
+       --d2x                    Use DKVP for input, XTAB for output.
+       --j2b                    Use JSON for input, PPRINT with --barred for output.
+       --j2c                    Use JSON for input, CSV for output.
+       --j2d                    Use JSON for input, DKVP for output.
+       --j2m                    Use JSON for input, markdown-tabular for output.
+       --j2n                    Use JSON for input, NIDX for output.
+       --j2p                    Use JSON for input, PPRINT for output.
+       --j2t                    Use JSON for input, TSV for output.
+       --j2x                    Use JSON for input, XTAB for output.
+       --n2b                    Use NIDX for input, PPRINT with `--barred` for
+                                output.
+       --n2c                    Use NIDX for input, CSV for output.
+       --n2d                    Use NIDX for input, DKVP for output.
+       --n2j                    Use NIDX for input, JSON for output.
+       --n2m                    Use NIDX for input, markdown-tabular for output.
+       --n2p                    Use NIDX for input, PPRINT for output.
+       --n2t                    Use NIDX for input, TSV for output.
+       --n2x                    Use NIDX for input, XTAB for output.
+       --p2c                    Use PPRINT for input, CSV for output.
+       --p2d                    Use PPRINT for input, DKVP for output.
+       --p2j                    Use PPRINT for input, JSON for output.
+       --p2m                    Use PPRINT for input, markdown-tabular for output.
+       --p2n                    Use PPRINT for input, NIDX for output.
+       --p2t                    Use PPRINT for input, TSV for output.
+       --p2x                    Use PPRINT for input, XTAB for output.
+       --t2b                    Use TSV for input, PPRINT with `--barred` for output.
+       --t2c                    Use TSV for input, CSV for output.
+       --t2d                    Use TSV for input, DKVP for output.
+       --t2j                    Use TSV for input, JSON for output.
+       --t2m                    Use TSV for input, markdown-tabular for output.
+       --t2n                    Use TSV for input, NIDX for output.
+       --t2p                    Use TSV for input, PPRINT for output.
+       --t2x                    Use TSV for input, XTAB for output.
+       --x2b                    Use XTAB for input, PPRINT with `--barred` for
+                                output.
+       --x2c                    Use XTAB for input, CSV for output.
+       --x2d                    Use XTAB for input, DKVP for output.
+       --x2j                    Use XTAB for input, JSON for output.
+       --x2m                    Use XTAB for input, markdown-tabular for output.
+       --x2n                    Use XTAB for input, NIDX for output.
+       --x2p                    Use XTAB for input, PPRINT for output.
+       --x2t                    Use XTAB for input, TSV for output.
+       -p                       Keystroke-saver for `--nidx --fs space --repifs`.
+       -T                       Keystroke-saver for `--nidx --fs tab`.
 
-   COMPRESSED I/O
-       Decompression done within the Miller process itself:
-       --gzin  Uncompress gzip within the Miller process. Done by default if file ends in ".gz".
-       --bz2in Uncompress bz2ip within the Miller process. Done by default if file ends in ".bz2".
-       --zin   Uncompress zlib within the Miller process. Done by default if file ends in ".z".
+JSON-ONLY FLAGS
+       These are flags which are applicable to JSON format.
 
-       Decompression done outside the Miller process:
-       --prepipe {command} You can, of course, already do without this for single input files,
-         e.g. "gunzip &lt; myfile.csv.gz | mlr ..."
-       --prepipex {command} Like --prepipe with one exception: doesn't insert '&lt;' between
-         command and filename at runtime. Useful for some commands like 'unzip -qc'
-         which don't read standard input.
+       --jlistwrap or --jl      Wrap JSON output in outermost `[ ]`.
+       --jvstack                Put one key-value pair per line for JSON output
+                                (multi-line output).
+       --no-jvstack             Put objects/arrays all on one line for JSON output.
 
-       Using --prepipe and --prepipex you can specify an action to be taken on each
-       input file. This prepipe command must be able to read from standard input; it
-       will be invoked with {command} &lt; {filename}.
+LEGACY FLAGS
+       These are flags which don't do anything in the current Miller version.
+       They are accepted as no-op flags in order to keep old scripts from breaking.
 
-       Examples:
-         mlr --prepipe gunzip
-         mlr --prepipe zcat -cf
-         mlr --prepipe xz -cd
-         mlr --prepipe cat
+       --jknquoteint            Type information from JSON input files is now
+                                preserved throughout the processing stream.
+       --jquoteall              Type information from JSON input files is now
+                                preserved throughout the processing stream.
+       --json-fatal-arrays-on-input
+                                Miller now supports arrays as of version 6.
+       --json-map-arrays-on-input
+                                Miller now supports arrays as of version 6.
+       --json-skip-arrays-on-input
+                                Miller now supports arrays as of version 6.
+       --jsonx                  The `--jvstack` flag is now default true in Miller 6.
+       --jvquoteall             Type information from JSON input files is now
+                                preserved throughout the processing stream.
+       --mmap                   Miller no longer uses memory-mapping to access data
+                                files.
+       --no-fflush              The current implementation of Miller does not use
+                                buffered output, so there is no longer anything to
+                                suppress here.
+       --no-mmap                Miller no longer uses memory-mapping to access data
+                                files.
+       --ojsonx                 The `--jvstack` flag is now default true in Miller 6.
 
-       Note that this feature is quite general and is not limited to decompression
-       utilities. You can use it to apply per-file filters of your choice.  For output
-       compression (or other) utilities, simply pipe the output:
-       mlr ... | {your compression command} &gt; outputfilenamegoeshere
+MISCELLANEOUS FLAGS
+       --from {filename}        Use this to specify an input file before the verb(s),
+                                rather than after. May be used more than once.
+                                Example: `mlr --from a.dat --from b.dat cat` is the
+                                same as `mlr cat a.dat b.dat`.
+       --load {filename}        Load DSL script file for all put/filter operations on
+                                the command line. If the name following `--load` is a
+                                directory, load all `*.mlr` files in that directory.
+                                This is just like `put -f` and `filter -f` except
+                                it's up-front on the command line, so you can do
+                                something like `alias mlr='mlr --load ~/myscripts'`
+                                if you like.
+       --mfrom {filenames}      Use this to specify one of more input files before
+                                the verb(s), rather than after. May be used more than
+                                once. The list of filename must end with `--`. This
+                                is useful for example since `--from *.csv` doesn't do
+                                what you might hope but `--mfrom *.csv --` does.
+       --mload {filenames}      Like `--load` but works with more than one filename,
+                                e.g. `--mload *.mlr --`.
+       --ofmt {format}          E.g. %.18f, %.0f, %9.6e. Please use sprintf-style
+                                codes for floating-point nummbers. If not specified,
+                                default formatting is used. See also the `fmtnum`
+                                function and the `format-values` verb.
+       --seed {n}               with `n` of the form `12345678` or `0xcafefeed`. For
+                                `put`/`filter` `urand`, `urandint`, and `urand32`.
+       -I                       Process files in-place. For each file name on the
+                                command line, output is written to a temp file in the
+                                same directory, which is then renamed over the
+                                original. Each file is processed in isolation: if the
+                                output format is CSV, CSV headers will be present in
+                                each output file, statistics are only over each
+                                file's own records; and so on.
+       -n                       Process no input files, nor standard input either.
+                                Useful for `mlr put` with `begin`/`end` statements
+                                only. (Same as `--from /dev/null`.) Also useful in
+                                `mlr -n put -v '...'` for analyzing abstract syntax
+                                trees (if that's your thing).
 
-       Lastly, note that if --prepipe or --prepipex is specified, it replaces any
-       decisions that might have been made based on the file suffix. Also,
-       --gzin/--bz2in/--zin are ignored if --prepipe is also specified.
+OUTPUT-COLORIZATION FLAGS
+       Miller uses colors to highlight outputs. You can specify color preferences.
+       Note: output colorization does not work on Windows.
 
-   COMMENTS IN DATA
-       --skip-comments                 Ignore commented lines (prefixed by "#")
-                                       within the input.
-       --skip-comments-with {string}   Ignore commented lines within input, with
-                                       specified prefix.
-       --pass-comments                 Immediately print commented lines (prefixed by "#")
-                                       within the input.
-       --pass-comments-with {string}   Immediately print commented lines within input, with
-                                       specified prefix.
+       Things having colors:
 
-       Notes:
-       * Comments are only honored at the start of a line.
-       * In the absence of any of the above four options, comments are data like
-         any other text.
-       * When pass-comments is used, comment lines are written to standard output
-         immediately upon being read; they are not part of the record stream.  Results
-         may be counterintuitive. A suggestion is to place comments at the start of
-         data files.
+       * Keys in CSV header lines, JSON keys, etc
+       * Values in CSV data lines, JSON scalar values, etc in regression-test output
+       * Some online-help strings
 
-   CSV-SPECIFIC OPTIONS
-       --implicit-csv-header Use 1,2,3,... as field labels, rather than from line 1
-                          of input files. Tip: combine with "label" to recreate
-                          missing headers.
-       --no-implicit-csv-header Do not use --implicit-csv-header. This is the default
-                          anyway -- the main use is for the flags to 'mlr join' if you have
-                          main file(s) which are headerless but you want to join in on
-                          a file which does have a CSV header. Then you could use
-                          'mlr --csv --implicit-csv-header join --no-implicit-csv-header
-                          -l your-join-in-with-header.csv ... your-headerless.csv'
-       --allow-ragged-csv-input|--ragged If a data line has fewer fields than the header line,
-                          fill remaining keys with empty string. If a data line has more
-                          fields than the header line, use integer field labels as in
-                          the implicit-header case.
-       --headerless-csv-output   Print only CSV data lines.
-       -N                 Keystroke-saver for --implicit-csv-header --headerless-csv-output.
+       Rules for coloring:
 
-   DOUBLE-QUOTING FOR CSV/CSVLITE OUTPUT
-       THIS IS STILL WIP FOR MILLER 6
-       --quote-all        Wrap all fields in double quotes
-       --quote-none       Do not wrap any fields in double quotes, even if they have
-                          OFS or ORS in them
-       --quote-minimal    Wrap fields in double quotes only if they have OFS or ORS
-                          in them (default)
-       --quote-numeric    Wrap fields in double quotes only if they have numbers
-                          in them
-       --quote-original   Wrap fields in double quotes if and only if they were
-                          quoted on input. This isn't sticky for computed fields:
-                          e.g. if fields a and b were quoted on input and you do
-                          "put '$c = $a . $b'" then field c won't inherit a or b's
-                          was-quoted-on-input flag.
+       * By default, colorize output only if writing to stdout and stdout is a TTY.
+           * Example: color: `mlr --csv cat foo.csv`
+           * Example: no color: `mlr --csv cat foo.csv &gt; bar.csv`
+           * Example: no color: `mlr --csv cat foo.csv | less`
+       * The default colors were chosen since they look OK with white or black terminal background,
+         and are differentiable with common varieties of human color vision.
 
-   NUMBER FORMATTING
-       THIS IS STILL WIP FOR MILLER 6
-         --ofmt {format}    E.g. %.18f, %.0f, %9.6e. Please use sprintf-style codes for
-                            floating-point nummbers. If not specified, default formatting is used.
-                            See also the fmtnum function within mlr put (mlr --help-all-functions);
-                            see also the format-values function.
+       Mechanisms for coloring:
 
-   OTHER OPTIONS
-       --seed {n} with n of the form 12345678 or 0xcafefeed. For put/filter
-                          urand()/urandint()/urand32().
-       --nr-progress-mod {m}, with m a positive integer: print filename and record
-                          count to os.Stderr every m input records.
-       --from {filename}  Use this to specify an input file before the verb(s),
-                          rather than after. May be used more than once. Example:
-                          "mlr --from a.dat --from b.dat cat" is the same as
-                          "mlr cat a.dat b.dat".
-       --mfrom {filenames} --  Use this to specify one of more input files before the verb(s),
-                          rather than after. May be used more than once.
-                          The list of filename must end with "--". This is useful
-                          for example since "--from *.csv" doesn't do what you might
-                          hope but "--mfrom *.csv --" does.
-       --load {filename}  Load DSL script file for all put/filter operations on the command line.
-                          If the name following --load is a directory, load all "*.mlr" files
-                          in that directory. This is just like "put -f" and "filter -f"
-                          except it's up-front on the command line, so you can do something like
-                          alias mlr='mlr --load ~/myscripts' if you like.
-       --mload {names} -- Like --load but works with more than one filename,
-                          e.g. '--mload *.mlr --'.
-       -n                 Process no input files, nor standard input either. Useful
-                          for mlr put with begin/end statements only. (Same as --from
-                          /dev/null.) Also useful in "mlr -n put -v '...'" for
-                          analyzing abstract syntax trees (if that's your thing).
-       -I                 Process files in-place. For each file name on the command
-                          line, output is written to a temp file in the same
-                          directory, which is then renamed over the original. Each
-                          file is processed in isolation: if the output format is
-                          CSV, CSV headers will be present in each output file
-                          statistics are only over each file's own records; and so on.
+       * Miller uses ANSI escape sequences only. This does not work on Windows except within Cygwin.
+       * Requires `TERM` environment variable to be set to non-empty string.
+       * Doesn't try to check to see whether the terminal is capable of 256-color
+         ANSI vs 16-color ANSI. Note that if colors are in the range 0..15
+         then 16-color ANSI escapes are used, so this is in the user's control.
+
+       How you can control colorization:
+
+       * Suppression/unsuppression:
+           * Environment variable `export MLR_NO_COLOR=true` means don't color even if stdout+TTY.
+           * Environment variable `export MLR_ALWAYS_COLOR=true` means do color even if not stdout+TTY.
+             For example, you might want to use this when piping mlr output to `less -r`.
+           * Command-line flags `--no-color` or `-M`, `--always-color` or `-C`.
+
+       * Color choices can be specified by using environment variables, or command-line flags,
+         with values 0..255:
+           * `export MLR_KEY_COLOR=208`, `MLR_VALUE_COLOR=33`, etc.:
+               `MLR_KEY_COLOR` `MLR_VALUE_COLOR` `MLR_PASS_COLOR` `MLR_FAIL_COLOR`
+               `MLR_REPL_PS1_COLOR` `MLR_REPL_PS2_COLOR` `MLR_HELP_COLOR`
+           * Command-line flags `--key-color 208`, `--value-color 33`, etc.:
+               `--key-color` `--value-color` `--pass-color` `--fail-color`
+               `--repl-ps1-color` `--repl-ps2-color` `--help-color`
+           * This is particularly useful if your terminal's background color clashes with current settings.
+
+       If environment-variable settings and command-line flags are both provided, the latter take precedence.
+
+       Please do mlr `--list-color-codes` to see the available color codes (like 170), and
+       `mlr --list-color-names` to see available names (like `orchid`).
+
+       --always-color or -C
+       --fail-color
+       --help-color
+       --key-color
+       --list-color-codes
+       --list-color-names
+       --no-color or -M
+       --pass-color
+       --value-color
+
+PPRINT-ONLY FLAGS
+       These are flags which are applicable to PPRINT output format.
+
+       --barred                 Prints a border around PPRINT output (not available
+                                for input).
+       --right                  Right-justifies all fields for PPRINT output.
+
+SEPARATOR FLAGS
+       Separator options:
+
+           --rs     --irs     --ors              Record separators, e.g. 'lf' or '\\r\\n'
+           --fs     --ifs     --ofs  --repifs    Field separators, e.g. comma
+           --ps     --ips     --ops              Pair separators, e.g. equals sign
+
+       TODO: auto-detect is still TBD for Miller 6
+
+       Notes about line endings:
+
+       * Default line endings (`--irs` and `--ors`) are "auto" which means autodetect from
+         the input file format, as long as the input file(s) have lines ending in either
+         LF (also known as linefeed, `\n`, `0x0a`, or Unix-style) or CRLF (also known as
+         carriage-return/linefeed pairs, `\r\n`, `0x0d 0x0a`, or Windows-style).
+       * If both `irs` and `ors` are `auto` (which is the default) then LF input will lead to LF
+         output and CRLF input will lead to CRLF output, regardless of the platform you're
+         running on.
+       * The line-ending autodetector triggers on the first line ending detected in the input
+         stream. E.g. if you specify a CRLF-terminated file on the command line followed by an
+         LF-terminated file then autodetected line endings will be CRLF.
+       * If you use `--ors {something else}` with (default or explicitly specified) `--irs auto`
+         then line endings are autodetected on input and set to what you specify on output.
+       * If you use `--irs {something else}` with (default or explicitly specified) `--ors auto`
+         then the output line endings used are LF on Unix/Linux/BSD/MacOSX, and CRLF on Windows.
+
+       Notes about all other separators:
+
+       * IPS/OPS are only used for DKVP and XTAB formats, since only in these formats
+         do key-value pairs appear juxtaposed.
+       * IRS/ORS are ignored for XTAB format. Nominally IFS and OFS are newlines;
+         XTAB records are separated by two or more consecutive IFS/OFS -- i.e.
+         a blank line. Everything above about `--irs/--ors/--rs auto` becomes `--ifs/--ofs/--fs`
+         auto for XTAB format. (XTAB's default IFS/OFS are "auto".)
+       * OFS must be single-character for PPRINT format. This is because it is used
+         with repetition for alignment; multi-character separators would make
+         alignment impossible.
+       * OPS may be multi-character for XTAB format, in which case alignment is
+         disabled.
+       * TSV is simply CSV using tab as field separator (`--fs tab`).
+       * FS/PS are ignored for markdown format; RS is used.
+       * All FS and PS options are ignored for JSON format, since they are not relevant
+         to the JSON format.
+       * You can specify separators in any of the following ways, shown by example:
+         - Type them out, quoting as necessary for shell escapes, e.g.
+           `--fs '|' --ips :`
+         - C-style escape sequences, e.g. `--rs '\r\n' --fs '\t'`.
+         - To avoid backslashing, you can use any of the following names:
+              TODO desc-to-chars map
+
+       * Default separators by format:
+            TODO default_xses
+
+       --fs {string}            Specify FS for input and output.
+       --ifs {string}           Specify FS for input.
+       --ips {string}           Specify PS for input.
+       --irs {string}           Specify RS for input.
+       --ofs {string}           Specify FS for output.
+       --ops {string}           Specify PS for output.
+       --ors {string}           Specify RS for output.
+       --ps {string}            Specify PS for input and output.
+       --repifs                 Let IFS be repeated: e.g. for splitting on multiple
+                                spaces.
+       --rs {string}            Specify RS for input and output.
 
 AUXILIARY COMMANDS
        Available subcommands:
@@ -407,80 +638,6 @@ AUXILIARY COMMANDS
          repl
        For more information, please invoke mlr {subcommand} --help.
 
-REPL
-       Usage: mlr repl [options] {zero or more data-file names}
-       -v Prints the expressions's AST (abstract syntax tree), which gives
-          full transparency on the precedence and associativity rules of
-          Miller's grammar, to stdout.
-
-       -d Like -v but uses a parenthesized-expression format for the AST.
-
-       -D Like -d but with output all on one line.
-
-       -w Show warnings about uninitialized variables
-
-       -q Don't show startup banner
-
-       -s Don't show prompts
-
-       --load {DSL script file} Load script file before presenting the prompt.
-          If the name following --load is a directory, load all "*.mlr" files
-          in that directory.
-
-       --mload {DSL script files} -- Like --load but works with more than one filename,
-          e.g. '--mload *.mlr --'.
-
-       -h|--help Show this message.
-
-       Or any --icsv, --ojson, etc. reader/writer options as for the main Miller command line.
-
-       Any data-file names are opened just as if you had waited and typed :open {filenames}
-       at the Miller REPL prompt.
-
-OUTPUT COLORIZATION
-       Things having colors:
-       * Keys in CSV header lines, JSON keys, etc
-       * Values in CSV data lines, JSON scalar values, etc
-        in regression-test output
-       * Some online-help strings
-
-       Rules for coloring:
-       * By default, colorize output only if writing to stdout and stdout is a TTY.
-         * Example: color: mlr --csv cat foo.csv
-         * Example: no color: mlr --csv cat foo.csv &gt; bar.csv
-         * Example: no color: mlr --csv cat foo.csv | less
-       * The default colors were chosen since they look OK with white or black terminal background,
-         and are differentiable with common varieties of human color vision.
-
-       Mechanisms for coloring:
-       * Miller uses ANSI escape sequences only. This does not work on Windows except on Cygwin.
-       * Requires TERM environment variable to be set to non-empty string.
-       * Doesn't try to check to see whether the terminal is capable of 256-color
-         ANSI vs 16-color ANSI. Note that if colors are in the range 0..15
-         then 16-color ANSI escapes are used, so this is in the user's control.
-
-       How you can control colorization:
-       * Suppression/unsuppression:
-         * Environment variable export MLR_NO_COLOR=true means don't color even if stdout+TTY.
-         * Environment variable export MLR_ALWAYS_COLOR=true means do color even if not stdout+TTY.
-           For example, you might want to use this when piping mlr output to less -r.
-         * Command-line flags --no-color or -M, --always-color or -C.
-
-       * Color choices can be specified by using environment variables, or command-line flags,
-         with values 0..255:
-         * export MLR_KEY_COLOR=208, MLR_VALUE_COLOR-33, etc.:
-           MLR_KEY_COLOR MLR_VALUE_COLOR MLR_PASS_COLOR MLR_FAIL_COLOR
-           MLR_REPL_PS1_COLOR MLR_REPL_PS2_COLOR MLR_HELP_COLOR
-         * Command-line flags --key-color 208, --value-color 33, etc.:
-           --key-color --value-color --pass-color --fail-color
-           --repl-ps1-color --repl-ps2-color --help-color
-         * This is particularly useful if your terminal's background color clashes with current settings.
-
-       If environment-variable settings and command-line flags are both provided,the latter take precedence.
-
-       Please do mlr --list-color-codes to see the available color codes (like 170), and
-       mlr --list-color-names to see available names (like orchid).
-
 MLRRC
        You can set up personal defaults via a $HOME/.mlrrc and/or ./.mlrrc.
        For example, if you usually process CSV, then you can put "--csv" in your .mlrrc file
@@ -514,6 +671,36 @@ MLRRC
        See also:
        https://miller.readthedocs.io/en/latest/customization.html
 
+REPL
+       Usage: mlr repl [options] {zero or more data-file names}
+       -v Prints the expressions's AST (abstract syntax tree), which gives
+          full transparency on the precedence and associativity rules of
+          Miller's grammar, to stdout.
+
+       -d Like -v but uses a parenthesized-expression format for the AST.
+
+       -D Like -d but with output all on one line.
+
+       -w Show warnings about uninitialized variables
+
+       -q Don't show startup banner
+
+       -s Don't show prompts
+
+       --load {DSL script file} Load script file before presenting the prompt.
+          If the name following --load is a directory, load all "*.mlr" files
+          in that directory.
+
+       --mload {DSL script files} -- Like --load but works with more than one filename,
+          e.g. '--mload *.mlr --'.
+
+       -h|--help Show this message.
+
+       Or any --icsv, --ojson, etc. reader/writer options as for the main Miller command line.
+
+       Any data-file names are opened just as if you had waited and typed :open {filenames}
+       at the Miller REPL prompt.
+
 VERBS
    altkv
        Usage: mlr altkv [options]
@@ -2547,5 +2734,5 @@ SEE ALSO
 
 
 
-                                  2021-09-05                         MILLER(1)
+                                  2021-09-08                         MILLER(1)
 </pre>
diff --git a/docs6/docs/manpage.txt b/docs6/docs/manpage.txt
index b8ad2fe58..59e1314dd 100644
--- a/docs6/docs/manpage.txt
+++ b/docs6/docs/manpage.txt
@@ -7,9 +7,9 @@ NAME
        as CSV and tabular JSON.
 
 SYNOPSIS
-       Usage: mlr [I/O options] {verb} [verb-dependent options ...] {zero or
-       more file names} Output of one verb may be chained as input to another
-       using "then", e.g.
+       Usage: mlr [flags] {verb} [verb-dependent options ...] {zero or more
+       file names} Output of one verb may be chained as input to another using
+       "then", e.g.
          mlr stats1 -a min,mean,max -f flag,u,v -g color then sort -f color
        Please see 'mlr help topics' for more information.  Please also see
        https://johnkerl.org/miller6
@@ -95,6 +95,43 @@ DATA FORMATS
        | fox jumped          | Record 2: "1":"fox", "2":"jumped"
        +---------------------+
 
+HELP OPTIONS
+       Type 'mlr help {topic}' for any of the following:
+       Essentials:
+         mlr help topics
+         mlr help basic-examples
+         mlr help data-formats
+       Flags:
+         mlr help flags
+       Verbs:
+         mlr help list-verbs
+         mlr help usage-verbs
+         mlr help verb
+       Functions:
+         mlr help list-functions
+         mlr help list-function-classes
+         mlr help list-functions-in-class
+         mlr help usage-functions
+         mlr help usage-functions-by-class
+         mlr help function
+       Keywords:
+         mlr help list-keywords
+         mlr help usage-keywords
+         mlr help keyword
+       Other:
+         mlr help auxents
+         mlr help mlrrc
+         mlr help output-colorization
+         mlr help type-arithmetic-info
+       Shorthands:
+         mlr -g = mlr help flags
+         mlr -l = mlr help list-verbs
+         mlr -L = mlr help usage-verbs
+         mlr -f = mlr help list-functions
+         mlr -F = mlr help usage-functions
+         mlr -k = mlr help list-keywords
+         mlr -K = mlr help usage-keywords
+
 VERB LIST
        altkv bar bootstrap cat check clean-whitespace count-distinct count
        count-similar cut decimate fill-down fill-empty filter flatten format-values
@@ -128,117 +165,171 @@ FUNCTION LIST
        version ! != !=~ % & && * ** + - . .* .+ .- ./ / // < << <= == =~ > >= >> >>>
        ?: ?? ??? ^ ^^ | || ~
 
-HELP OPTIONS
-       Type 'mlr help {topic}' for any of the following:
-         mlr help topics
-         mlr help auxents
-         mlr help basic-examples
-         mlr help comments-in-data
-         mlr help compressed-data
-         mlr help csv-options
-         mlr help data-format-options
-         mlr help data-formats
-         mlr help double-quoting
-         mlr help format-conversion
-         mlr help function
-         mlr help keyword
-         mlr help list-functions
-         mlr help list-function-classes
-         mlr help list-functions-in-class
-         mlr help list-functions-as-paragraph
-         mlr help list-functions-as-table
-         mlr help list-keywords
-         mlr help list-keywords-as-paragraph
-         mlr help list-verbs
-         mlr help list-verbs-as-paragraph
-         mlr help misc
-         mlr help mlrrc
-         mlr help number-formatting
-         mlr help output-colorization
-         mlr help separator-options
-         mlr help type-arithmetic-info
-         mlr help usage-functions
-         mlr help usage-functions-by-class
-         mlr help usage-keywords
-         mlr help usage-verbs
-         mlr help verb
-       Shorthands:
-         mlr -l = mlr help list-verbs
-         mlr -L = mlr help usage-verbs
-         mlr -f = mlr help list-functions
-         mlr -F = mlr help usage-functions
-         mlr -k = mlr help list-keywords
-         mlr -K = mlr help usage-keywords
+COMMENTS-IN-DATA FLAGS
+       Miller lets you put comments in your data, such as
 
-OPTIONS
-       In the following option flags, the version with "i" designates the
-       input stream, "o" the output stream, and the version without prefix
-       sets the option for both input and output stream. For example: --irs
-       sets the input record separator, --ors the output record separator, and
-       --rs sets both the input and output separator to the given value.
+           # This is a comment for a CSV file
+           a,b,c
+           1,2,3
+           4,5,6
 
-   DATA-FORMAT OPTIONS
-       --idkvp   --odkvp   --dkvp      Delimited key-value pairs, e.g "a=1,b=2"
-                                        (Miller's default format).
+       Notes:
 
-       --inidx   --onidx   --nidx      Implicitly-integer-indexed fields (Unix-toolkit style).
-       -T                              Synonymous with "--nidx --fs tab".
+       * Comments are only honored at the start of a line.
+       * In the absence of any of the below four options, comments are data like
+         any other text. (The comments-in-data feature is opt-in.)
+       * When `--pass-comments` is used, comment lines are written to standard output
+         immediately upon being read; they are not part of the record stream.  Results
+         may be counterintuitive. A suggestion is to place comments at the start of
+         data files.
 
-       --icsv    --ocsv    --csv       Comma-separated value (or tab-separated with --fs tab, etc.)
+       --pass-comments          Immediately print commented lines (prefixed by `#`)
+                                within the input.
+       --pass-comments-with {string}
+                                Immediately print commented lines within input, with
+                                specified prefix.
+       --skip-comments          Ignore commented lines (prefixed by `#`) within the
+                                input.
+       --skip-comments-with {string}
+                                Ignore commented lines within input, with specified
+                                prefix.
 
-       --itsv    --otsv    --tsv       Keystroke-savers for "--icsv --ifs tab",
-                                       "--ocsv --ofs tab", "--csv --fs tab".
-       --iasv    --oasv    --asv       Similar but using ASCII FS 0x1f and RS 0x1e\n",
-       --iusv    --ousv    --usv       Similar but using Unicode FS U+241F (UTF-8 0xe2909f)\n",
-                                       and RS U+241E (UTF-8 0xe2909e)\n",
+COMPRESSED-DATA FLAGS
+       Miller offers a few different ways to handle reading data files which have been compressed.
 
-       --icsvlite --ocsvlite --csvlite Comma-separated value (or tab-separated with --fs tab, etc.).
-                                              The 'lite' CSV does not handle RFC-CSV double-quoting rules; is
-                                              slightly faster and handles heterogeneity in the input stream via
-                                              empty newline followed by new header line. See also
-                                               https://johnkerl.org/miller6/file-formats.html#csv-tsv-asv-usv-etc
+       * Decompression done within the Miller process itself: `--bz2in` `--gzin` `--zin`
+       * Decompression done outside the Miller process: `--prepipe` `--prepipex`
 
-       --itsvlite --otsvlite --tsvlite Keystroke-savers for "--icsvlite --ifs tab",
-                                       "--ocsvlite --ofs tab", "--csvlite --fs tab".
-       -t                              Synonymous with --tsvlite.
-       --iasvlite --oasvlite --asvlite Similar to --itsvlite et al. but using ASCII FS 0x1f and RS 0x1e\n",
-       --iusvlite --ousvlite --usvlite Similar to --itsvlite et al. but using Unicode FS U+241F (UTF-8 0xe2909f)\n",
-                                       and RS U+241E (UTF-8 0xe2909e)\n",
+       Using `--prepipe` and `--prepipex` you can specify an action to be
+       taken on each input file.  The prepipe command must be able to read from
+       standard input; it will be invoked with `{command} < {filename}`.  The
+       prepipex command must take a filename as argument; it will be invoked with
+       `{command} {filename}`.
 
-       --ipprint --opprint --pprint    Pretty-printed tabular (produces no
-                                       output until all input is in).
-                           --right     Right-justifies all fields for PPRINT output.
-                           --barred    Prints a border around PPRINT output
-                                       (only available for output).
+       Examples:
 
-                 --omd                 Markdown-tabular (only available for output).
+           mlr --prepipe gunzip
+           mlr --prepipe zcat -cf
+           mlr --prepipe xz -cd
+           mlr --prepipe cat
 
-       --ixtab   --oxtab   --xtab      Pretty-printed vertical-tabular.
-                           --xvright   Right-justifies values for XTAB format.
+       Note that this feature is quite general and is not limited to decompression
+       utilities. You can use it to apply per-file filters of your choice.  For output
+       compression (or other) utilities, simply pipe the output:
+       `mlr ... | {your compression command} > outputfilenamegoeshere`
 
-       --ijson   --ojson   --json      JSON tabular: sequence or list of one-level
-                                       maps: {...}{...} or [{...},{...}].
-                           --jvstack   Put one key-value pair per line for JSON output.
-                        --no-jvstack   Put objects/arrays all on one line for JSON output.
-                     --jsonx --ojsonx  Keystroke-savers for --json --jvstack
-                     --jsonx --ojsonx  and --ojson --jvstack, respectively.
-                           --jlistwrap Wrap JSON output in outermost [ ].
-                   --flatsep {string} Separator for flattening multi-level JSON keys,
-                                       e.g. '{"a":{"b":3}}' becomes a:b => 3 for
-                                       non-JSON formats. Defaults to ..\n",
+       Lastly, note that if `--prepipe` or `--prepipex` is specified, it replaces any
+       decisions that might have been made based on the file suffix. Likewise,
+       `--gzin`/`--bz2in`/`--zin` are ignored if `--prepipe` is also specified.
 
-       -p is a keystroke-saver for --nidx --fs space --repifs
+       --bz2in                  Uncompress bzip2 within the Miller process. Done by
+                                default if file ends in `.bz2`.
+       --gzin                   Uncompress gzip within the Miller process. Done by
+                                default if file ends in `.gz`.
+       --prepipe {decompression command}
+                                You can, of course, already do without this for
+                                single input files, e.g. `gunzip < myfile.csv.gz |
+                                mlr ...`. Allowed at the command line, but not in
+                                `.mlrrc` to avoid unexpected code execution.
+       --prepipe-bz2            Same as `--prepipe bz2`, except this is allowed in
+                                `.mlrrc`.
+       --prepipe-gunzip         Same as `--prepipe gunzip`, except this is allowed in
+                                `.mlrrc`.
+       --prepipe-zcat           Same as `--prepipe zcat`, except this is allowed in
+                                `.mlrrc`.
+       --prepipex {decompression command}
+                                Like `--prepipe` with one exception: doesn't insert
+                                `<` between command and filename at runtime. Useful
+                                for some commands like `unzip -qc` which don't read
+                                standard input. Allowed at the command line, but not
+                                in `.mlrrc` to avoid unexpected code execution.
+       --zin                    Uncompress zlib within the Miller process. Done by
+                                default if file ends in `.z`.
 
-       Examples: --csv for CSV-formatted input and output; --icsv --opprint for
+CSV-ONLY FLAGS
+       --allow-ragged-csv-input or --ragged
+                                If a data line has fewer fields than the header line,
+                                fill remaining keys with empty string. If a data line
+                                has more fields than the header line, use integer
+                                field labels as in the implicit-header case.
+       --headerless-csv-output  Print only CSV data lines; do not print CSV header
+                                lines.
+       --implicit-csv-header    Use 1,2,3,... as field labels, rather than from line
+                                1 of input files. Tip: combine with `label` to
+                                recreate missing headers.
+       --no-implicit-csv-header Opposite of `--implicit-csv-header`. This is the
+                                default anyway -- the main use is for the flags to
+                                `mlr join` if you have main file(s) which are
+                                headerless but you want to join in on a file which
+                                does have a CSV header. Then you could use `mlr --csv
+                                --implicit-csv-header join --no-implicit-csv-header
+                                -l your-join-in-with-header.csv ...
+                                your-headerless.csv`.
+       -N                       Keystroke-saver for `--implicit-csv-header
+                                --headerless-csv-output`.
+
+FILE-FORMAT FLAGS
+       TO DO: brief list of formats w/ xref to m6 webdocs.
+
+       Examples: `--csv` for CSV-formatted input and output; `--icsv --opprint` for
        CSV-formatted input and pretty-printed output.
 
-       Please use --iformat1 --oformat2 rather than --format1 --oformat2.
-       The latter sets up input and output flags for format1, not all of which
-       are overridden in all cases by setting output format to format2.
+       Please use `--iformat1 --oformat2` rather than `--format1 --oformat2`.
+       The latter sets up input and output flags for `format1`, not all of which
+       are overridden in all cases by setting output format to `format2`.
 
-   FORMAT-CONVERSION KEYSTROKE-SAVERS
+       --asv or --asvlite       Use ASV format for input and output data.
+       --csv or -c              Use CSV format for input and output data.
+       --csvlite                Use CSV-lite format for input and output data.
+       --dkvp                   Use DKVP format for input and output data.
+       --iasv or --iasvlite     Use ASV format for input data.
+       --icsv                   Use CSV format for input data.
+       --icsvlite               Use CSV-lite format for input data.
+       --idkvp                  Use DKVP format for input data.
+       --ijson                  Use JSON format for input data.
+       --inidx                  Use NIDX format for input data.
+       --io {format name}       Use format name for input and output data. For
+                                example: `--io csv` is the same as `--csv`.
+       --ipprint                Use PPRINT format for input data.
+       --itsv                   Use TSV format for input data.
+       --itsvlite               Use TSV-lite format for input data.
+       --iusv or --iusvlite     Use USV format for input data.
+       --ixtab                  Use XTAB format for input data.
+       --json or -j             Use JSON format for input and output data.
+       --nidx                   Use NIDX format for input and output data.
+       --oasv or --oasvlite     Use ASV format for output data.
+       --ocsv                   Use CSV format for output data.
+       --ocsvlite               Use CSV-lite format for output data.
+       --odkvp                  Use DKVP format for output data.
+       --ojson                  Use JSON format for output data.
+       --omd                    Use markdown-tabular format for output data.
+       --onidx                  Use NIDX format for output data.
+       --opprint                Use PPRINT format for output data.
+       --otsv                   Use TSV format for output data.
+       --otsvlite               Use TSV-lite format for output data.
+       --ousv or --ousvlite     Use USV format for output data.
+       --oxtab                  Use XTAB format for output data.
+       --pprint                 Use PPRINT format for input and output data.
+       --tsv                    Use TSV format for input and output data.
+       --tsvlite or -t          Use TSV-lite format for input and output data.
+       --usv or --usvlite       Use USV format for input and output data.
+       --xtab                   Use XTAB format for input and output data.
+       -i {format name}         Use format name for input data. For example: `-i csv`
+                                is the same as `--icsv`.
+       -o {format name}         Use format name for output data. For example: `-o
+                                csv` is the same as `--ocsv`.
+
+FLATTEN-UNFLATTEN FLAGS
+       --flatsep or --jflatsep or --oflatsep {string}
+                                Separator for flattening multi-level JSON keys, e.g.
+                                `{"a":{"b":3}}` becomes `a:b => 3` for non-JSON
+                                formats. Defaults to `.`.
+       --no-auto-flatten
+       --no-auto-unflatten
+
+FORMAT-CONVERSION KEYSTROKE-SAVER FLAGS
        As keystroke-savers for format-conversion you may use the following:
-       --c2t --c2d --c2n --c2j --c2x --c2p --c2m
+             --c2t --c2d --c2n --c2j --c2x --c2p --c2m
        --t2c       --t2d --t2n --t2j --t2x --t2p --t2m
        --d2c --d2t       --d2n --d2j --d2x --d2p --d2m
        --n2c --n2t --n2d       --n2j --n2x --n2p --n2m
@@ -249,130 +340,270 @@ OPTIONS
        PPRINT, and markdown, respectively. Note that markdown format is available for
        output only.
 
-   SEPARATORS
-       THIS IS STILL TBD FOR MILLER 6
+       --c2b                    Use CSV for input, PPRINT with `--barred` for output.
+       --c2d                    Use CSV for input, DKVP for output.
+       --c2j                    Use CSV for input, JSON for output.
+       --c2m                    Use CSV for input, markdown-tabular for output.
+       --c2n                    Use CSV for input, NIDX for output.
+       --c2p                    Use CSV for input, PPRINT for output.
+       --c2t                    Use CSV for input, TSV for output.
+       --c2x                    Use CSV for input, XTAB for output.
+       --d2b                    Use DKVP for input, PPRINT with `--barred` for
+                                output.
+       --d2c                    Use DKVP for input, CSV for output.
+       --d2j                    Use DKVP for input, JSON for output.
+       --d2m                    Use DKVP for input, markdown-tabular for output.
+       --d2n                    Use DKVP for input, NIDX for output.
+       --d2p                    Use DKVP for input, PPRINT for output.
+       --d2t                    Use DKVP for input, TSV for output.
+       --d2x                    Use DKVP for input, XTAB for output.
+       --j2b                    Use JSON for input, PPRINT with --barred for output.
+       --j2c                    Use JSON for input, CSV for output.
+       --j2d                    Use JSON for input, DKVP for output.
+       --j2m                    Use JSON for input, markdown-tabular for output.
+       --j2n                    Use JSON for input, NIDX for output.
+       --j2p                    Use JSON for input, PPRINT for output.
+       --j2t                    Use JSON for input, TSV for output.
+       --j2x                    Use JSON for input, XTAB for output.
+       --n2b                    Use NIDX for input, PPRINT with `--barred` for
+                                output.
+       --n2c                    Use NIDX for input, CSV for output.
+       --n2d                    Use NIDX for input, DKVP for output.
+       --n2j                    Use NIDX for input, JSON for output.
+       --n2m                    Use NIDX for input, markdown-tabular for output.
+       --n2p                    Use NIDX for input, PPRINT for output.
+       --n2t                    Use NIDX for input, TSV for output.
+       --n2x                    Use NIDX for input, XTAB for output.
+       --p2c                    Use PPRINT for input, CSV for output.
+       --p2d                    Use PPRINT for input, DKVP for output.
+       --p2j                    Use PPRINT for input, JSON for output.
+       --p2m                    Use PPRINT for input, markdown-tabular for output.
+       --p2n                    Use PPRINT for input, NIDX for output.
+       --p2t                    Use PPRINT for input, TSV for output.
+       --p2x                    Use PPRINT for input, XTAB for output.
+       --t2b                    Use TSV for input, PPRINT with `--barred` for output.
+       --t2c                    Use TSV for input, CSV for output.
+       --t2d                    Use TSV for input, DKVP for output.
+       --t2j                    Use TSV for input, JSON for output.
+       --t2m                    Use TSV for input, markdown-tabular for output.
+       --t2n                    Use TSV for input, NIDX for output.
+       --t2p                    Use TSV for input, PPRINT for output.
+       --t2x                    Use TSV for input, XTAB for output.
+       --x2b                    Use XTAB for input, PPRINT with `--barred` for
+                                output.
+       --x2c                    Use XTAB for input, CSV for output.
+       --x2d                    Use XTAB for input, DKVP for output.
+       --x2j                    Use XTAB for input, JSON for output.
+       --x2m                    Use XTAB for input, markdown-tabular for output.
+       --x2n                    Use XTAB for input, NIDX for output.
+       --x2p                    Use XTAB for input, PPRINT for output.
+       --x2t                    Use XTAB for input, TSV for output.
+       -p                       Keystroke-saver for `--nidx --fs space --repifs`.
+       -T                       Keystroke-saver for `--nidx --fs tab`.
 
-   COMPRESSED I/O
-       Decompression done within the Miller process itself:
-       --gzin  Uncompress gzip within the Miller process. Done by default if file ends in ".gz".
-       --bz2in Uncompress bz2ip within the Miller process. Done by default if file ends in ".bz2".
-       --zin   Uncompress zlib within the Miller process. Done by default if file ends in ".z".
+JSON-ONLY FLAGS
+       These are flags which are applicable to JSON format.
 
-       Decompression done outside the Miller process:
-       --prepipe {command} You can, of course, already do without this for single input files,
-         e.g. "gunzip < myfile.csv.gz | mlr ..."
-       --prepipex {command} Like --prepipe with one exception: doesn't insert '<' between
-         command and filename at runtime. Useful for some commands like 'unzip -qc'
-         which don't read standard input.
+       --jlistwrap or --jl      Wrap JSON output in outermost `[ ]`.
+       --jvstack                Put one key-value pair per line for JSON output
+                                (multi-line output).
+       --no-jvstack             Put objects/arrays all on one line for JSON output.
 
-       Using --prepipe and --prepipex you can specify an action to be taken on each
-       input file. This prepipe command must be able to read from standard input; it
-       will be invoked with {command} < {filename}.
+LEGACY FLAGS
+       These are flags which don't do anything in the current Miller version.
+       They are accepted as no-op flags in order to keep old scripts from breaking.
 
-       Examples:
-         mlr --prepipe gunzip
-         mlr --prepipe zcat -cf
-         mlr --prepipe xz -cd
-         mlr --prepipe cat
+       --jknquoteint            Type information from JSON input files is now
+                                preserved throughout the processing stream.
+       --jquoteall              Type information from JSON input files is now
+                                preserved throughout the processing stream.
+       --json-fatal-arrays-on-input
+                                Miller now supports arrays as of version 6.
+       --json-map-arrays-on-input
+                                Miller now supports arrays as of version 6.
+       --json-skip-arrays-on-input
+                                Miller now supports arrays as of version 6.
+       --jsonx                  The `--jvstack` flag is now default true in Miller 6.
+       --jvquoteall             Type information from JSON input files is now
+                                preserved throughout the processing stream.
+       --mmap                   Miller no longer uses memory-mapping to access data
+                                files.
+       --no-fflush              The current implementation of Miller does not use
+                                buffered output, so there is no longer anything to
+                                suppress here.
+       --no-mmap                Miller no longer uses memory-mapping to access data
+                                files.
+       --ojsonx                 The `--jvstack` flag is now default true in Miller 6.
 
-       Note that this feature is quite general and is not limited to decompression
-       utilities. You can use it to apply per-file filters of your choice.  For output
-       compression (or other) utilities, simply pipe the output:
-       mlr ... | {your compression command} > outputfilenamegoeshere
+MISCELLANEOUS FLAGS
+       --from {filename}        Use this to specify an input file before the verb(s),
+                                rather than after. May be used more than once.
+                                Example: `mlr --from a.dat --from b.dat cat` is the
+                                same as `mlr cat a.dat b.dat`.
+       --load {filename}        Load DSL script file for all put/filter operations on
+                                the command line. If the name following `--load` is a
+                                directory, load all `*.mlr` files in that directory.
+                                This is just like `put -f` and `filter -f` except
+                                it's up-front on the command line, so you can do
+                                something like `alias mlr='mlr --load ~/myscripts'`
+                                if you like.
+       --mfrom {filenames}      Use this to specify one of more input files before
+                                the verb(s), rather than after. May be used more than
+                                once. The list of filename must end with `--`. This
+                                is useful for example since `--from *.csv` doesn't do
+                                what you might hope but `--mfrom *.csv --` does.
+       --mload {filenames}      Like `--load` but works with more than one filename,
+                                e.g. `--mload *.mlr --`.
+       --ofmt {format}          E.g. %.18f, %.0f, %9.6e. Please use sprintf-style
+                                codes for floating-point nummbers. If not specified,
+                                default formatting is used. See also the `fmtnum`
+                                function and the `format-values` verb.
+       --seed {n}               with `n` of the form `12345678` or `0xcafefeed`. For
+                                `put`/`filter` `urand`, `urandint`, and `urand32`.
+       -I                       Process files in-place. For each file name on the
+                                command line, output is written to a temp file in the
+                                same directory, which is then renamed over the
+                                original. Each file is processed in isolation: if the
+                                output format is CSV, CSV headers will be present in
+                                each output file, statistics are only over each
+                                file's own records; and so on.
+       -n                       Process no input files, nor standard input either.
+                                Useful for `mlr put` with `begin`/`end` statements
+                                only. (Same as `--from /dev/null`.) Also useful in
+                                `mlr -n put -v '...'` for analyzing abstract syntax
+                                trees (if that's your thing).
 
-       Lastly, note that if --prepipe or --prepipex is specified, it replaces any
-       decisions that might have been made based on the file suffix. Also,
-       --gzin/--bz2in/--zin are ignored if --prepipe is also specified.
+OUTPUT-COLORIZATION FLAGS
+       Miller uses colors to highlight outputs. You can specify color preferences.
+       Note: output colorization does not work on Windows.
 
-   COMMENTS IN DATA
-       --skip-comments                 Ignore commented lines (prefixed by "#")
-                                       within the input.
-       --skip-comments-with {string}   Ignore commented lines within input, with
-                                       specified prefix.
-       --pass-comments                 Immediately print commented lines (prefixed by "#")
-                                       within the input.
-       --pass-comments-with {string}   Immediately print commented lines within input, with
-                                       specified prefix.
+       Things having colors:
 
-       Notes:
-       * Comments are only honored at the start of a line.
-       * In the absence of any of the above four options, comments are data like
-         any other text.
-       * When pass-comments is used, comment lines are written to standard output
-         immediately upon being read; they are not part of the record stream.  Results
-         may be counterintuitive. A suggestion is to place comments at the start of
-         data files.
+       * Keys in CSV header lines, JSON keys, etc
+       * Values in CSV data lines, JSON scalar values, etc in regression-test output
+       * Some online-help strings
 
-   CSV-SPECIFIC OPTIONS
-       --implicit-csv-header Use 1,2,3,... as field labels, rather than from line 1
-                          of input files. Tip: combine with "label" to recreate
-                          missing headers.
-       --no-implicit-csv-header Do not use --implicit-csv-header. This is the default
-                          anyway -- the main use is for the flags to 'mlr join' if you have
-                          main file(s) which are headerless but you want to join in on
-                          a file which does have a CSV header. Then you could use
-                          'mlr --csv --implicit-csv-header join --no-implicit-csv-header
-                          -l your-join-in-with-header.csv ... your-headerless.csv'
-       --allow-ragged-csv-input|--ragged If a data line has fewer fields than the header line,
-                          fill remaining keys with empty string. If a data line has more
-                          fields than the header line, use integer field labels as in
-                          the implicit-header case.
-       --headerless-csv-output   Print only CSV data lines.
-       -N                 Keystroke-saver for --implicit-csv-header --headerless-csv-output.
+       Rules for coloring:
 
-   DOUBLE-QUOTING FOR CSV/CSVLITE OUTPUT
-       THIS IS STILL WIP FOR MILLER 6
-       --quote-all        Wrap all fields in double quotes
-       --quote-none       Do not wrap any fields in double quotes, even if they have
-                          OFS or ORS in them
-       --quote-minimal    Wrap fields in double quotes only if they have OFS or ORS
-                          in them (default)
-       --quote-numeric    Wrap fields in double quotes only if they have numbers
-                          in them
-       --quote-original   Wrap fields in double quotes if and only if they were
-                          quoted on input. This isn't sticky for computed fields:
-                          e.g. if fields a and b were quoted on input and you do
-                          "put '$c = $a . $b'" then field c won't inherit a or b's
-                          was-quoted-on-input flag.
+       * By default, colorize output only if writing to stdout and stdout is a TTY.
+           * Example: color: `mlr --csv cat foo.csv`
+           * Example: no color: `mlr --csv cat foo.csv > bar.csv`
+           * Example: no color: `mlr --csv cat foo.csv | less`
+       * The default colors were chosen since they look OK with white or black terminal background,
+         and are differentiable with common varieties of human color vision.
 
-   NUMBER FORMATTING
-       THIS IS STILL WIP FOR MILLER 6
-         --ofmt {format}    E.g. %.18f, %.0f, %9.6e. Please use sprintf-style codes for
-                            floating-point nummbers. If not specified, default formatting is used.
-                            See also the fmtnum function within mlr put (mlr --help-all-functions);
-                            see also the format-values function.
+       Mechanisms for coloring:
 
-   OTHER OPTIONS
-       --seed {n} with n of the form 12345678 or 0xcafefeed. For put/filter
-                          urand()/urandint()/urand32().
-       --nr-progress-mod {m}, with m a positive integer: print filename and record
-                          count to os.Stderr every m input records.
-       --from {filename}  Use this to specify an input file before the verb(s),
-                          rather than after. May be used more than once. Example:
-                          "mlr --from a.dat --from b.dat cat" is the same as
-                          "mlr cat a.dat b.dat".
-       --mfrom {filenames} --  Use this to specify one of more input files before the verb(s),
-                          rather than after. May be used more than once.
-                          The list of filename must end with "--". This is useful
-                          for example since "--from *.csv" doesn't do what you might
-                          hope but "--mfrom *.csv --" does.
-       --load {filename}  Load DSL script file for all put/filter operations on the command line.
-                          If the name following --load is a directory, load all "*.mlr" files
-                          in that directory. This is just like "put -f" and "filter -f"
-                          except it's up-front on the command line, so you can do something like
-                          alias mlr='mlr --load ~/myscripts' if you like.
-       --mload {names} -- Like --load but works with more than one filename,
-                          e.g. '--mload *.mlr --'.
-       -n                 Process no input files, nor standard input either. Useful
-                          for mlr put with begin/end statements only. (Same as --from
-                          /dev/null.) Also useful in "mlr -n put -v '...'" for
-                          analyzing abstract syntax trees (if that's your thing).
-       -I                 Process files in-place. For each file name on the command
-                          line, output is written to a temp file in the same
-                          directory, which is then renamed over the original. Each
-                          file is processed in isolation: if the output format is
-                          CSV, CSV headers will be present in each output file
-                          statistics are only over each file's own records; and so on.
+       * Miller uses ANSI escape sequences only. This does not work on Windows except within Cygwin.
+       * Requires `TERM` environment variable to be set to non-empty string.
+       * Doesn't try to check to see whether the terminal is capable of 256-color
+         ANSI vs 16-color ANSI. Note that if colors are in the range 0..15
+         then 16-color ANSI escapes are used, so this is in the user's control.
+
+       How you can control colorization:
+
+       * Suppression/unsuppression:
+           * Environment variable `export MLR_NO_COLOR=true` means don't color even if stdout+TTY.
+           * Environment variable `export MLR_ALWAYS_COLOR=true` means do color even if not stdout+TTY.
+             For example, you might want to use this when piping mlr output to `less -r`.
+           * Command-line flags `--no-color` or `-M`, `--always-color` or `-C`.
+
+       * Color choices can be specified by using environment variables, or command-line flags,
+         with values 0..255:
+           * `export MLR_KEY_COLOR=208`, `MLR_VALUE_COLOR=33`, etc.:
+               `MLR_KEY_COLOR` `MLR_VALUE_COLOR` `MLR_PASS_COLOR` `MLR_FAIL_COLOR`
+               `MLR_REPL_PS1_COLOR` `MLR_REPL_PS2_COLOR` `MLR_HELP_COLOR`
+           * Command-line flags `--key-color 208`, `--value-color 33`, etc.:
+               `--key-color` `--value-color` `--pass-color` `--fail-color`
+               `--repl-ps1-color` `--repl-ps2-color` `--help-color`
+           * This is particularly useful if your terminal's background color clashes with current settings.
+
+       If environment-variable settings and command-line flags are both provided, the latter take precedence.
+
+       Please do mlr `--list-color-codes` to see the available color codes (like 170), and
+       `mlr --list-color-names` to see available names (like `orchid`).
+
+       --always-color or -C
+       --fail-color
+       --help-color
+       --key-color
+       --list-color-codes
+       --list-color-names
+       --no-color or -M
+       --pass-color
+       --value-color
+
+PPRINT-ONLY FLAGS
+       These are flags which are applicable to PPRINT output format.
+
+       --barred                 Prints a border around PPRINT output (not available
+                                for input).
+       --right                  Right-justifies all fields for PPRINT output.
+
+SEPARATOR FLAGS
+       Separator options:
+
+           --rs     --irs     --ors              Record separators, e.g. 'lf' or '\\r\\n'
+           --fs     --ifs     --ofs  --repifs    Field separators, e.g. comma
+           --ps     --ips     --ops              Pair separators, e.g. equals sign
+
+       TODO: auto-detect is still TBD for Miller 6
+
+       Notes about line endings:
+
+       * Default line endings (`--irs` and `--ors`) are "auto" which means autodetect from
+         the input file format, as long as the input file(s) have lines ending in either
+         LF (also known as linefeed, `\n`, `0x0a`, or Unix-style) or CRLF (also known as
+         carriage-return/linefeed pairs, `\r\n`, `0x0d 0x0a`, or Windows-style).
+       * If both `irs` and `ors` are `auto` (which is the default) then LF input will lead to LF
+         output and CRLF input will lead to CRLF output, regardless of the platform you're
+         running on.
+       * The line-ending autodetector triggers on the first line ending detected in the input
+         stream. E.g. if you specify a CRLF-terminated file on the command line followed by an
+         LF-terminated file then autodetected line endings will be CRLF.
+       * If you use `--ors {something else}` with (default or explicitly specified) `--irs auto`
+         then line endings are autodetected on input and set to what you specify on output.
+       * If you use `--irs {something else}` with (default or explicitly specified) `--ors auto`
+         then the output line endings used are LF on Unix/Linux/BSD/MacOSX, and CRLF on Windows.
+
+       Notes about all other separators:
+
+       * IPS/OPS are only used for DKVP and XTAB formats, since only in these formats
+         do key-value pairs appear juxtaposed.
+       * IRS/ORS are ignored for XTAB format. Nominally IFS and OFS are newlines;
+         XTAB records are separated by two or more consecutive IFS/OFS -- i.e.
+         a blank line. Everything above about `--irs/--ors/--rs auto` becomes `--ifs/--ofs/--fs`
+         auto for XTAB format. (XTAB's default IFS/OFS are "auto".)
+       * OFS must be single-character for PPRINT format. This is because it is used
+         with repetition for alignment; multi-character separators would make
+         alignment impossible.
+       * OPS may be multi-character for XTAB format, in which case alignment is
+         disabled.
+       * TSV is simply CSV using tab as field separator (`--fs tab`).
+       * FS/PS are ignored for markdown format; RS is used.
+       * All FS and PS options are ignored for JSON format, since they are not relevant
+         to the JSON format.
+       * You can specify separators in any of the following ways, shown by example:
+         - Type them out, quoting as necessary for shell escapes, e.g.
+           `--fs '|' --ips :`
+         - C-style escape sequences, e.g. `--rs '\r\n' --fs '\t'`.
+         - To avoid backslashing, you can use any of the following names:
+              TODO desc-to-chars map
+
+       * Default separators by format:
+            TODO default_xses
+
+       --fs {string}            Specify FS for input and output.
+       --ifs {string}           Specify FS for input.
+       --ips {string}           Specify PS for input.
+       --irs {string}           Specify RS for input.
+       --ofs {string}           Specify FS for output.
+       --ops {string}           Specify PS for output.
+       --ors {string}           Specify RS for output.
+       --ps {string}            Specify PS for input and output.
+       --repifs                 Let IFS be repeated: e.g. for splitting on multiple
+                                spaces.
+       --rs {string}            Specify RS for input and output.
 
 AUXILIARY COMMANDS
        Available subcommands:
@@ -386,80 +617,6 @@ AUXILIARY COMMANDS
          repl
        For more information, please invoke mlr {subcommand} --help.
 
-REPL
-       Usage: mlr repl [options] {zero or more data-file names}
-       -v Prints the expressions's AST (abstract syntax tree), which gives
-          full transparency on the precedence and associativity rules of
-          Miller's grammar, to stdout.
-
-       -d Like -v but uses a parenthesized-expression format for the AST.
-
-       -D Like -d but with output all on one line.
-
-       -w Show warnings about uninitialized variables
-
-       -q Don't show startup banner
-
-       -s Don't show prompts
-
-       --load {DSL script file} Load script file before presenting the prompt.
-          If the name following --load is a directory, load all "*.mlr" files
-          in that directory.
-
-       --mload {DSL script files} -- Like --load but works with more than one filename,
-          e.g. '--mload *.mlr --'.
-
-       -h|--help Show this message.
-
-       Or any --icsv, --ojson, etc. reader/writer options as for the main Miller command line.
-
-       Any data-file names are opened just as if you had waited and typed :open {filenames}
-       at the Miller REPL prompt.
-
-OUTPUT COLORIZATION
-       Things having colors:
-       * Keys in CSV header lines, JSON keys, etc
-       * Values in CSV data lines, JSON scalar values, etc
-        in regression-test output
-       * Some online-help strings
-
-       Rules for coloring:
-       * By default, colorize output only if writing to stdout and stdout is a TTY.
-         * Example: color: mlr --csv cat foo.csv
-         * Example: no color: mlr --csv cat foo.csv > bar.csv
-         * Example: no color: mlr --csv cat foo.csv | less
-       * The default colors were chosen since they look OK with white or black terminal background,
-         and are differentiable with common varieties of human color vision.
-
-       Mechanisms for coloring:
-       * Miller uses ANSI escape sequences only. This does not work on Windows except on Cygwin.
-       * Requires TERM environment variable to be set to non-empty string.
-       * Doesn't try to check to see whether the terminal is capable of 256-color
-         ANSI vs 16-color ANSI. Note that if colors are in the range 0..15
-         then 16-color ANSI escapes are used, so this is in the user's control.
-
-       How you can control colorization:
-       * Suppression/unsuppression:
-         * Environment variable export MLR_NO_COLOR=true means don't color even if stdout+TTY.
-         * Environment variable export MLR_ALWAYS_COLOR=true means do color even if not stdout+TTY.
-           For example, you might want to use this when piping mlr output to less -r.
-         * Command-line flags --no-color or -M, --always-color or -C.
-
-       * Color choices can be specified by using environment variables, or command-line flags,
-         with values 0..255:
-         * export MLR_KEY_COLOR=208, MLR_VALUE_COLOR-33, etc.:
-           MLR_KEY_COLOR MLR_VALUE_COLOR MLR_PASS_COLOR MLR_FAIL_COLOR
-           MLR_REPL_PS1_COLOR MLR_REPL_PS2_COLOR MLR_HELP_COLOR
-         * Command-line flags --key-color 208, --value-color 33, etc.:
-           --key-color --value-color --pass-color --fail-color
-           --repl-ps1-color --repl-ps2-color --help-color
-         * This is particularly useful if your terminal's background color clashes with current settings.
-
-       If environment-variable settings and command-line flags are both provided,the latter take precedence.
-
-       Please do mlr --list-color-codes to see the available color codes (like 170), and
-       mlr --list-color-names to see available names (like orchid).
-
 MLRRC
        You can set up personal defaults via a $HOME/.mlrrc and/or ./.mlrrc.
        For example, if you usually process CSV, then you can put "--csv" in your .mlrrc file
@@ -493,6 +650,36 @@ MLRRC
        See also:
        https://miller.readthedocs.io/en/latest/customization.html
 
+REPL
+       Usage: mlr repl [options] {zero or more data-file names}
+       -v Prints the expressions's AST (abstract syntax tree), which gives
+          full transparency on the precedence and associativity rules of
+          Miller's grammar, to stdout.
+
+       -d Like -v but uses a parenthesized-expression format for the AST.
+
+       -D Like -d but with output all on one line.
+
+       -w Show warnings about uninitialized variables
+
+       -q Don't show startup banner
+
+       -s Don't show prompts
+
+       --load {DSL script file} Load script file before presenting the prompt.
+          If the name following --load is a directory, load all "*.mlr" files
+          in that directory.
+
+       --mload {DSL script files} -- Like --load but works with more than one filename,
+          e.g. '--mload *.mlr --'.
+
+       -h|--help Show this message.
+
+       Or any --icsv, --ojson, etc. reader/writer options as for the main Miller command line.
+
+       Any data-file names are opened just as if you had waited and typed :open {filenames}
+       at the Miller REPL prompt.
+
 VERBS
    altkv
        Usage: mlr altkv [options]
@@ -2526,4 +2713,4 @@ SEE ALSO
 
 
 
-                                  2021-09-05                         MILLER(1)
+                                  2021-09-08                         MILLER(1)
diff --git a/docs6/docs/mk-flag-info.rb b/docs6/docs/mk-flag-info.rb
index 8ffae3ac3..c6b07cf7a 100755
--- a/docs6/docs/mk-flag-info.rb
+++ b/docs6/docs/mk-flag-info.rb
@@ -48,10 +48,6 @@ EOF
       help = `mlr help show-help-for-flag '#{flag}'`
       puts "* `#{headline}`: #{help}"
     end
-
-    #puts '```'
-    #system("mlr help list-flags-for-section '#{section_name}'")
-    #puts '```'
   end
 
   puts
diff --git a/docs6/docs/online-help.md b/docs6/docs/online-help.md
index d7e12135f..8b82f16a4 100644
--- a/docs6/docs/online-help.md
+++ b/docs6/docs/online-help.md
@@ -38,36 +38,34 @@ Please also see https://johnkerl.org/miller6
 </pre>
 <pre class="pre-non-highlight-in-pair">
 Type 'mlr help {topic}' for any of the following:
+Essentials:
   mlr help topics
-  mlr help auxents
   mlr help basic-examples
   mlr help data-formats
-  mlr help function
-  mlr help keyword
+Flags:
+  mlr help flags
+Verbs:
+  mlr help list-verbs
+  mlr help usage-verbs
+  mlr help verb
+Functions:
   mlr help list-functions
   mlr help list-function-classes
   mlr help list-functions-in-class
-  mlr help list-functions-as-paragraph
-  mlr help list-keywords
-  mlr help list-keywords-as-paragraph
-  mlr help list-verbs
-  mlr help list-verbs-as-paragraph
-  mlr help mlrrc
-  mlr help number-formatting
-  mlr help type-arithmetic-info
   mlr help usage-functions
   mlr help usage-functions-by-class
+  mlr help function
+Keywords:
+  mlr help list-keywords
   mlr help usage-keywords
-  mlr help usage-verbs
-  mlr help verb
-  mlr help comments-in-data
-  mlr help compressed-data
-  mlr help data-format-options
-  mlr help double-quoting
-  mlr help format-conversion
-  mlr help separator-options
+  mlr help keyword
+Other:
+  mlr help auxents
+  mlr help mlrrc
   mlr help output-colorization
+  mlr help type-arithmetic-info
 Shorthands:
+  mlr -g = mlr help flags
   mlr -l = mlr help list-verbs
   mlr -L = mlr help usage-verbs
   mlr -f = mlr help list-functions
@@ -81,36 +79,34 @@ Shorthands:
 </pre>
 <pre class="pre-non-highlight-in-pair">
 Type 'mlr help {topic}' for any of the following:
+Essentials:
   mlr help topics
-  mlr help auxents
   mlr help basic-examples
   mlr help data-formats
-  mlr help function
-  mlr help keyword
+Flags:
+  mlr help flags
+Verbs:
+  mlr help list-verbs
+  mlr help usage-verbs
+  mlr help verb
+Functions:
   mlr help list-functions
   mlr help list-function-classes
   mlr help list-functions-in-class
-  mlr help list-functions-as-paragraph
-  mlr help list-keywords
-  mlr help list-keywords-as-paragraph
-  mlr help list-verbs
-  mlr help list-verbs-as-paragraph
-  mlr help mlrrc
-  mlr help number-formatting
-  mlr help type-arithmetic-info
   mlr help usage-functions
   mlr help usage-functions-by-class
+  mlr help function
+Keywords:
+  mlr help list-keywords
   mlr help usage-keywords
-  mlr help usage-verbs
-  mlr help verb
-  mlr help comments-in-data
-  mlr help compressed-data
-  mlr help data-format-options
-  mlr help double-quoting
-  mlr help format-conversion
-  mlr help separator-options
+  mlr help keyword
+Other:
+  mlr help auxents
+  mlr help mlrrc
   mlr help output-colorization
+  mlr help type-arithmetic-info
 Shorthands:
+  mlr -g = mlr help flags
   mlr -l = mlr help list-verbs
   mlr -L = mlr help usage-verbs
   mlr -f = mlr help list-functions
@@ -121,8 +117,14 @@ Shorthands:
 
 Etc.
 
+## Command-line flags
+
+This is a command-line version of the [List of command-line flags](reference-main-flag-list.md) page.
+See `mlr help flags` for a full listing.
+
 ## Per-verb help
 
+This is a command-line version of the [List of verbs](reference-verbs.md) page.
 Given the name of a verb (from `mlr -l`) you can invoke it with `--help` or `-h` -- or, use `mlr help verb`:
 
 <pre class="pre-highlight-in-pair">
@@ -177,6 +179,7 @@ Etc.
 
 ## Per-function help
 
+This is a command-line version of the [DSL built-in functions](reference-dsl-builtin-functions.md) page.
 Given the name of a DSL function (from `mlr -f`) you can use `mlr help function` for details:
 
 <pre class="pre-highlight-in-pair">
diff --git a/docs6/docs/online-help.md.in b/docs6/docs/online-help.md.in
index fc3858505..d7f0d45cc 100644
--- a/docs6/docs/online-help.md.in
+++ b/docs6/docs/online-help.md.in
@@ -20,8 +20,14 @@ GENMD_EOF
 
 Etc.
 
+## Command-line flags
+
+This is a command-line version of the [List of command-line flags](reference-main-flag-list.md) page.
+See `mlr help flags` for a full listing.
+
 ## Per-verb help
 
+This is a command-line version of the [List of verbs](reference-verbs.md) page.
 Given the name of a verb (from `mlr -l`) you can invoke it with `--help` or `-h` -- or, use `mlr help verb`:
 
 GENMD_RUN_COMMAND
@@ -40,6 +46,7 @@ Etc.
 
 ## Per-function help
 
+This is a command-line version of the [DSL built-in functions](reference-dsl-builtin-functions.md) page.
 Given the name of a DSL function (from `mlr -f`) you can use `mlr help function` for details:
 
 GENMD_RUN_COMMAND
diff --git a/docs6/docs/proofreads.txt b/docs6/docs/proofreads.txt
index 33a6d44ef..d614d862c 100644
--- a/docs6/docs/proofreads.txt
+++ b/docs6/docs/proofreads.txt
@@ -9,10 +9,6 @@ E flatten/unflatten page
 ? twi-dm re all-contribs: all-contributors.org
 
 C flags LUTs
-  o bootstrap internal-only list-sections and help-for-section and a prototype autogen.rb
-  o implement autogen for docs6 flags page
-    - internal-only help-verbs as support for docs6 autogen
-    - audit backticking
   o audit `mlr --list` section output
     - file-format top
     - flatten/unflatten top
@@ -21,12 +17,30 @@ C flags LUTs
     - then nilabend on empty help
   o connect to climain & remove old if-else-if chains
     - needs forReader / forWriter flags
-  o width-32 programmatic w/ splits on "\n"
-    - uncolor " or "
   o assert non-nulls on all flag/section/table elements
   o width-80 audits on mlr --list
   o desc-to-chars map
   o default_xses
+  o work off all TODOs in src/auxents/help/entry.go, src/cli/flag_types.go, src/cli/option_parse.go
+    - also just be happy w/ the code
+  ? mlr help topics autogen namee from FLAG_TABLE -- ?
+
+// ----------------------------------------------------------------
+// TODO: REMOVE
+// Callsites:
+// * src/climain/mlrcli_parse.go
+//   ParseCommandLine
+//     MainOptions (--cpuprofile, --version, etc)
+//     ParseReaderOptions
+//     ParseWriterOptions
+//     ParseReaderWriterOptions
+//     ParseMiscOptions
+//     help.ParseTerminalUsage
+// * handleMlrrcLine
+// * nest/tee/join/put/filter:
+//     ParseReaderOptions
+//     ParseWriterOptions
+//   !! must use only cli package, not climain package
 
 * nikos materials -> fold in
 
diff --git a/docs6/docs/reference-main-flag-list.md b/docs6/docs/reference-main-flag-list.md
index 30f209b7f..e6be89a7a 100644
--- a/docs6/docs/reference-main-flag-list.md
+++ b/docs6/docs/reference-main-flag-list.md
@@ -162,8 +162,6 @@ are overridden in all cases by setting output format to `format2`.
 `: Use XTAB format for input data.
 * `--json or -j
 `: Use JSON format for input and output data.
-* `--jsonx
-`: TODO
 * `--nidx
 `: Use NIDX format for input and output data.
 * `--oasv or --oasvlite
@@ -176,8 +174,6 @@ are overridden in all cases by setting output format to `format2`.
 `: Use DKVP format for output data.
 * `--ojson
 `: Use JSON format for output data.
-* `--ojsonx
-`: TODO
 * `--omd
 `: Use markdown-tabular format for output data.
 * `--onidx
@@ -268,10 +264,12 @@ They are accepted as no-op flags in order to keep old scripts from breaking.
 `: Type information from JSON input files is now preserved throughout the processing stream.
 * `--json-fatal-arrays-on-input
 `: Miller now supports arrays as of version 6.
-* `--json-skip-arrays-on-input
+* `--json-map-arrays-on-input
 `: Miller now supports arrays as of version 6.
 * `--json-skip-arrays-on-input
 `: Miller now supports arrays as of version 6.
+* `--jsonx
+`: The `--jvstack` flag is now default true in Miller 6.
 * `--jvquoteall
 `: Type information from JSON input files is now preserved throughout the processing stream.
 * `--mmap
@@ -280,6 +278,8 @@ They are accepted as no-op flags in order to keep old scripts from breaking.
 `: The current implementation of Miller does not use buffered output, so there is no longer anything to suppress here.
 * `--no-mmap
 `: Miller no longer uses memory-mapping to access data files.
+* `--ojsonx
+`: The `--jvstack` flag is now default true in Miller 6.
 
 ## Miscellaneous flags
 
@@ -294,8 +294,8 @@ They are accepted as no-op flags in order to keep old scripts from breaking.
 `: Use this to specify one of more input files before the verb(s), rather than after. May be used more than once.  The list of filename must end with `--`. This is useful for example since `--from *.csv` doesn't do what you might hope but `--mfrom *.csv --` does.
 * `--mload {filenames}
 `: Like `--load` but works with more than one filename, e.g. `--mload *.mlr --`.
-* `--ofmt
-`: 
+* `--ofmt {format}
+`: E.g. %.18f, %.0f, %9.6e. Please use sprintf-style codes for floating-point nummbers. If not specified, default formatting is used.  See also the `fmtnum` function and the `format-values` verb.
 * `--seed {n}
 `: with `n` of the form `12345678` or `0xcafefeed`. For `put`/`filter` `urand`, `urandint`, and `urand32`.
 * `-I
@@ -341,13 +341,13 @@ How you can control colorization:
 
 * Color choices can be specified by using environment variables, or command-line flags,
   with values 0..255:
-  	* `export MLR_KEY_COLOR=208`, `MLR_VALUE_COLOR=33`, etc.:
-    	`MLR_KEY_COLOR` `MLR_VALUE_COLOR` `MLR_PASS_COLOR` `MLR_FAIL_COLOR`
-    	`MLR_REPL_PS1_COLOR` `MLR_REPL_PS2_COLOR` `MLR_HELP_COLOR`
-  	* Command-line flags `--key-color 208`, `--value-color 33`, etc.:
-    	`--key-color` `--value-color` `--pass-color` `--fail-color`
-    	`--repl-ps1-color` `--repl-ps2-color` `--help-color`
-  	* This is particularly useful if your terminal's background color clashes with current settings.
+    * `export MLR_KEY_COLOR=208`, `MLR_VALUE_COLOR=33`, etc.:
+        `MLR_KEY_COLOR` `MLR_VALUE_COLOR` `MLR_PASS_COLOR` `MLR_FAIL_COLOR`
+        `MLR_REPL_PS1_COLOR` `MLR_REPL_PS2_COLOR` `MLR_HELP_COLOR`
+    * Command-line flags `--key-color 208`, `--value-color 33`, etc.:
+        `--key-color` `--value-color` `--pass-color` `--fail-color`
+        `--repl-ps1-color` `--repl-ps2-color` `--help-color`
+    * This is particularly useful if your terminal's background color clashes with current settings.
 
 If environment-variable settings and command-line flags are both provided, the latter take precedence.
 
diff --git a/go/src/auxents/help/entry.go b/go/src/auxents/help/entry.go
index 7dad52838..866a387a5 100644
--- a/go/src/auxents/help/entry.go
+++ b/go/src/auxents/help/entry.go
@@ -21,82 +21,174 @@ import (
 type tZaryHandlerFunc func()
 type tUnaryHandlerFunc func(arg string)
 
-type shorthandInfo struct {
-	shorthand string
-	longhand  string
+type tHandlerLookupTable struct {
+	sections []tHandlerInfoSection
 }
 
-type handlerInfo struct {
-	name             string
-	zaryHandlerFunc  tZaryHandlerFunc
-	unaryHandlerFunc tUnaryHandlerFunc
+type tHandlerInfoSection struct {
+	name         string
+	handlerInfos []tHandlerInfo
 
 	// Some handlers are used only for webdoc/manpage autogen and needn't
 	// clutter up the on-line help experience for the interactive user
 	internal bool
 }
 
+type tHandlerInfo struct {
+	name             string
+	zaryHandlerFunc  tZaryHandlerFunc
+	unaryHandlerFunc tUnaryHandlerFunc
+}
+
+type tShorthandTable struct {
+	shorthandInfos []tShorthandInfo
+}
+
+type tShorthandInfo struct {
+	shorthand string
+	longhand  string
+}
+
 // We get a Golang "initialization loop" if this is defined statically. So, we
 // use a "package init" function.
-var shorthandLookupTable = []shorthandInfo{}
-var handlerLookupTable = []handlerInfo{}
+var handlerLookupTable = tHandlerLookupTable{}
+var shorthandLookupTable = tShorthandTable{}
 
 func init() {
-	// For things like 'mlr -f', invoked through the CLI parser which does not
-	// go through our HelpMain().
-	shorthandLookupTable = []shorthandInfo{
-		{shorthand: "-l", longhand: "list-verbs"},
-		{shorthand: "-L", longhand: "usage-verbs"},
-		{shorthand: "-f", longhand: "list-functions"},
-		{shorthand: "-F", longhand: "usage-functions"},
-		{shorthand: "-k", longhand: "list-keywords"},
-		{shorthand: "-K", longhand: "usage-keywords"},
-	}
 
 	// For things like 'mlr help foo', invoked through the auxent framework
 	// which goes through our HelpMain().
-	handlerLookupTable = []handlerInfo{
-		{name: "topics", zaryHandlerFunc: listTopics},
-		{name: "auxents", zaryHandlerFunc: helpAuxents},
-		{name: "basic-examples", zaryHandlerFunc: helpBasicExamples},
-		{name: "data-formats", zaryHandlerFunc: helpDataFormats},
-		{name: "function", unaryHandlerFunc: helpForFunction},
-		{name: "keyword", unaryHandlerFunc: helpForKeyword},
-		{name: "list-functions", zaryHandlerFunc: listFunctions},
-		{name: "list-function-classes", zaryHandlerFunc: listFunctionClasses},
-		{name: "list-functions-in-class", unaryHandlerFunc: listFunctionsInClass},
-		{name: "list-functions-as-paragraph", zaryHandlerFunc: listFunctionsAsParagraph},
-		{name: "list-keywords", zaryHandlerFunc: listKeywords},
-		{name: "list-keywords-as-paragraph", zaryHandlerFunc: listKeywordsAsParagraph},
-		{name: "list-verbs", zaryHandlerFunc: listVerbs},
-		{name: "list-verbs-as-paragraph", zaryHandlerFunc: listVerbsAsParagraph},
-		{name: "mlrrc", zaryHandlerFunc: helpMlrrc},
-		{name: "number-formatting", zaryHandlerFunc: helpNumberFormatting},
-		{name: "type-arithmetic-info", zaryHandlerFunc: helpTypeArithmeticInfo},
-		{name: "usage-functions", zaryHandlerFunc: usageFunctions},
-		{name: "usage-functions-by-class", zaryHandlerFunc: usageFunctionsByClass},
-		{name: "usage-keywords", zaryHandlerFunc: usageKeywords},
-		{name: "usage-verbs", zaryHandlerFunc: usageVerbs},
-		{name: "verb", unaryHandlerFunc: helpForVerb},
+	handlerLookupTable = tHandlerLookupTable{
+		sections: []tHandlerInfoSection{
+			{
+				name: "Essentials",
+				handlerInfos: []tHandlerInfo{
+					{name: "topics", zaryHandlerFunc: listTopics},
+					{name: "basic-examples", zaryHandlerFunc: helpBasicExamples},
+					{name: "data-formats", zaryHandlerFunc: helpDataFormats},
+				},
+			},
+			{
+				name: "Flags",
+				handlerInfos: []tHandlerInfo{
+					{name: "flags", zaryHandlerFunc: showFlagHelp},
+					// Per-section entries will be computed and installed below
+				},
+			},
+			{
+				name: "Verbs",
+				handlerInfos: []tHandlerInfo{
+					{name: "list-verbs", zaryHandlerFunc: listVerbs},
+					{name: "usage-verbs", zaryHandlerFunc: usageVerbs},
+					{name: "verb", unaryHandlerFunc: helpForVerb},
+				},
+			},
+			{
+				name: "Functions",
+				handlerInfos: []tHandlerInfo{
+					{name: "list-functions", zaryHandlerFunc: listFunctions},
+					{name: "list-function-classes", zaryHandlerFunc: listFunctionClasses},
+					{name: "list-functions-in-class", unaryHandlerFunc: listFunctionsInClass},
+					{name: "usage-functions", zaryHandlerFunc: usageFunctions},
+					{name: "usage-functions-by-class", zaryHandlerFunc: usageFunctionsByClass},
+					{name: "function", unaryHandlerFunc: helpForFunction},
+				},
+			},
+			{
+				name: "Keywords",
+				handlerInfos: []tHandlerInfo{
+					{name: "list-keywords", zaryHandlerFunc: listKeywords},
+					{name: "usage-keywords", zaryHandlerFunc: usageKeywords},
+					{name: "keyword", unaryHandlerFunc: helpForKeyword},
+				},
+			},
+			{
+				name: "Other",
+				handlerInfos: []tHandlerInfo{
+					{name: "auxents", zaryHandlerFunc: helpAuxents},
+					{name: "mlrrc", zaryHandlerFunc: helpMlrrc},
+					{name: "output-colorization", zaryHandlerFunc: helpOutputColorization},
+					{name: "type-arithmetic-info", zaryHandlerFunc: helpTypeArithmeticInfo},
+				},
+			},
+			{
+				name:     "Internal/docgen",
+				internal: true,
+				handlerInfos: []tHandlerInfo{
+					{name: "list-verbs-as-paragraph", zaryHandlerFunc: listVerbsAsParagraph},
+					{name: "list-functions-as-paragraph", zaryHandlerFunc: listFunctionsAsParagraph},
+					{name: "list-keywords-as-paragraph", zaryHandlerFunc: listKeywordsAsParagraph},
+					{name: "list-functions-as-table", zaryHandlerFunc: listFunctionsAsTable},
+					{name: "list-flag-sections", zaryHandlerFunc: listFlagSections},
+					{name: "print-info-for-section", unaryHandlerFunc: printInfoForSection},
+					{name: "list-flags-for-section", unaryHandlerFunc: listFlagsForSection},
+					{name: "show-help-for-section", unaryHandlerFunc: showHelpForSection},
+					{name: "show-help-for-section-via-downdash", unaryHandlerFunc: showHelpForSectionViaDowndash},
+					{name: "show-headline-for-flag", unaryHandlerFunc: showHeadlineForFlag},
+					{name: "show-help-for-flag", unaryHandlerFunc: showHelpForFlag},
+				},
+			},
+		},
+	}
 
-		// TODO: to flags-sections
-		{name: "comments-in-data", zaryHandlerFunc: helpCommentsInData},
-		{name: "compressed-data", zaryHandlerFunc: helpCompressedDataOptions},
-		{name: "data-format-options", zaryHandlerFunc: helpDataFormatOptions},
-		{name: "double-quoting", zaryHandlerFunc: helpDoubleQuoting},
-		{name: "format-conversion", zaryHandlerFunc: helpFormatConversionKeystrokeSaverOptions},
-		{name: "separator-options", zaryHandlerFunc: helpSeparatorOptions},
+	// This is a wee bit clever. The rest of the topics in the table have names
+	// manually keyed in. But we want to produce `mlr help csv-only-flags` for
+	// flag-section named "CSV-only flags", etc. Here we can't key in the names
+	// since we want to compute them dynamically from cli.FLAG_TABLE which is
+	// Miller's wqy of tracking command-line flags.
 
-		// Internal-only
-		{name: "list-functions-as-table", zaryHandlerFunc: listFunctionsAsTable, internal: true},
-		{name: "list-flag-sections", zaryHandlerFunc: listFlagSections, internal: true},
-		{name: "print-info-for-section", unaryHandlerFunc: printInfoForSection, internal: true},
-		{name: "list-flags-for-section", unaryHandlerFunc: listFlagsForSection, internal: true},
-		{name: "show-headline-for-flag", unaryHandlerFunc: showHeadlineForFlag, internal: true},
-		{name: "show-help-for-flag", unaryHandlerFunc: showHelpForFlag, internal: true},
+	//{name: "comments-in-data-flags"},
+	//{name: "compressed-data-flags"},
+	//{name: "csv-only-flags"},
+	//{name: "file-format-flags"},
+	//{name: "flatten-unflatten-flags"},
+	//{name: "format-conversion-keystroke-saver-flags"},
+	//{name: "json-only-flags"},
+	//{name: "legacy-flags"},
+	//{name: "miscellaneous-flags"},
+	//{name: "output-colorization-flags"},
+	//{name: "pprint-only-flags"},
+	//{name: "separator-flags"},
 
-		// TBD: have an info-only handler in addition to flags-section
-		{name: "output-colorization", zaryHandlerFunc: helpOutputColorization},
+	// For this file's topic-lookup table, find and extend the section called "Flags".
+	for i, section := range handlerLookupTable.sections {
+		if section.name != "Flags" {
+			continue
+		}
+
+		// Ask the flags table for a list of flag-section names, downcased and
+		// with spaces replaced with dashes -- "downdashed" -- making the
+		// punctuation/casing style for online help.
+		downdashSectionNames := cli.FLAG_TABLE.GetDowndashSectionNames()
+		// Note: `j, _` rather than `_, downdashSectionName` since the latter
+		// is a data copy while the former allows us to do a reference. The
+		// former won't produce correct lookup-table data.
+		for j, _ := range downdashSectionNames {
+			downdashSectionName := downdashSectionNames[j]
+			// Patch a new entry into the "Flags" section of our lookup table.
+			entry := tHandlerInfo{
+				name: downdashSectionName,
+				// Make a function which passes in "csv-only-flags" etc. to the FLAG_TABLE.
+				zaryHandlerFunc: func() {
+					showHelpForSectionViaDowndash(downdashSectionName)
+				},
+			}
+			handlerLookupTable.sections[i].handlerInfos = append(handlerLookupTable.sections[i].handlerInfos, entry)
+		}
+	}
+
+	// For things like 'mlr -f', invoked through the CLI parser which does not
+	// go through our HelpMain().
+	shorthandLookupTable = tShorthandTable{
+		shorthandInfos: []tShorthandInfo{
+			{shorthand: "-g", longhand: "flags"},
+			{shorthand: "-l", longhand: "list-verbs"},
+			{shorthand: "-L", longhand: "usage-verbs"},
+			{shorthand: "-f", longhand: "list-functions"},
+			{shorthand: "-F", longhand: "usage-functions"},
+			{shorthand: "-k", longhand: "list-keywords"},
+			{shorthand: "-K", longhand: "usage-keywords"},
+		},
 	}
 }
 
@@ -115,25 +207,27 @@ func HelpMain(args []string) int {
 
 	// "mlr help something" where we recognize the something
 	name := args[0]
-	for _, info := range handlerLookupTable {
-		if info.name == name {
-			if info.zaryHandlerFunc != nil {
-				if len(args) != 1 {
-					fmt.Printf("mlr help %s takes no additional argument.\n", name)
+	for _, section := range handlerLookupTable.sections {
+		for _, info := range section.handlerInfos {
+			if info.name == name {
+				if info.zaryHandlerFunc != nil {
+					if len(args) != 1 {
+						fmt.Printf("mlr help %s takes no additional argument.\n", name)
+						return 0
+					}
+					info.zaryHandlerFunc()
 					return 0
 				}
-				info.zaryHandlerFunc()
-				return 0
-			}
-			if info.unaryHandlerFunc != nil {
-				if len(args) < 2 {
-					fmt.Printf("mlr help %s takes at least one required argument.\n", name)
+				if info.unaryHandlerFunc != nil {
+					if len(args) < 2 {
+						fmt.Printf("mlr help %s takes at least one required argument.\n", name)
+						return 0
+					}
+					for _, arg := range args[1:] {
+						info.unaryHandlerFunc(arg)
+					}
 					return 0
 				}
-				for _, arg := range args[1:] {
-					info.unaryHandlerFunc(arg)
-				}
-				return 0
 			}
 		}
 	}
@@ -164,12 +258,14 @@ func ParseTerminalUsage(arg string) bool {
 		return true
 	}
 	// "mlr -l" is shorthand for "mlr help list-verbs", etc.
-	for _, sinfo := range shorthandLookupTable {
+	for _, sinfo := range shorthandLookupTable.shorthandInfos {
 		if sinfo.shorthand == arg {
-			for _, info := range handlerLookupTable {
-				if info.name == sinfo.longhand {
-					info.zaryHandlerFunc()
-					return true
+			for _, section := range handlerLookupTable.sections {
+				for _, info := range section.handlerInfos {
+					if info.name == sinfo.longhand {
+						info.zaryHandlerFunc()
+						return true
+					}
 				}
 			}
 		}
@@ -185,17 +281,25 @@ func handleDefault() {
 // ----------------------------------------------------------------
 func listTopics() {
 	fmt.Println("Type 'mlr help {topic}' for any of the following:")
-	for _, info := range handlerLookupTable {
-		if !info.internal {
-			fmt.Printf("  mlr help %s\n", info.name)
+	for _, section := range handlerLookupTable.sections {
+		if !section.internal {
+			fmt.Printf("%s:\n", section.name)
+			for _, info := range section.handlerInfos {
+				fmt.Printf("  mlr help %s\n", info.name)
+			}
 		}
 	}
 	fmt.Println("Shorthands:")
-	for _, info := range shorthandLookupTable {
+	for _, info := range shorthandLookupTable.shorthandInfos {
 		fmt.Printf("  mlr %s = mlr help %s\n", info.shorthand, info.longhand)
 	}
 }
 
+// ----------------------------------------------------------------
+func showFlagHelp() {
+	cli.FLAG_TABLE.ShowHelp()
+}
+
 // ----------------------------------------------------------------
 func helpAuxents() {
 	fmt.Print(`Miller has a few otherwise-standalone executables packaged within it.
@@ -219,16 +323,6 @@ mlr --icsv --opprint --from example.csv sort -nr index then cut -f shape,quantit
 `)
 }
 
-// ----------------------------------------------------------------
-func helpCommentsInData() {
-	cli.CommentsInDataPrintInfo()
-}
-
-// ----------------------------------------------------------------
-func helpCompressedDataOptions() {
-	cli.CompressedDataPrintInfo()
-}
-
 // ----------------------------------------------------------------
 func helpDataFormats() {
 	fmt.Printf(
@@ -295,36 +389,6 @@ NIDX: implicitly numerically indexed (Unix-toolkit style)
 `)
 }
 
-// ----------------------------------------------------------------
-func helpDataFormatOptions() {
-	cli.FileFormatPrintInfo()
-}
-
-// ----------------------------------------------------------------
-// TBD FOR MILLER 6:
-
-func helpDoubleQuoting() {
-	fmt.Printf("THIS IS STILL WIP FOR MILLER 6\n")
-	fmt.Println(
-		`--quote-all        Wrap all fields in double quotes
---quote-none       Do not wrap any fields in double quotes, even if they have
-                   OFS or ORS in them
---quote-minimal    Wrap fields in double quotes only if they have OFS or ORS
-                   in them (default)
---quote-numeric    Wrap fields in double quotes only if they have numbers
-                   in them
---quote-original   Wrap fields in double quotes if and only if they were
-                   quoted on input. This isn't sticky for computed fields:
-                   e.g. if fields a and b were quoted on input and you do
-                   "put '$c = $a . $b'" then field c won't inherit a or b's
-                   was-quoted-on-input flag.`)
-}
-
-// ----------------------------------------------------------------
-func helpFormatConversionKeystrokeSaverOptions() {
-	cli.FileFormatPrintInfo()
-}
-
 // ----------------------------------------------------------------
 func helpMlrrc() {
 	fmt.Print(
@@ -367,24 +431,6 @@ func helpOutputColorization() {
 	cli.OutputColorizationPrintInfo()
 }
 
-// ----------------------------------------------------------------
-// TBD FOR MILLER 6:
-
-func helpNumberFormatting() {
-	fmt.Printf("THIS IS STILL WIP FOR MILLER 6\n")
-	fmt.Printf("  --ofmt {format}    E.g. %%.18f, %%.0f, %%9.6e. Please use sprintf-style codes for\n")
-	fmt.Printf("                     floating-point nummbers. If not specified, default formatting is used.\n")
-	fmt.Printf("                     See also the fmtnum function within mlr put (mlr --help-all-functions);\n")
-	fmt.Printf("                     see also the format-values function.\n")
-}
-
-// ----------------------------------------------------------------
-// TBD FOR MILLER 6:
-
-func helpSeparatorOptions() {
-	cli.SeparatorPrintInfo()
-}
-
 // ----------------------------------------------------------------
 func helpTypeArithmeticInfo() {
 	mlrvals := []*types.Mlrval{
@@ -420,10 +466,14 @@ func helpTypeArithmeticInfo() {
 }
 
 // ----------------------------------------------------------------
-// listFlagSections is for webdoc/manpage autogen.
+// listFlagSections et al. are for webdoc/manpage autogen in the miller/docs
+// and miller/man subdirectories. Unlike showFlagHelp where all looping over
+// the flags table, its sections, and flags within each section is done within
+// this Go program, by contrast the following few methods expose the hierarchy
+// to standard output, letting the calling programs (nominally Ruby autogen
+// scripts) control their own looping and formatting.
 
 func listFlagSections() {
-	// xxx temp factorization
 	cli.FLAG_TABLE.ListFlagSections()
 }
 
@@ -443,18 +493,66 @@ func listFlagsForSection(sectionName string) {
 	}
 }
 
+// For manpage autogen: just produce text
+func showHelpForSection(sectionName string) {
+	if !cli.FLAG_TABLE.ShowHelpForSection(sectionName) {
+		fmt.Printf(
+			"mlr: flag-section \"%s\" not found. Please use \"mlr help list-flag-sections\" for a list.\n",
+			sectionName)
+	}
+}
+
+// For on-the-fly `mlr help foo-bar-flags` where `Foo-bar flags` is the name of
+// a section in the FLAG_TABLE. See the func-init block at the top of this
+// file.
+func showHelpForSectionViaDowndash(downdashSectionName string) {
+	if !cli.FLAG_TABLE.ShowHelpForSectionViaDowndash(downdashSectionName) {
+		fmt.Printf("mlr: flag-section \"%s\" not found.\n", downdashSectionName)
+	}
+}
+
+// For webdocs autogen: we want the headline separately so we can backtick it.
 func showHeadlineForFlag(flagName string) {
 	if !cli.FLAG_TABLE.ShowHeadlineForFlag(flagName) {
 		fmt.Printf("mlr: flag \"%s\" not found..\n", flagName)
 	}
 }
 
+// For webdocs autogen
 func showHelpForFlag(flagName string) {
 	if !cli.FLAG_TABLE.ShowHelpForFlag(flagName) {
 		fmt.Printf("mlr: flag \"%s\" not found..\n", flagName)
 	}
 }
 
+// ----------------------------------------------------------------
+func listVerbs() {
+	if isatty.IsTerminal(os.Stdout.Fd()) {
+		transformers.ListVerbNamesAsParagraph()
+	} else {
+		transformers.ListVerbNamesVertically()
+	}
+}
+
+func listVerbsAsParagraph() {
+	transformers.ListVerbNamesAsParagraph()
+}
+
+func helpForVerb(arg string) {
+	transformerSetup := transformers.LookUp(arg)
+	if transformerSetup != nil {
+		transformerSetup.UsageFunc(os.Stdout, true, 0)
+	} else {
+		fmt.Printf(
+			"mlr: verb \"%s\" not found. Please use \"mlr help list-verbs\" for a list.\n",
+			arg)
+	}
+}
+
+func usageVerbs() {
+	transformers.UsageVerbs()
+}
+
 // ----------------------------------------------------------------
 func listFunctions() {
 	if isatty.IsTerminal(os.Stdout.Fd()) {
@@ -512,31 +610,3 @@ func usageKeywords() {
 func helpForKeyword(arg string) {
 	cst.UsageForKeyword(arg)
 }
-
-// ----------------------------------------------------------------
-func listVerbs() {
-	if isatty.IsTerminal(os.Stdout.Fd()) {
-		transformers.ListVerbNamesAsParagraph()
-	} else {
-		transformers.ListVerbNamesVertically()
-	}
-}
-
-func listVerbsAsParagraph() {
-	transformers.ListVerbNamesAsParagraph()
-}
-
-func helpForVerb(arg string) {
-	transformerSetup := transformers.LookUp(arg)
-	if transformerSetup != nil {
-		transformerSetup.UsageFunc(os.Stdout, true, 0)
-	} else {
-		fmt.Printf(
-			"mlr: verb \"%s\" not found. Please use \"mlr help list-verbs\" for a list.\n",
-			arg)
-	}
-}
-
-func usageVerbs() {
-	transformers.UsageVerbs()
-}
diff --git a/go/src/cli/flag_types.go b/go/src/cli/flag_types.go
index 58b7fea3f..fb773b4e7 100644
--- a/go/src/cli/flag_types.go
+++ b/go/src/cli/flag_types.go
@@ -1,7 +1,39 @@
-// TODO: comment
-// TODO: note complexity b/c serving many uses: main CLI, .mlrrc, some verbs; OLH/man/docs autogen
-// TODO: why not go flags
-// TODO: auto-alpha
+// ================================================================
+// Miller support for command-line flags.
+//
+// * Flags are used for several purposes:
+//
+//   o Command-line parsing the main mlr program.
+//
+//   o Record-reader and record-writer options for a few verbs such as join and
+//     tee. E.g. `mlr --csv join -f foo.tsv --tsv ...`: the main input files are
+//     CSV but the join-in file is TSV>
+//
+//   o Processing .mlrrc files.
+//
+//   o Autogenerating on-line help for `mlr help flags`.
+//
+//   o Autogenerating the manpage for `man mlr`.
+//
+//   o Autogenerating webdocs (mkdocs).
+//
+// * For these reasons, flags are organized into tables; for documentation
+//   purposes, flags are organized into sections (see src/cli/option_parse.go).
+//
+// * The Flag struct separates out flag name (e.g. `--csv`), any alternate
+//   names (e.g. `-c`), any arguments the flag may take, a help string, and a
+//   command-line parser function.
+//
+// * The tabular structure may seem overwrought; in fact it has been a blessing
+//   to develop the tabular structure since these flags objects need to serve
+//   so many roles as listed above.
+//
+// * I don't use Go flags for a few reasons. The most important one is that I
+//   need to handle repeated flags, e.g. --from can be used more than once for
+//   mlr, and -f/-n/-r etc can be used more than once for mlr sort, etc. I also
+//   insist on total control of flag formatting including alphabetization of
+//   flags for on-line help and documentation systems.
+// ================================================================
 
 package cli
 
@@ -11,36 +43,23 @@ import (
 	"strings"
 
 	"mlr/src/colorizer"
+	"mlr/src/lib"
 )
 
 // ----------------------------------------------------------------
-// Callsites:
-// * src/climain/mlrcli_parse.go
-//   ParseCommandLine
-//     MainOptions (--cpuprofile, --version, etc)
-//     ParseReaderOptions
-//     ParseWriterOptions
-//     ParseReaderWriterOptions
-//     ParseMiscOptions
-//     help.ParseTerminalUsage
-// * handleMlrrcLine
-// * nest/tee/join/put/filter:
-//     ParseReaderOptions
-//     ParseWriterOptions
-//   !! must use only cli package, not cli package
-
-// sections:
-//   how to factor
-//   reader/writer/readerwriter/misc
-//   -> split necessary for verbs & what they do/don't accept
-//   vs
-//   data-format options, --x2y, separators, compressed, comments in data,
-//   csv-specific, number-formatting, other
-//   -> split useful for on-line help
-
-// ----------------------------------------------------------------
-// TODO: comment
+// Data types used within the flags table.
 
+// FlagParser is a function which takes a flag such as `--foo`.
+//
+// * It should assume that a flag.Owns method has already been invoked to be
+//   sure that this function is indeed the right one to call for `--foo`.
+//
+// * The FlagParser function is responsible for advancing *pargi by 1 (if
+//   `--foo`) or 2 (if `--foo bar`), checking to see if argc is long enough in
+//   the latter case, and mutating the options struct.
+//
+// * Successful handling of the flag is indicated by this function making a
+//   non-zero increment of *pargi.
 type FlagParser func(
 	args []string,
 	argc int,
@@ -48,35 +67,62 @@ type FlagParser func(
 	options *TOptions,
 )
 
-type SectionInfoPrinter func()
-
+// ----------------------------------------------------------------
+// FlagTable holds all the flags for Miller, organized into sections.
 type FlagTable struct {
 	sections []*FlagSection
 }
 
+// FlagSection holds all the flags in a given cateogory, where these
+// categories exist for documentation purposes.
+//
+// The name should be right-cased for webdocs. For on-line help and
+// manpage use, it will get fully uppercased.
+//
+// The infoPrinter provides summary/overview for all flags in the
+// section, for on-line help / webdocs.
 type FlagSection struct {
-	name string // TODO: lowercase? capcase? upper? make methods?
-	// xxx common-info func
-	infoPrinter SectionInfoPrinter
+	name        string
+	infoPrinter func()
 	flags       []Flag
 }
 
+// Flag is a container for all runtime as well as documentation information for
+// a flag.
 type Flag struct {
-	// More common case: the flag has just one spelling, like "--ifs".
+	// In most cases, the flag has just one spelling, like "--ifs".
 	name string
 
-	// Less common case: the flag has more than one spelling, like "-h" and "--help",
-	// or "-c" and "--csv".
+	// In some cases, the flag has more than one spelling, like "-h" and
+	// "--help", or "-c" and "--csv". The altNames field can be omitted from
+	// struct initializers, which in Go means it will read as nil.
 	altNames []string
 
-	// If not "", a name for the flag's argument, for on-line help. E.g. the "bar" in ""--foo {bar}".
-	arg  string
+	// If not "", a name for the flag's argument, for on-line help. E.g. the
+	// "bar" in ""--foo {bar}". It should always be written in curly braces.
+	arg string
 
+	// Help string for `mlr help flags`, `man mlr`, and webdocs.
+	// * It should be all one line within the source code. The text will be
+	//   reformatted as a paragraph for on-line help / manpage, so there should
+	//   be no attempt at line-breaking within the help string.
+	// * Any code bits should be marked with backticks. These look OK for
+	//   on-line help / manpage, and render marvelously for webdocs which
+	//   take markdown.
+	// * After changing flags you can run `sh build-go-src-test-man-doc.sh`
+	//   followed by `git diff` to see how the output looks. See also
+	//   the README.md files in the docs6 and man6 directories for how
+	//   to look at the autogenned docs pre-commit.
 	help string
 
+	// A function for parsing the command line, as described above.
 	parser FlagParser
-	// TODO: comment
-	// reader, writer, reader/writer, misc = neither
+
+	// Any flag intended for record-reading only (e.g. for `mlr join`)
+	// should set forReader = true.
+	// Any flag intended for record-writing only (e.g. for `mlr tee`)
+	// should set forWriter = true.
+	// TODO: rethink this to make the normal case non-error-prone.
 	forReader bool
 	forWriter bool
 }
@@ -84,8 +130,9 @@ type Flag struct {
 // ================================================================
 // FlagTable methods
 
-// Sort organizes the sections in the table alphabetically, to make on-line help
-// easier to read.
+// Sort organizes the sections in the table alphabetically, to make on-line
+// help easier to read. This is done from func-init context so on-line help
+// will always be easy to navigate.
 func (ft *FlagTable) Sort() {
 	// Go sort API: for ascending sort, return true if element i < element j.
 	sort.Slice(ft.sections, func(i, j int) bool {
@@ -93,6 +140,10 @@ func (ft *FlagTable) Sort() {
 	})
 }
 
+// Parse is for parsing a flag on the command line. Given say `--foo`, if a
+// Flag object is found which owns the flag, and if its parser accepts it (e.g.
+// `bar` is present and spelt correctly if the flag-parser expects `--foo bar`)
+// then the return value is true, else false.
 func (ft *FlagTable) Parse(
 	args []string,
 	argc int,
@@ -106,16 +157,18 @@ func (ft *FlagTable) Parse(
 				// Let the flag-parser advance *pargi, depending on how many
 				// arguments follow the flag. E.g. `--ifs pipe` will advance
 				// *pargi by 2; `-I` will advance it by 1.
+				oargi := *pargi
 				flag.parser(args, argc, pargi, options)
-				return true
+				nargi := *pargi
+				return nargi > oargi
 			}
 		}
 	}
 	return false
 }
 
-// TODO: more options for OLH
-func (ft *FlagTable) ListTemp() {
+// ShowHelp prints all-in-one on-line help, nominally for `mlr help flags`.
+func (ft *FlagTable) ShowHelp() {
 	for i, section := range ft.sections {
 		if i > 0 {
 			fmt.Println()
@@ -127,14 +180,45 @@ func (ft *FlagTable) ListTemp() {
 	}
 }
 
-// TODO: comment more. For webdoc/manpage autogen.
+// ListFlagSections exposes some of the flags-table structure, so Ruby autogen
+// scripts for on-line help and webdocs can traverse the structure with looping
+// inside their own code.
 func (ft *FlagTable) ListFlagSections() {
 	for _, section := range ft.sections {
 		fmt.Println(section.name)
 	}
 }
 
-// TODO: comment more. For webdoc/manpage autogen.
+// PrintInfoForSection exposes some of the flags-table structure, so Ruby
+// autogen scripts for on-line help and webdocs can traverse the structure with
+// looping inside their own code.
+func (ft *FlagTable) ShowHelpForSection(sectionName string) bool {
+	for _, section := range ft.sections {
+		if sectionName == section.name {
+			section.PrintInfo()
+			section.ShowHelpForFlags()
+			return true
+		}
+	}
+	return false
+}
+
+// TODO: comment
+func (ft *FlagTable) ShowHelpForSectionViaDowndash(downdashSectionName string) bool {
+	for _, section := range ft.sections {
+		if downdashSectionName == section.GetDowndashSectionName() {
+			fmt.Println(colorizer.MaybeColorizeHelp(strings.ToUpper(section.name), true))
+			section.PrintInfo()
+			section.ShowHelpForFlags()
+			return true
+		}
+	}
+	return false
+}
+
+// PrintInfoForSection exposes some of the flags-table structure, so Ruby
+// autogen scripts for on-line help and webdocs can traverse the structure with
+// looping inside their own code.
 func (ft *FlagTable) PrintInfoForSection(sectionName string) bool {
 	for _, section := range ft.sections {
 		if sectionName == section.name {
@@ -145,7 +229,9 @@ func (ft *FlagTable) PrintInfoForSection(sectionName string) bool {
 	return false
 }
 
-// TODO: comment more. For webdoc/manpage autogen.
+// ListFlagsForSection exposes some of the flags-table structure, so Ruby
+// autogen scripts for on-line help and webdocs can traverse the structure with
+// looping inside their own code.
 func (ft *FlagTable) ListFlagsForSection(sectionName string) bool {
 	for _, section := range ft.sections {
 		if sectionName == section.name {
@@ -156,7 +242,10 @@ func (ft *FlagTable) ListFlagsForSection(sectionName string) bool {
 	return false
 }
 
-// TODO: comment more. For webdoc/manpage autogen.
+// Given flag named `--foo`, altName `-f`, and argument spec `{bar}`, the
+// headline is `--foo or -f {bar}`. This is the bit which is highlighted in
+// on-line help; its length is also used for alignment decisions in the on-line
+// help and the manapge.
 func (ft *FlagTable) ShowHeadlineForFlag(flagName string) bool {
 	for _, fs := range ft.sections {
 		for _, flag := range fs.flags {
@@ -169,7 +258,10 @@ func (ft *FlagTable) ShowHeadlineForFlag(flagName string) bool {
 	return false
 }
 
-// TODO: comment more. For webdoc/manpage autogen.
+// TODO: individualize these comments
+// ShowHelpForFlag prints the flag's help-string all on one line.  This is for
+// webdoc usage where the browser does dynamic line-wrapping, as the user
+// resizes the browser window.
 func (ft *FlagTable) ShowHelpForFlag(flagName string) bool {
 	for _, fs := range ft.sections {
 		for _, flag := range fs.flags {
@@ -182,13 +274,25 @@ func (ft *FlagTable) ShowHelpForFlag(flagName string) bool {
 	return false
 }
 
+// Map "CSV-only flags" to "csv-only-flags" etc. for the benefit of per-section
+// help in `mlr help topics`.
+func (ft *FlagTable) GetDowndashSectionNames() []string {
+	downdashSectionNames := make([]string, len(ft.sections))
+	for i, fs := range ft.sections {
+		// Get names like "CSV-only flags" from the FLAG_TABLE.
+		// Downcase and replace spaces with dashes to get names like
+		// "csv-only-flags"
+		downdashSectionNames[i] = fs.GetDowndashSectionName()
+	}
+	return downdashSectionNames
+}
+
 // ================================================================
 // FlagSection methods
 
-// TODO: more options for OLH
-
 // Sort organizes the flags in the section alphabetically, to make on-line help
-// easier to read.
+// easier to read.  This is done from func-init context so on-line help will
+// always be easy to navigate.
 func (fs *FlagSection) Sort() {
 	// Go sort API: for ascending sort, return true if element i < element j.
 	sort.Slice(fs.flags, func(i, j int) bool {
@@ -196,6 +300,17 @@ func (fs *FlagSection) Sort() {
 	})
 }
 
+// ShowHelpForFlags prints all-in-one on-line help, nominally for `mlr help
+// flags`.
+func (fs *FlagSection) ShowHelpForFlags() {
+	for _, flag := range fs.flags {
+		flag.ShowHelp()
+	}
+}
+
+// PrintInfo exposes some of the flags-table structure, so Ruby autogen scripts
+// for on-line help and webdocs can traverse the structure with looping inside
+// their own code.
 func (fs *FlagSection) PrintInfo() {
 	// TODO: remove with nilabend check
 	if fs.infoPrinter != nil {
@@ -204,18 +319,19 @@ func (fs *FlagSection) PrintInfo() {
 	}
 }
 
-// TODO: more options for OLH
+// ListFlags exposes some of the flags-table structure, so Ruby autogen scripts
+// for on-line help and webdocs can traverse the structure with looping inside
+// their own code.
 func (fs *FlagSection) ListFlags() {
 	for _, flag := range fs.flags {
 		fmt.Println(flag.name)
 	}
 }
 
-// TODO: more options for OLH
-func (fs *FlagSection) ShowHelpForFlags() {
-	for _, flag := range fs.flags {
-		flag.ListTemp()
-	}
+// Map "CSV-only flags" to "csv-only-flags" etc. for the benefit of per-section
+// help in `mlr help topics`.
+func (fs *FlagSection) GetDowndashSectionName() string {
+	return strings.ReplaceAll(strings.ToLower(fs.name), " ", "-")
 }
 
 // ================================================================
@@ -236,17 +352,82 @@ func (flag *Flag) Owns(input string) bool {
 	return false
 }
 
-func (flag *Flag) ListTemp() {
-	displayText := fmt.Sprintf("%-31s", flag.GetHeadline())
-	// TODO: abend if flag.help == ""
-	help := flag.help
-	if help == "" {
-		help = "TODO WRITEME"
+// ShowHelp produces formatting for `mlr help flags` and manpage use.
+// Example:
+// * Flag name is `--foo`
+// * altName is `-f`
+// * Argument spec is `{bar}`
+// * Help string is "Lorem ipsum dolor sit amet, consectetur adipiscing elit,
+//   sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim
+//   ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip
+//   ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate
+//   velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat
+//   cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id
+//   est laborum."
+// * The headline (see the GetHeadline function) is `--foo or -f {bar}`.
+// * We place the headline left in a 25-character column, colorized with the
+//   help color.
+// * We format the help text as 55-character lines and place them
+//   to the right.
+// * The result looks like
+//
+//   --foo or -f {bar}        Lorem ipsum dolor sit amet, consectetur adipiscing
+//                            elit, sed do eiusmod tempor incididunt ut labore et
+//                            dolore magna aliqua. Ut enim ad minim veniam, quis
+//                            nostrud exercitation ullamco laboris nisi ut aliquip
+//                            ex ea commodo consequat. Duis aute irure dolor in
+//                            reprehenderit in voluptate velit esse cillum dolore
+//                            eu fugiat nulla pariatur. Excepteur sint occaecat
+//                            cupidatat non proident, sunt in culpa qui officia
+//                            deserunt mollit anim id est laborum.
+//
+// * If the headline is too long we put the first help line a line below like this:
+//
+//   --foo-flag-is-very-very-long {bar}
+//                            Lorem ipsum dolor sit amet, consectetur adipiscing
+//                            elit, sed do eiusmod tempor incididunt ut labore et
+//                            dolore magna aliqua. Ut enim ad minim veniam, quis
+//                            nostrud exercitation ullamco laboris nisi ut aliquip
+//                            ex ea commodo consequat. Duis aute irure dolor in
+//                            reprehenderit in voluptate velit esse cillum dolore
+//                            eu fugiat nulla pariatur. Excepteur sint occaecat
+//                            cupidatat non proident, sunt in culpa qui officia
+//                            deserunt mollit anim id est laborum.
+//
+
+func (flag *Flag) ShowHelp() {
+	headline := flag.GetHeadline()
+	displayHeadline := fmt.Sprintf("%-25s", headline)
+	broken := len(headline) >= 25
+
+	helpLines := lib.FormatAsParagraph(flag.help, 55)
+
+	if broken {
+		fmt.Printf("%s\n", colorizer.MaybeColorizeHelp(displayHeadline, true))
+		for _, helpLine := range helpLines {
+			fmt.Printf("%25s%s\n", " ", helpLine)
+		}
+	} else {
+		fmt.Printf("%s", colorizer.MaybeColorizeHelp(displayHeadline, true))
+		if len(helpLines) == 0 {
+			fmt.Println()
+		}
+		for i, helpLine := range helpLines {
+			if i == 0 {
+				fmt.Printf("%s\n", helpLine)
+			} else {
+				fmt.Printf("%25s%s\n", " ", helpLine)
+			}
+		}
 	}
-	fmt.Printf("%s %s\n", colorizer.MaybeColorizeHelp(displayText, true), help)
 }
 
-// TODO: comment
+// GetHeadline puts together the flag name, any altNames, and any argument spec
+// into a single string for the left column of online help / manpage content.
+// Given flag named `--foo`, altName `-f`, and argument spec `{bar}`, the
+// headline is `--foo or -f {bar}`. This is the bit which is highlighted in
+// on-line help; its length is also used for alignment decisions in the on-line
+// help and the manapge.
 func (flag *Flag) GetHeadline() string {
 	displayNames := make([]string, 1)
 	displayNames[0] = flag.name
@@ -261,6 +442,10 @@ func (flag *Flag) GetHeadline() string {
 	return displayText
 }
 
+// Gets the help string all on one line (just in case anyone typed it in using
+// multiline string-literal backtick notation in Go). This is suitable for
+// webdoc use where we create all one line, and the browser dynamically
+// line-wraps as the user resizes the window.
 func (flag *Flag) GetHelpOneLine() string {
 	return strings.Join(strings.Split(flag.help, "\n"), " ")
 }
@@ -273,5 +458,3 @@ func (flag *Flag) GetHelpOneLine() string {
 func NoOpParse1(args []string, argc int, pargi *int, options *TOptions) {
 	*pargi += 1
 }
-
-var NoOpHelp string = "No-op pass-through for backward compatibility with Miller 5."
diff --git a/go/src/cli/option_parse.go b/go/src/cli/option_parse.go
index be266cce0..93631671d 100644
--- a/go/src/cli/option_parse.go
+++ b/go/src/cli/option_parse.go
@@ -1247,9 +1247,10 @@ func ParseMiscOptions(
 		}
 		argi += 2
 
-	} else if args[argi] == "--list" {
+		// TODO: some terminal/main/something
+	} else if args[argi] == "-g" {
 		argi += 1
-		FLAG_TABLE.ListTemp()
+		FLAG_TABLE.ShowHelp()
 		os.Exit(0)
 
 	}
@@ -1265,6 +1266,7 @@ var FLAG_TABLE = FlagTable{
 		&SeparatorFlagSection,
 		&FileFormatFlagSection,
 		&FormatConversionKeystrokeSaverFlagSection,
+		// TODO: &HelpFlags, here or in climain?
 		&JSONOnlyFlagSection,
 		&CSVOnlyFlagSection,
 		&PPRINTOnlyFlagSection,
@@ -1294,19 +1296,19 @@ TODO: auto-detect is still TBD for Miller 6
 
 Notes about line endings:
 
-* Default line endings (`+"`--irs`"+` and `+"`--ors`"+`) are "auto" which means autodetect from
+* Default line endings (` + "`--irs`" + ` and ` + "`--ors`" + `) are "auto" which means autodetect from
   the input file format, as long as the input file(s) have lines ending in either
-  LF (also known as linefeed, `+"`\\n`"+`, `+"`0x0a`"+`, or Unix-style) or CRLF (also known as
-  carriage-return/linefeed pairs, `+"`\\r\\n`"+`, `+"`0x0d 0x0a`"+`, or Windows-style).
-* If both `+"`irs`"+` and `+"`ors`"+` are `+"`auto`"+` (which is the default) then LF input will lead to LF
+  LF (also known as linefeed, ` + "`\\n`" + `, ` + "`0x0a`" + `, or Unix-style) or CRLF (also known as
+  carriage-return/linefeed pairs, ` + "`\\r\\n`" + `, ` + "`0x0d 0x0a`" + `, or Windows-style).
+* If both ` + "`irs`" + ` and ` + "`ors`" + ` are ` + "`auto`" + ` (which is the default) then LF input will lead to LF
   output and CRLF input will lead to CRLF output, regardless of the platform you're
   running on.
 * The line-ending autodetector triggers on the first line ending detected in the input
   stream. E.g. if you specify a CRLF-terminated file on the command line followed by an
   LF-terminated file then autodetected line endings will be CRLF.
-* If you use `+"`--ors {something else}`"+` with (default or explicitly specified) `+"`--irs auto`"+`
+* If you use ` + "`--ors {something else}`" + ` with (default or explicitly specified) ` + "`--irs auto`" + `
   then line endings are autodetected on input and set to what you specify on output.
-* If you use `+"`--irs {something else}`"+` with (default or explicitly specified) `+"`--ors auto`"+`
+* If you use ` + "`--irs {something else}`" + ` with (default or explicitly specified) ` + "`--ors auto`" + `
   then the output line endings used are LF on Unix/Linux/BSD/MacOSX, and CRLF on Windows.
 
 Notes about all other separators:
@@ -1315,21 +1317,21 @@ Notes about all other separators:
   do key-value pairs appear juxtaposed.
 * IRS/ORS are ignored for XTAB format. Nominally IFS and OFS are newlines;
   XTAB records are separated by two or more consecutive IFS/OFS -- i.e.
-  a blank line. Everything above about `+"`--irs/--ors/--rs auto`"+` becomes `+"`--ifs/--ofs/--fs`"+`
+  a blank line. Everything above about ` + "`--irs/--ors/--rs auto`" + ` becomes ` + "`--ifs/--ofs/--fs`" + `
   auto for XTAB format. (XTAB's default IFS/OFS are "auto".)
 * OFS must be single-character for PPRINT format. This is because it is used
   with repetition for alignment; multi-character separators would make
   alignment impossible.
 * OPS may be multi-character for XTAB format, in which case alignment is
   disabled.
-* TSV is simply CSV using tab as field separator (`+"`--fs tab`"+`).
+* TSV is simply CSV using tab as field separator (` + "`--fs tab`" + `).
 * FS/PS are ignored for markdown format; RS is used.
 * All FS and PS options are ignored for JSON format, since they are not relevant
   to the JSON format.
 * You can specify separators in any of the following ways, shown by example:
   - Type them out, quoting as necessary for shell escapes, e.g.
-    `+"`--fs '|' --ips :`"+`
-  - C-style escape sequences, e.g. `+"`--rs '\\r\\n' --fs '\\t'`"+`.
+    ` + "`--fs '|' --ips :`" + `
+  - C-style escape sequences, e.g. ` + "`--rs '\\r\\n' --fs '\\t'`" + `.
   - To avoid backslashing, you can use any of the following names:
 	  TODO desc-to-chars map
 
@@ -1359,7 +1361,7 @@ var SeparatorFlagSection = FlagSection{
 
 		{
 			name: "--ifs",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Specify FS for input.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1373,7 +1375,7 @@ var SeparatorFlagSection = FlagSection{
 
 		{
 			name: "--ips",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Specify PS for input.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1387,7 +1389,7 @@ var SeparatorFlagSection = FlagSection{
 
 		{
 			name: "--irs",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Specify RS for input.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1412,7 +1414,7 @@ var SeparatorFlagSection = FlagSection{
 
 		{
 			name: "--ors",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Specify RS for output.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1424,7 +1426,7 @@ var SeparatorFlagSection = FlagSection{
 
 		{
 			name: "--ofs",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Specify FS for output.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1436,7 +1438,7 @@ var SeparatorFlagSection = FlagSection{
 
 		{
 			name: "--ops",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Specify PS for output.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1448,7 +1450,7 @@ var SeparatorFlagSection = FlagSection{
 
 		{
 			name: "--rs",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Specify RS for input and output.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1464,7 +1466,7 @@ var SeparatorFlagSection = FlagSection{
 
 		{
 			name: "--fs",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Specify FS for input and output.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1480,7 +1482,7 @@ var SeparatorFlagSection = FlagSection{
 
 		{
 			name: "--ps",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Specify PS for input and output.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1506,7 +1508,7 @@ func JSONOnlyPrintInfo() {
 func init() { JSONOnlyFlagSection.Sort() }
 
 var JSONOnlyFlagSection = FlagSection{
-	name: "JSON-only flags",
+	name:        "JSON-only flags",
 	infoPrinter: JSONOnlyPrintInfo,
 	flags: []Flag{
 
@@ -1531,16 +1533,15 @@ var JSONOnlyFlagSection = FlagSection{
 		},
 
 		{
-			name: "--jlistwrap",
+			name:     "--jlistwrap",
 			altNames: []string{"--jl"},
-			help: "Wrap JSON output in outermost `[ ]`.",
+			help:     "Wrap JSON output in outermost `[ ]`.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				options.WriterOptions.WrapJSONOutputInOuterList = true
 				*pargi += 1
 			},
 			forWriter: true,
 		},
-
 	},
 }
 
@@ -1554,7 +1555,7 @@ func PPRINTOnlyPrintInfo() {
 func init() { PPRINTOnlyFlagSection.Sort() }
 
 var PPRINTOnlyFlagSection = FlagSection{
-	name: "PPRINT-only flags",
+	name:        "PPRINT-only flags",
 	infoPrinter: PPRINTOnlyPrintInfo,
 	flags: []Flag{
 
@@ -1589,31 +1590,42 @@ They are accepted as no-op flags in order to keep old scripts from breaking.`)
 func init() { LegacyFlagSection.Sort() }
 
 var LegacyFlagSection = FlagSection{
-	name: "Legacy flags",
+	name:        "Legacy flags",
 	infoPrinter: LegacyFlagInfoPrint,
 	flags: []Flag{
 
 		{
 			name:      "--mmap",
-			help: "Miller no longer uses memory-mapping to access data files.",
+			help:      "Miller no longer uses memory-mapping to access data files.",
 			parser:    NoOpParse1,
 			forReader: true,
 		},
 
 		{
 			name:      "--no-mmap",
-			help: "Miller no longer uses memory-mapping to access data files.",
+			help:      "Miller no longer uses memory-mapping to access data files.",
 			parser:    NoOpParse1,
 			forReader: true,
 		},
 
 		{
 			name:      "--no-fflush",
-			help: "The current implementation of Miller does not use buffered output, so there is no longer anything to suppress here.",
+			help:      "The current implementation of Miller does not use buffered output, so there is no longer anything to suppress here.",
 			parser:    NoOpParse1,
 			forWriter: true,
 		},
 
+		{
+			name:   "--jsonx",
+			help:   "The `--jvstack` flag is now default true in Miller 6.",
+			parser: NoOpParse1,
+		},
+		{
+			name:   "--ojsonx",
+			help:   "The `--jvstack` flag is now default true in Miller 6.",
+			parser: NoOpParse1,
+		},
+
 		{
 			name:   "--jknquoteint",
 			help:   "Type information from JSON input files is now preserved throughout the processing stream.",
@@ -1639,7 +1651,7 @@ var LegacyFlagSection = FlagSection{
 		},
 
 		{
-			name:   "--json-skip-arrays-on-input",
+			name:   "--json-map-arrays-on-input",
 			help:   "Miller now supports arrays as of version 6.",
 			parser: NoOpParse1,
 		},
@@ -1659,12 +1671,12 @@ func FileFormatPrintInfo() {
 	// TODO
 	fmt.Println(`TO DO: brief list of formats w/ xref to m6 webdocs.
 
-Examples: `+"`--csv`"+` for CSV-formatted input and output; `+"`--icsv --opprint`"+` for
+Examples: ` + "`--csv`" + ` for CSV-formatted input and output; ` + "`--icsv --opprint`" + ` for
 CSV-formatted input and pretty-printed output.
 
-Please use `+"`--iformat1 --oformat2`"+` rather than `+"`--format1 --oformat2`"+`.
-The latter sets up input and output flags for `+"`format1`"+`, not all of which
-are overridden in all cases by setting output format to `+"`format2`"+`.`)
+Please use ` + "`--iformat1 --oformat2`" + ` rather than ` + "`--format1 --oformat2`" + `.
+The latter sets up input and output flags for ` + "`format1`" + `, not all of which
+are overridden in all cases by setting output format to ` + "`format2`" + `.`)
 }
 
 //--idkvp   --odkvp   --dkvp      Delimited key-value pairs, e.g "a=1,b=2"
@@ -1708,10 +1720,6 @@ are overridden in all cases by setting output format to `+"`format2`"+`.`)
 //--ijson   --ojson   --json      JSON tabular: sequence or list of one-level
 //                                maps: {...}{...} or [{...},{...}].
 
-//              --jsonx --ojsonx  Keystroke-savers for --json --jvstack
-
-//              --jsonx --ojsonx  and --ojson --jvstack, respectively.
-
 func init() { FileFormatFlagSection.Sort() }
 
 var FileFormatFlagSection = FlagSection{
@@ -1835,7 +1843,7 @@ var FileFormatFlagSection = FlagSection{
 
 		{
 			name: "-i",
-			arg: "{format name}",
+			arg:  "{format name}",
 			help: "Use format name for input data. For example: `-i csv` is the same as `--icsv`.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1890,7 +1898,7 @@ var FileFormatFlagSection = FlagSection{
 
 		{
 			name: "-o",
-			arg: "{format name}",
+			arg:  "{format name}",
 			help: "Use format name for output data.  For example: `-o csv` is the same as `--ocsv`.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -1993,15 +2001,6 @@ var FileFormatFlagSection = FlagSection{
 				*pargi += 1
 			},
 		},
-		{
-			name: "--ojsonx",
-			help: "TODO", // move to legacy
-			parser: func(args []string, argc int, pargi *int, options *TOptions) {
-				// --jvstack is now the default in Miller 6 so this is just for backward compatibility
-				options.WriterOptions.OutputFileFormat = "json"
-				*pargi += 1
-			},
-		},
 
 		{
 			name: "--onidx",
@@ -2034,7 +2033,7 @@ var FileFormatFlagSection = FlagSection{
 
 		{
 			name: "--io",
-			arg: "{format name}",
+			arg:  "{format name}",
 			help: "Use format name for input and output data. For example: `--io csv` is the same as `--csv`.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -2159,16 +2158,6 @@ var FileFormatFlagSection = FlagSection{
 				*pargi += 1
 			},
 		},
-		{
-			name: "--jsonx",
-			help: "TODO", // move to legacy
-			parser: func(args []string, argc int, pargi *int, options *TOptions) {
-				// --jvstack is now the default in Miller 6 so this is just for backward compatibility
-				options.ReaderOptions.InputFileFormat = "json"
-				options.WriterOptions.OutputFileFormat = "json"
-				*pargi += 1
-			},
-		},
 
 		{
 			name: "--nidx",
@@ -2212,17 +2201,20 @@ var FileFormatFlagSection = FlagSection{
 // FORMAT-CONVERSION KEYSTROKE-SAVER FLAGS
 
 func FormatConversionKeystrokeSaverPrintInfo() {
-	fmt.Println(`As keystroke-savers for format-conversion you may use the following:
-      --c2t --c2d --c2n --c2j --c2x --c2p --c2m
---t2c       --t2d --t2n --t2j --t2x --t2p --t2m
---d2c --d2t       --d2n --d2j --d2x --d2p --d2m
---n2c --n2t --n2d       --n2j --n2x --n2p --n2m
---j2c --j2t --j2d --j2n       --j2x --j2p --j2m
---x2c --x2t --x2d --x2n --x2j       --x2p --x2m
---p2c --p2t --p2d --p2n --p2j --p2x       --p2m
-The letters c t d n j x p m refer to formats CSV, TSV, DKVP, NIDX, JSON, XTAB,
-PPRINT, and markdown, respectively. Note that markdown format is available for
-output only.`)
+	fmt.Println(`As keystroke-savers for format-conversion you may use the following.
+The letters c, t, j, d, n, x, p, and m refer to formats CSV, TSV, DKVP, NIDX,
+JSON, XTAB, PPRINT, and markdown, respectively. Note that markdown format is
+available for output only.
+
+| In \ out | CSV   | TSV   | JSON   | DKVP   | NIDX   | XTAB   | PPRINT | Markdown |
+| CSV      |       | --c2t | --c2j  | --c2d  | --c2n  | --c2x  | --c2p  | --c2m    |
+| TSV      | --t2c |       | --t2j  | --t2d  | --t2n  | --t2x  | --t2p  | --t2m    |
+| JSON     | --j2c | --j2t |        | --j2d  | --j2n  | --j2x  | --j2p  | --j2m    |
+| DKVP     | --d2c | --d2t | --d2j  |        | --d2n  | --d2x  | --d2p  | --d2m    |
+| NIDX     | --n2c | --n2t | --n2j  | --n2d  |        | --n2x  | --n2p  | --n2m    |
+| XTAB     | --x2c | --x2t | --x2j  | --x2d  | --x2n  |        | --x2p  | --x2m    |
+| PPRINT   | --p2c | --p2t | --p2j  | --p2d  | --p2n  | --p2x  |        | --p2m    |
+`)
 }
 
 func init() { FormatConversionKeystrokeSaverFlagSection.Sort() }
@@ -2896,7 +2888,7 @@ var CSVOnlyFlagSection = FlagSection{
 		{
 			name:     "--allow-ragged-csv-input",
 			altNames: []string{"--ragged"},
-			help: "If a data line has fewer fields than the header line, fill remaining keys with empty string. If a data line has more fields than the header line, use integer field labels as in the implicit-header case.",
+			help:     "If a data line has fewer fields than the header line, fill remaining keys with empty string. If a data line has more fields than the header line, use integer field labels as in the implicit-header case.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				options.ReaderOptions.AllowRaggedCSVInput = true
 				*pargi += 1
@@ -2967,6 +2959,23 @@ var CSVOnlyFlagSection = FlagSection{
 		//	},
 		//},
 
+		//func helpDoubleQuoting() {
+		//    fmt.Printf("THIS IS STILL WIP FOR MILLER 6\n")
+		//    fmt.Println(
+		//        `--quote-all        Wrap all fields in double quotes
+		//--quote-none       Do not wrap any fields in double quotes, even if they have
+		//                   OFS or ORS in them
+		//--quote-minimal    Wrap fields in double quotes only if they have OFS or ORS
+		//                   in them (default)
+		//--quote-numeric    Wrap fields in double quotes only if they have numbers
+		//                   in them
+		//--quote-original   Wrap fields in double quotes if and only if they were
+		//                   quoted on input. This isn't sticky for computed fields:
+		//                   e.g. if fields a and b were quoted on input and you do
+		//                   "put '$c = $a . $b'" then field c won't inherit a or b's
+		//                   was-quoted-on-input flag.`)
+		//}
+
 	},
 }
 
@@ -2976,14 +2985,14 @@ var CSVOnlyFlagSection = FlagSection{
 func CompressedDataPrintInfo() {
 	fmt.Print(`Miller offers a few different ways to handle reading data files which have been compressed.
 
-* Decompression done within the Miller process itself: `+"`--bz2in`"+` `+"`--gzin`"+` `+"`--zin`"+`
-* Decompression done outside the Miller process: `+"`--prepipe`"+` `+"`--prepipex`"+`
+* Decompression done within the Miller process itself: ` + "`--bz2in`" + ` ` + "`--gzin`" + ` ` + "`--zin`" + `
+* Decompression done outside the Miller process: ` + "`--prepipe`" + ` ` + "`--prepipex`" + `
 
-Using `+"`--prepipe`"+` and `+"`--prepipex`"+` you can specify an action to be
+Using ` + "`--prepipe`" + ` and ` + "`--prepipex`" + ` you can specify an action to be
 taken on each input file.  The prepipe command must be able to read from
-standard input; it will be invoked with `+"`{command} < {filename}`"+`.  The
+standard input; it will be invoked with ` + "`{command} < {filename}`" + `.  The
 prepipex command must take a filename as argument; it will be invoked with
-`+"`{command} {filename}`"+`.
+` + "`{command} {filename}`" + `.
 
 Examples:
 
@@ -2995,11 +3004,11 @@ Examples:
 Note that this feature is quite general and is not limited to decompression
 utilities. You can use it to apply per-file filters of your choice.  For output
 compression (or other) utilities, simply pipe the output:
-`+"`mlr ... | {your compression command} > outputfilenamegoeshere`"+`
+` + "`mlr ... | {your compression command} > outputfilenamegoeshere`" + `
 
-Lastly, note that if `+"`--prepipe`"+` or `+"`--prepipex`"+` is specified, it replaces any
+Lastly, note that if ` + "`--prepipe`" + ` or ` + "`--prepipex`" + ` is specified, it replaces any
 decisions that might have been made based on the file suffix. Likewise,
-`+"`--gzin`"+`/`+"`--bz2in`"+`/`+"`--zin`"+` are ignored if `+"`--prepipe`"+` is also specified.
+` + "`--gzin`" + `/` + "`--bz2in`" + `/` + "`--zin`" + ` are ignored if ` + "`--prepipe`" + ` is also specified.
 `)
 }
 
@@ -3012,7 +3021,7 @@ var CompressedDataFlagSection = FlagSection{
 
 		{
 			name: "--prepipe",
-			arg: "{decompression command}",
+			arg:  "{decompression command}",
 			help: "You can, of course, already do without this for single input files, e.g. `gunzip < myfile.csv.gz | mlr ...`.  Allowed at the command line, but not in `.mlrrc` to avoid unexpected code execution.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -3024,7 +3033,7 @@ var CompressedDataFlagSection = FlagSection{
 
 		{
 			name: "--prepipex",
-			arg: "{decompression command}",
+			arg:  "{decompression command}",
 			help: "Like `--prepipe` with one exception: doesn't insert `<` between command and filename at runtime. Useful for some commands like `unzip -qc` which don't read standard input.  Allowed at the command line, but not in `.mlrrc` to avoid unexpected code execution.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -3125,7 +3134,7 @@ var CommentsInDataFlagSection = FlagSection{
 
 		{
 			name: "--skip-comments",
-			help: "Ignore commented lines (prefixed by `"+DEFAULT_COMMENT_STRING+"`) within the input.",
+			help: "Ignore commented lines (prefixed by `" + DEFAULT_COMMENT_STRING + "`) within the input.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				options.ReaderOptions.CommentString = DEFAULT_COMMENT_STRING
 				options.ReaderOptions.CommentHandling = SkipComments
@@ -3135,7 +3144,7 @@ var CommentsInDataFlagSection = FlagSection{
 
 		{
 			name: "--skip-comments-with",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Ignore commented lines within input, with specified prefix.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -3147,7 +3156,7 @@ var CommentsInDataFlagSection = FlagSection{
 
 		{
 			name: "--pass-comments",
-			help: "Immediately print commented lines (prefixed by `"+DEFAULT_COMMENT_STRING+"`) within the input.",
+			help: "Immediately print commented lines (prefixed by `" + DEFAULT_COMMENT_STRING + "`) within the input.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				options.ReaderOptions.CommentString = DEFAULT_COMMENT_STRING
 				options.ReaderOptions.CommentHandling = PassComments
@@ -3157,7 +3166,7 @@ var CommentsInDataFlagSection = FlagSection{
 
 		{
 			name: "--pass-comments-with",
-			arg: "{string}",
+			arg:  "{string}",
 			help: "Immediately print commented lines within input, with specified prefix.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -3185,16 +3194,16 @@ Things having colors:
 Rules for coloring:
 
 * By default, colorize output only if writing to stdout and stdout is a TTY.
-    * Example: color: `+"`mlr --csv cat foo.csv`"+`
-    * Example: no color: `+"`mlr --csv cat foo.csv > bar.csv`"+`
-    * Example: no color: `+"`mlr --csv cat foo.csv | less`"+`
+    * Example: color: ` + "`mlr --csv cat foo.csv`" + `
+    * Example: no color: ` + "`mlr --csv cat foo.csv > bar.csv`" + `
+    * Example: no color: ` + "`mlr --csv cat foo.csv | less`" + `
 * The default colors were chosen since they look OK with white or black terminal background,
   and are differentiable with common varieties of human color vision.
 
 Mechanisms for coloring:
 
 * Miller uses ANSI escape sequences only. This does not work on Windows except within Cygwin.
-* Requires `+"`TERM`"+` environment variable to be set to non-empty string.
+* Requires ` + "`TERM`" + ` environment variable to be set to non-empty string.
 * Doesn't try to check to see whether the terminal is capable of 256-color
   ANSI vs 16-color ANSI. Note that if colors are in the range 0..15
   then 16-color ANSI escapes are used, so this is in the user's control.
@@ -3202,25 +3211,25 @@ Mechanisms for coloring:
 How you can control colorization:
 
 * Suppression/unsuppression:
-    * Environment variable `+"`export MLR_NO_COLOR=true`"+` means don't color even if stdout+TTY.
-    * Environment variable `+"`export MLR_ALWAYS_COLOR=true`"+` means do color even if not stdout+TTY.
-      For example, you might want to use this when piping mlr output to `+"`less -r`"+`.
-    * Command-line flags `+"`--no-color`"+` or `+"`-M`"+`, `+"`--always-color`"+` or `+"`-C`"+`.
+    * Environment variable ` + "`export MLR_NO_COLOR=true`" + ` means don't color even if stdout+TTY.
+    * Environment variable ` + "`export MLR_ALWAYS_COLOR=true`" + ` means do color even if not stdout+TTY.
+      For example, you might want to use this when piping mlr output to ` + "`less -r`" + `.
+    * Command-line flags ` + "`--no-color`" + ` or ` + "`-M`" + `, ` + "`--always-color`" + ` or ` + "`-C`" + `.
 
 * Color choices can be specified by using environment variables, or command-line flags,
   with values 0..255:
-  	* `+"`export MLR_KEY_COLOR=208`"+`, `+"`MLR_VALUE_COLOR=33`"+`, etc.:
-    	`+"`MLR_KEY_COLOR`"+` `+"`MLR_VALUE_COLOR`"+` `+"`MLR_PASS_COLOR`"+` `+"`MLR_FAIL_COLOR`"+`
-    	`+"`MLR_REPL_PS1_COLOR`"+` `+"`MLR_REPL_PS2_COLOR`"+` `+"`MLR_HELP_COLOR`"+`
-  	* Command-line flags `+"`--key-color 208`"+`, `+"`--value-color 33`"+`, etc.:
-    	`+"`--key-color`"+` `+"`--value-color`"+` `+"`--pass-color`"+` `+"`--fail-color`"+`
-    	`+"`--repl-ps1-color`"+` `+"`--repl-ps2-color`"+` `+"`--help-color`"+`
-  	* This is particularly useful if your terminal's background color clashes with current settings.
+    * ` + "`export MLR_KEY_COLOR=208`" + `, ` + "`MLR_VALUE_COLOR=33`" + `, etc.:
+        ` + "`MLR_KEY_COLOR`" + ` ` + "`MLR_VALUE_COLOR`" + ` ` + "`MLR_PASS_COLOR`" + ` ` + "`MLR_FAIL_COLOR`" + `
+        ` + "`MLR_REPL_PS1_COLOR`" + ` ` + "`MLR_REPL_PS2_COLOR`" + ` ` + "`MLR_HELP_COLOR`" + `
+    * Command-line flags ` + "`--key-color 208`" + `, ` + "`--value-color 33`" + `, etc.:
+        ` + "`--key-color`" + ` ` + "`--value-color`" + ` ` + "`--pass-color`" + ` ` + "`--fail-color`" + `
+        ` + "`--repl-ps1-color`" + ` ` + "`--repl-ps2-color`" + ` ` + "`--help-color`" + `
+    * This is particularly useful if your terminal's background color clashes with current settings.
 
 If environment-variable settings and command-line flags are both provided, the latter take precedence.
 
-Please do mlr `+"`--list-color-codes`"+` to see the available color codes (like 170), and
-`+"`mlr --list-color-names`"+` to see available names (like `+"`orchid`"+`).
+Please do mlr ` + "`--list-color-codes`" + ` to see the available color codes (like 170), and
+` + "`mlr --list-color-names`" + ` to see available names (like ` + "`orchid`" + `).
 `)
 }
 
@@ -3356,8 +3365,8 @@ var FlattenUnflattenFlagSection = FlagSection{
 		{
 			name:     "--flatsep",
 			altNames: []string{"--jflatsep", "--oflatsep"}, // TODO: really need all for miller5 back-compat?
-			arg: "{string}",
-			help: "Separator for flattening multi-level JSON keys, e.g. `{\"a\":{\"b\":3}}` becomes `a:b => 3` for non-JSON formats. Defaults to `"+DEFAULT_JSON_FLATTEN_SEPARATOR+"`.",
+			arg:      "{string}",
+			help:     "Separator for flattening multi-level JSON keys, e.g. `{\"a\":{\"b\":3}}` becomes `a:b => 3` for non-JSON formats. Defaults to `" + DEFAULT_JSON_FLATTEN_SEPARATOR + "`.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
 				options.WriterOptions.FLATSEP = SeparatorFromArg(args[*pargi+1])
@@ -3429,7 +3438,7 @@ var MiscFlagSection = FlagSection{
 
 		{
 			name: "--from",
-			arg: "{filename}",
+			arg:  "{filename}",
 			help: "Use this to specify an input file before the verb(s), rather than after. May be used more than once. Example: `mlr --from a.dat --from b.dat cat` is the same as `mlr cat a.dat b.dat`.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -3440,7 +3449,7 @@ var MiscFlagSection = FlagSection{
 
 		{
 			name: "--mfrom",
-			arg: "{filenames}",
+			arg:  "{filenames}",
 			help: "Use this to specify one of more input files before the verb(s), rather than after. May be used more than once.  The list of filename must end with `--`. This is useful for example since `--from *.csv` doesn't do what you might hope but `--mfrom *.csv --` does.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -3457,6 +3466,8 @@ var MiscFlagSection = FlagSection{
 
 		{
 			name: "--ofmt",
+			arg:  "{format}",
+			help: "E.g. %.18f, %.0f, %9.6e. Please use sprintf-style codes for floating-point nummbers. If not specified, default formatting is used.  See also the `fmtnum` function and the `format-values` verb.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
 				options.WriterOptions.FPOFMT = args[*pargi+1]
@@ -3467,7 +3478,7 @@ var MiscFlagSection = FlagSection{
 		// TODO: move to another (or new) section
 		{
 			name: "--load",
-			arg: "{filename}",
+			arg:  "{filename}",
 			help: "Load DSL script file for all put/filter operations on the command line.  If the name following `--load` is a directory, load all `*.mlr` files in that directory. This is just like `put -f` and `filter -f` except it's up-front on the command line, so you can do something like `alias mlr='mlr --load ~/myscripts'` if you like.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -3478,7 +3489,7 @@ var MiscFlagSection = FlagSection{
 
 		{
 			name: "--mload",
-			arg: "{filenames}",
+			arg:  "{filenames}",
 			help: "Like `--load` but works with more than one filename, e.g. `--mload *.mlr --`.",
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
 				CheckArgCount(args, *pargi, argc, 2)
@@ -3518,7 +3529,7 @@ var MiscFlagSection = FlagSection{
 
 		{
 			name: "--seed",
-			arg: "{n}",
+			arg:  "{n}",
 			help: "with `n` of the form `12345678` or `0xcafefeed`. For `put`/`filter` `urand`, `urandint`, and `urand32`.",
 
 			parser: func(args []string, argc int, pargi *int, options *TOptions) {
diff --git a/go/src/climain/mlrcli_parse.go b/go/src/climain/mlrcli_parse.go
index 074a87a77..f32d1fbdb 100644
--- a/go/src/climain/mlrcli_parse.go
+++ b/go/src/climain/mlrcli_parse.go
@@ -49,18 +49,18 @@ func ParseCommandLine(args []string) (
 			os.Exit(0)
 
 			// TODO
-			//		} else if cli.FLAG_TABLE.Parse(args, argc, &argi, &options) {
-			//			// handled
+		} else if cli.FLAG_TABLE.Parse(args, argc, &argi, &options) {
+			// handled
 
-		} else if cli.ParseReaderOptions(args, argc, &argi, &options.ReaderOptions) {
-			// handled
-		} else if cli.ParseWriterOptions(args, argc, &argi, &options.WriterOptions) {
-			// handled
-		} else if cli.ParseReaderWriterOptions(args, argc, &argi,
-			&options.ReaderOptions, &options.WriterOptions) {
-			// handled
-		} else if cli.ParseMiscOptions(args, argc, &argi, &options) {
-			// handled
+			//} else if cli.ParseReaderOptions(args, argc, &argi, &options.ReaderOptions) {
+			//// handled
+			//} else if cli.ParseWriterOptions(args, argc, &argi, &options.WriterOptions) {
+			//// handled
+			//} else if cli.ParseReaderWriterOptions(args, argc, &argi,
+			//&options.ReaderOptions, &options.WriterOptions) {
+			//// handled
+			//} else if cli.ParseMiscOptions(args, argc, &argi, &options) {
+			//// handled
 		} else {
 			// unhandled
 			fmt.Fprintf(os.Stderr, "%s: option \"%s\" not recognized.\n", "mlr", args[argi])
diff --git a/go/src/lib/paragraph.go b/go/src/lib/paragraph.go
index 4ca9f7409..b608461a2 100644
--- a/go/src/lib/paragraph.go
+++ b/go/src/lib/paragraph.go
@@ -1,7 +1,9 @@
 package lib
 
 import (
+	"bytes"
 	"fmt"
+	"strings"
 )
 
 // For online help contexts like printing all the built-in DSL functions, or
@@ -31,3 +33,39 @@ func PrintWordsAsParagraph(words []string) {
 
 	fmt.Printf("\n")
 }
+
+// For online help contexts like printing all the built-in DSL functions, or
+// the list of all verbs. Max width is nominally 80.
+func FormatAsParagraph(text string, maxWidth int) []string {
+	lines := make([]string, 0)
+	words := strings.Fields(text)
+
+	separator := " "
+	separatorlen := len(separator)
+	linelen := 0
+	j := 0
+
+	var buffer bytes.Buffer
+	for _, word := range words {
+		wordlen := len(word)
+		linelen += separatorlen + wordlen
+		if linelen >= maxWidth {
+			line := buffer.String()
+			lines = append(lines, line)
+			buffer.Reset()
+			linelen = separatorlen + wordlen
+			j = 0
+		}
+		if j > 0 {
+			buffer.WriteString(separator)
+		}
+		buffer.WriteString(word)
+		j++
+	}
+	line := buffer.String()
+	if line != "" {
+		lines = append(lines, line)
+	}
+
+	return lines
+}
diff --git a/man6/manpage.txt b/man6/manpage.txt
index b8ad2fe58..59e1314dd 100644
--- a/man6/manpage.txt
+++ b/man6/manpage.txt
@@ -7,9 +7,9 @@ NAME
        as CSV and tabular JSON.
 
 SYNOPSIS
-       Usage: mlr [I/O options] {verb} [verb-dependent options ...] {zero or
-       more file names} Output of one verb may be chained as input to another
-       using "then", e.g.
+       Usage: mlr [flags] {verb} [verb-dependent options ...] {zero or more
+       file names} Output of one verb may be chained as input to another using
+       "then", e.g.
          mlr stats1 -a min,mean,max -f flag,u,v -g color then sort -f color
        Please see 'mlr help topics' for more information.  Please also see
        https://johnkerl.org/miller6
@@ -95,6 +95,43 @@ DATA FORMATS
        | fox jumped          | Record 2: "1":"fox", "2":"jumped"
        +---------------------+
 
+HELP OPTIONS
+       Type 'mlr help {topic}' for any of the following:
+       Essentials:
+         mlr help topics
+         mlr help basic-examples
+         mlr help data-formats
+       Flags:
+         mlr help flags
+       Verbs:
+         mlr help list-verbs
+         mlr help usage-verbs
+         mlr help verb
+       Functions:
+         mlr help list-functions
+         mlr help list-function-classes
+         mlr help list-functions-in-class
+         mlr help usage-functions
+         mlr help usage-functions-by-class
+         mlr help function
+       Keywords:
+         mlr help list-keywords
+         mlr help usage-keywords
+         mlr help keyword
+       Other:
+         mlr help auxents
+         mlr help mlrrc
+         mlr help output-colorization
+         mlr help type-arithmetic-info
+       Shorthands:
+         mlr -g = mlr help flags
+         mlr -l = mlr help list-verbs
+         mlr -L = mlr help usage-verbs
+         mlr -f = mlr help list-functions
+         mlr -F = mlr help usage-functions
+         mlr -k = mlr help list-keywords
+         mlr -K = mlr help usage-keywords
+
 VERB LIST
        altkv bar bootstrap cat check clean-whitespace count-distinct count
        count-similar cut decimate fill-down fill-empty filter flatten format-values
@@ -128,117 +165,171 @@ FUNCTION LIST
        version ! != !=~ % & && * ** + - . .* .+ .- ./ / // < << <= == =~ > >= >> >>>
        ?: ?? ??? ^ ^^ | || ~
 
-HELP OPTIONS
-       Type 'mlr help {topic}' for any of the following:
-         mlr help topics
-         mlr help auxents
-         mlr help basic-examples
-         mlr help comments-in-data
-         mlr help compressed-data
-         mlr help csv-options
-         mlr help data-format-options
-         mlr help data-formats
-         mlr help double-quoting
-         mlr help format-conversion
-         mlr help function
-         mlr help keyword
-         mlr help list-functions
-         mlr help list-function-classes
-         mlr help list-functions-in-class
-         mlr help list-functions-as-paragraph
-         mlr help list-functions-as-table
-         mlr help list-keywords
-         mlr help list-keywords-as-paragraph
-         mlr help list-verbs
-         mlr help list-verbs-as-paragraph
-         mlr help misc
-         mlr help mlrrc
-         mlr help number-formatting
-         mlr help output-colorization
-         mlr help separator-options
-         mlr help type-arithmetic-info
-         mlr help usage-functions
-         mlr help usage-functions-by-class
-         mlr help usage-keywords
-         mlr help usage-verbs
-         mlr help verb
-       Shorthands:
-         mlr -l = mlr help list-verbs
-         mlr -L = mlr help usage-verbs
-         mlr -f = mlr help list-functions
-         mlr -F = mlr help usage-functions
-         mlr -k = mlr help list-keywords
-         mlr -K = mlr help usage-keywords
+COMMENTS-IN-DATA FLAGS
+       Miller lets you put comments in your data, such as
 
-OPTIONS
-       In the following option flags, the version with "i" designates the
-       input stream, "o" the output stream, and the version without prefix
-       sets the option for both input and output stream. For example: --irs
-       sets the input record separator, --ors the output record separator, and
-       --rs sets both the input and output separator to the given value.
+           # This is a comment for a CSV file
+           a,b,c
+           1,2,3
+           4,5,6
 
-   DATA-FORMAT OPTIONS
-       --idkvp   --odkvp   --dkvp      Delimited key-value pairs, e.g "a=1,b=2"
-                                        (Miller's default format).
+       Notes:
 
-       --inidx   --onidx   --nidx      Implicitly-integer-indexed fields (Unix-toolkit style).
-       -T                              Synonymous with "--nidx --fs tab".
+       * Comments are only honored at the start of a line.
+       * In the absence of any of the below four options, comments are data like
+         any other text. (The comments-in-data feature is opt-in.)
+       * When `--pass-comments` is used, comment lines are written to standard output
+         immediately upon being read; they are not part of the record stream.  Results
+         may be counterintuitive. A suggestion is to place comments at the start of
+         data files.
 
-       --icsv    --ocsv    --csv       Comma-separated value (or tab-separated with --fs tab, etc.)
+       --pass-comments          Immediately print commented lines (prefixed by `#`)
+                                within the input.
+       --pass-comments-with {string}
+                                Immediately print commented lines within input, with
+                                specified prefix.
+       --skip-comments          Ignore commented lines (prefixed by `#`) within the
+                                input.
+       --skip-comments-with {string}
+                                Ignore commented lines within input, with specified
+                                prefix.
 
-       --itsv    --otsv    --tsv       Keystroke-savers for "--icsv --ifs tab",
-                                       "--ocsv --ofs tab", "--csv --fs tab".
-       --iasv    --oasv    --asv       Similar but using ASCII FS 0x1f and RS 0x1e\n",
-       --iusv    --ousv    --usv       Similar but using Unicode FS U+241F (UTF-8 0xe2909f)\n",
-                                       and RS U+241E (UTF-8 0xe2909e)\n",
+COMPRESSED-DATA FLAGS
+       Miller offers a few different ways to handle reading data files which have been compressed.
 
-       --icsvlite --ocsvlite --csvlite Comma-separated value (or tab-separated with --fs tab, etc.).
-                                              The 'lite' CSV does not handle RFC-CSV double-quoting rules; is
-                                              slightly faster and handles heterogeneity in the input stream via
-                                              empty newline followed by new header line. See also
-                                               https://johnkerl.org/miller6/file-formats.html#csv-tsv-asv-usv-etc
+       * Decompression done within the Miller process itself: `--bz2in` `--gzin` `--zin`
+       * Decompression done outside the Miller process: `--prepipe` `--prepipex`
 
-       --itsvlite --otsvlite --tsvlite Keystroke-savers for "--icsvlite --ifs tab",
-                                       "--ocsvlite --ofs tab", "--csvlite --fs tab".
-       -t                              Synonymous with --tsvlite.
-       --iasvlite --oasvlite --asvlite Similar to --itsvlite et al. but using ASCII FS 0x1f and RS 0x1e\n",
-       --iusvlite --ousvlite --usvlite Similar to --itsvlite et al. but using Unicode FS U+241F (UTF-8 0xe2909f)\n",
-                                       and RS U+241E (UTF-8 0xe2909e)\n",
+       Using `--prepipe` and `--prepipex` you can specify an action to be
+       taken on each input file.  The prepipe command must be able to read from
+       standard input; it will be invoked with `{command} < {filename}`.  The
+       prepipex command must take a filename as argument; it will be invoked with
+       `{command} {filename}`.
 
-       --ipprint --opprint --pprint    Pretty-printed tabular (produces no
-                                       output until all input is in).
-                           --right     Right-justifies all fields for PPRINT output.
-                           --barred    Prints a border around PPRINT output
-                                       (only available for output).
+       Examples:
 
-                 --omd                 Markdown-tabular (only available for output).
+           mlr --prepipe gunzip
+           mlr --prepipe zcat -cf
+           mlr --prepipe xz -cd
+           mlr --prepipe cat
 
-       --ixtab   --oxtab   --xtab      Pretty-printed vertical-tabular.
-                           --xvright   Right-justifies values for XTAB format.
+       Note that this feature is quite general and is not limited to decompression
+       utilities. You can use it to apply per-file filters of your choice.  For output
+       compression (or other) utilities, simply pipe the output:
+       `mlr ... | {your compression command} > outputfilenamegoeshere`
 
-       --ijson   --ojson   --json      JSON tabular: sequence or list of one-level
-                                       maps: {...}{...} or [{...},{...}].
-                           --jvstack   Put one key-value pair per line for JSON output.
-                        --no-jvstack   Put objects/arrays all on one line for JSON output.
-                     --jsonx --ojsonx  Keystroke-savers for --json --jvstack
-                     --jsonx --ojsonx  and --ojson --jvstack, respectively.
-                           --jlistwrap Wrap JSON output in outermost [ ].
-                   --flatsep {string} Separator for flattening multi-level JSON keys,
-                                       e.g. '{"a":{"b":3}}' becomes a:b => 3 for
-                                       non-JSON formats. Defaults to ..\n",
+       Lastly, note that if `--prepipe` or `--prepipex` is specified, it replaces any
+       decisions that might have been made based on the file suffix. Likewise,
+       `--gzin`/`--bz2in`/`--zin` are ignored if `--prepipe` is also specified.
 
-       -p is a keystroke-saver for --nidx --fs space --repifs
+       --bz2in                  Uncompress bzip2 within the Miller process. Done by
+                                default if file ends in `.bz2`.
+       --gzin                   Uncompress gzip within the Miller process. Done by
+                                default if file ends in `.gz`.
+       --prepipe {decompression command}
+                                You can, of course, already do without this for
+                                single input files, e.g. `gunzip < myfile.csv.gz |
+                                mlr ...`. Allowed at the command line, but not in
+                                `.mlrrc` to avoid unexpected code execution.
+       --prepipe-bz2            Same as `--prepipe bz2`, except this is allowed in
+                                `.mlrrc`.
+       --prepipe-gunzip         Same as `--prepipe gunzip`, except this is allowed in
+                                `.mlrrc`.
+       --prepipe-zcat           Same as `--prepipe zcat`, except this is allowed in
+                                `.mlrrc`.
+       --prepipex {decompression command}
+                                Like `--prepipe` with one exception: doesn't insert
+                                `<` between command and filename at runtime. Useful
+                                for some commands like `unzip -qc` which don't read
+                                standard input. Allowed at the command line, but not
+                                in `.mlrrc` to avoid unexpected code execution.
+       --zin                    Uncompress zlib within the Miller process. Done by
+                                default if file ends in `.z`.
 
-       Examples: --csv for CSV-formatted input and output; --icsv --opprint for
+CSV-ONLY FLAGS
+       --allow-ragged-csv-input or --ragged
+                                If a data line has fewer fields than the header line,
+                                fill remaining keys with empty string. If a data line
+                                has more fields than the header line, use integer
+                                field labels as in the implicit-header case.
+       --headerless-csv-output  Print only CSV data lines; do not print CSV header
+                                lines.
+       --implicit-csv-header    Use 1,2,3,... as field labels, rather than from line
+                                1 of input files. Tip: combine with `label` to
+                                recreate missing headers.
+       --no-implicit-csv-header Opposite of `--implicit-csv-header`. This is the
+                                default anyway -- the main use is for the flags to
+                                `mlr join` if you have main file(s) which are
+                                headerless but you want to join in on a file which
+                                does have a CSV header. Then you could use `mlr --csv
+                                --implicit-csv-header join --no-implicit-csv-header
+                                -l your-join-in-with-header.csv ...
+                                your-headerless.csv`.
+       -N                       Keystroke-saver for `--implicit-csv-header
+                                --headerless-csv-output`.
+
+FILE-FORMAT FLAGS
+       TO DO: brief list of formats w/ xref to m6 webdocs.
+
+       Examples: `--csv` for CSV-formatted input and output; `--icsv --opprint` for
        CSV-formatted input and pretty-printed output.
 
-       Please use --iformat1 --oformat2 rather than --format1 --oformat2.
-       The latter sets up input and output flags for format1, not all of which
-       are overridden in all cases by setting output format to format2.
+       Please use `--iformat1 --oformat2` rather than `--format1 --oformat2`.
+       The latter sets up input and output flags for `format1`, not all of which
+       are overridden in all cases by setting output format to `format2`.
 
-   FORMAT-CONVERSION KEYSTROKE-SAVERS
+       --asv or --asvlite       Use ASV format for input and output data.
+       --csv or -c              Use CSV format for input and output data.
+       --csvlite                Use CSV-lite format for input and output data.
+       --dkvp                   Use DKVP format for input and output data.
+       --iasv or --iasvlite     Use ASV format for input data.
+       --icsv                   Use CSV format for input data.
+       --icsvlite               Use CSV-lite format for input data.
+       --idkvp                  Use DKVP format for input data.
+       --ijson                  Use JSON format for input data.
+       --inidx                  Use NIDX format for input data.
+       --io {format name}       Use format name for input and output data. For
+                                example: `--io csv` is the same as `--csv`.
+       --ipprint                Use PPRINT format for input data.
+       --itsv                   Use TSV format for input data.
+       --itsvlite               Use TSV-lite format for input data.
+       --iusv or --iusvlite     Use USV format for input data.
+       --ixtab                  Use XTAB format for input data.
+       --json or -j             Use JSON format for input and output data.
+       --nidx                   Use NIDX format for input and output data.
+       --oasv or --oasvlite     Use ASV format for output data.
+       --ocsv                   Use CSV format for output data.
+       --ocsvlite               Use CSV-lite format for output data.
+       --odkvp                  Use DKVP format for output data.
+       --ojson                  Use JSON format for output data.
+       --omd                    Use markdown-tabular format for output data.
+       --onidx                  Use NIDX format for output data.
+       --opprint                Use PPRINT format for output data.
+       --otsv                   Use TSV format for output data.
+       --otsvlite               Use TSV-lite format for output data.
+       --ousv or --ousvlite     Use USV format for output data.
+       --oxtab                  Use XTAB format for output data.
+       --pprint                 Use PPRINT format for input and output data.
+       --tsv                    Use TSV format for input and output data.
+       --tsvlite or -t          Use TSV-lite format for input and output data.
+       --usv or --usvlite       Use USV format for input and output data.
+       --xtab                   Use XTAB format for input and output data.
+       -i {format name}         Use format name for input data. For example: `-i csv`
+                                is the same as `--icsv`.
+       -o {format name}         Use format name for output data. For example: `-o
+                                csv` is the same as `--ocsv`.
+
+FLATTEN-UNFLATTEN FLAGS
+       --flatsep or --jflatsep or --oflatsep {string}
+                                Separator for flattening multi-level JSON keys, e.g.
+                                `{"a":{"b":3}}` becomes `a:b => 3` for non-JSON
+                                formats. Defaults to `.`.
+       --no-auto-flatten
+       --no-auto-unflatten
+
+FORMAT-CONVERSION KEYSTROKE-SAVER FLAGS
        As keystroke-savers for format-conversion you may use the following:
-       --c2t --c2d --c2n --c2j --c2x --c2p --c2m
+             --c2t --c2d --c2n --c2j --c2x --c2p --c2m
        --t2c       --t2d --t2n --t2j --t2x --t2p --t2m
        --d2c --d2t       --d2n --d2j --d2x --d2p --d2m
        --n2c --n2t --n2d       --n2j --n2x --n2p --n2m
@@ -249,130 +340,270 @@ OPTIONS
        PPRINT, and markdown, respectively. Note that markdown format is available for
        output only.
 
-   SEPARATORS
-       THIS IS STILL TBD FOR MILLER 6
+       --c2b                    Use CSV for input, PPRINT with `--barred` for output.
+       --c2d                    Use CSV for input, DKVP for output.
+       --c2j                    Use CSV for input, JSON for output.
+       --c2m                    Use CSV for input, markdown-tabular for output.
+       --c2n                    Use CSV for input, NIDX for output.
+       --c2p                    Use CSV for input, PPRINT for output.
+       --c2t                    Use CSV for input, TSV for output.
+       --c2x                    Use CSV for input, XTAB for output.
+       --d2b                    Use DKVP for input, PPRINT with `--barred` for
+                                output.
+       --d2c                    Use DKVP for input, CSV for output.
+       --d2j                    Use DKVP for input, JSON for output.
+       --d2m                    Use DKVP for input, markdown-tabular for output.
+       --d2n                    Use DKVP for input, NIDX for output.
+       --d2p                    Use DKVP for input, PPRINT for output.
+       --d2t                    Use DKVP for input, TSV for output.
+       --d2x                    Use DKVP for input, XTAB for output.
+       --j2b                    Use JSON for input, PPRINT with --barred for output.
+       --j2c                    Use JSON for input, CSV for output.
+       --j2d                    Use JSON for input, DKVP for output.
+       --j2m                    Use JSON for input, markdown-tabular for output.
+       --j2n                    Use JSON for input, NIDX for output.
+       --j2p                    Use JSON for input, PPRINT for output.
+       --j2t                    Use JSON for input, TSV for output.
+       --j2x                    Use JSON for input, XTAB for output.
+       --n2b                    Use NIDX for input, PPRINT with `--barred` for
+                                output.
+       --n2c                    Use NIDX for input, CSV for output.
+       --n2d                    Use NIDX for input, DKVP for output.
+       --n2j                    Use NIDX for input, JSON for output.
+       --n2m                    Use NIDX for input, markdown-tabular for output.
+       --n2p                    Use NIDX for input, PPRINT for output.
+       --n2t                    Use NIDX for input, TSV for output.
+       --n2x                    Use NIDX for input, XTAB for output.
+       --p2c                    Use PPRINT for input, CSV for output.
+       --p2d                    Use PPRINT for input, DKVP for output.
+       --p2j                    Use PPRINT for input, JSON for output.
+       --p2m                    Use PPRINT for input, markdown-tabular for output.
+       --p2n                    Use PPRINT for input, NIDX for output.
+       --p2t                    Use PPRINT for input, TSV for output.
+       --p2x                    Use PPRINT for input, XTAB for output.
+       --t2b                    Use TSV for input, PPRINT with `--barred` for output.
+       --t2c                    Use TSV for input, CSV for output.
+       --t2d                    Use TSV for input, DKVP for output.
+       --t2j                    Use TSV for input, JSON for output.
+       --t2m                    Use TSV for input, markdown-tabular for output.
+       --t2n                    Use TSV for input, NIDX for output.
+       --t2p                    Use TSV for input, PPRINT for output.
+       --t2x                    Use TSV for input, XTAB for output.
+       --x2b                    Use XTAB for input, PPRINT with `--barred` for
+                                output.
+       --x2c                    Use XTAB for input, CSV for output.
+       --x2d                    Use XTAB for input, DKVP for output.
+       --x2j                    Use XTAB for input, JSON for output.
+       --x2m                    Use XTAB for input, markdown-tabular for output.
+       --x2n                    Use XTAB for input, NIDX for output.
+       --x2p                    Use XTAB for input, PPRINT for output.
+       --x2t                    Use XTAB for input, TSV for output.
+       -p                       Keystroke-saver for `--nidx --fs space --repifs`.
+       -T                       Keystroke-saver for `--nidx --fs tab`.
 
-   COMPRESSED I/O
-       Decompression done within the Miller process itself:
-       --gzin  Uncompress gzip within the Miller process. Done by default if file ends in ".gz".
-       --bz2in Uncompress bz2ip within the Miller process. Done by default if file ends in ".bz2".
-       --zin   Uncompress zlib within the Miller process. Done by default if file ends in ".z".
+JSON-ONLY FLAGS
+       These are flags which are applicable to JSON format.
 
-       Decompression done outside the Miller process:
-       --prepipe {command} You can, of course, already do without this for single input files,
-         e.g. "gunzip < myfile.csv.gz | mlr ..."
-       --prepipex {command} Like --prepipe with one exception: doesn't insert '<' between
-         command and filename at runtime. Useful for some commands like 'unzip -qc'
-         which don't read standard input.
+       --jlistwrap or --jl      Wrap JSON output in outermost `[ ]`.
+       --jvstack                Put one key-value pair per line for JSON output
+                                (multi-line output).
+       --no-jvstack             Put objects/arrays all on one line for JSON output.
 
-       Using --prepipe and --prepipex you can specify an action to be taken on each
-       input file. This prepipe command must be able to read from standard input; it
-       will be invoked with {command} < {filename}.
+LEGACY FLAGS
+       These are flags which don't do anything in the current Miller version.
+       They are accepted as no-op flags in order to keep old scripts from breaking.
 
-       Examples:
-         mlr --prepipe gunzip
-         mlr --prepipe zcat -cf
-         mlr --prepipe xz -cd
-         mlr --prepipe cat
+       --jknquoteint            Type information from JSON input files is now
+                                preserved throughout the processing stream.
+       --jquoteall              Type information from JSON input files is now
+                                preserved throughout the processing stream.
+       --json-fatal-arrays-on-input
+                                Miller now supports arrays as of version 6.
+       --json-map-arrays-on-input
+                                Miller now supports arrays as of version 6.
+       --json-skip-arrays-on-input
+                                Miller now supports arrays as of version 6.
+       --jsonx                  The `--jvstack` flag is now default true in Miller 6.
+       --jvquoteall             Type information from JSON input files is now
+                                preserved throughout the processing stream.
+       --mmap                   Miller no longer uses memory-mapping to access data
+                                files.
+       --no-fflush              The current implementation of Miller does not use
+                                buffered output, so there is no longer anything to
+                                suppress here.
+       --no-mmap                Miller no longer uses memory-mapping to access data
+                                files.
+       --ojsonx                 The `--jvstack` flag is now default true in Miller 6.
 
-       Note that this feature is quite general and is not limited to decompression
-       utilities. You can use it to apply per-file filters of your choice.  For output
-       compression (or other) utilities, simply pipe the output:
-       mlr ... | {your compression command} > outputfilenamegoeshere
+MISCELLANEOUS FLAGS
+       --from {filename}        Use this to specify an input file before the verb(s),
+                                rather than after. May be used more than once.
+                                Example: `mlr --from a.dat --from b.dat cat` is the
+                                same as `mlr cat a.dat b.dat`.
+       --load {filename}        Load DSL script file for all put/filter operations on
+                                the command line. If the name following `--load` is a
+                                directory, load all `*.mlr` files in that directory.
+                                This is just like `put -f` and `filter -f` except
+                                it's up-front on the command line, so you can do
+                                something like `alias mlr='mlr --load ~/myscripts'`
+                                if you like.
+       --mfrom {filenames}      Use this to specify one of more input files before
+                                the verb(s), rather than after. May be used more than
+                                once. The list of filename must end with `--`. This
+                                is useful for example since `--from *.csv` doesn't do
+                                what you might hope but `--mfrom *.csv --` does.
+       --mload {filenames}      Like `--load` but works with more than one filename,
+                                e.g. `--mload *.mlr --`.
+       --ofmt {format}          E.g. %.18f, %.0f, %9.6e. Please use sprintf-style
+                                codes for floating-point nummbers. If not specified,
+                                default formatting is used. See also the `fmtnum`
+                                function and the `format-values` verb.
+       --seed {n}               with `n` of the form `12345678` or `0xcafefeed`. For
+                                `put`/`filter` `urand`, `urandint`, and `urand32`.
+       -I                       Process files in-place. For each file name on the
+                                command line, output is written to a temp file in the
+                                same directory, which is then renamed over the
+                                original. Each file is processed in isolation: if the
+                                output format is CSV, CSV headers will be present in
+                                each output file, statistics are only over each
+                                file's own records; and so on.
+       -n                       Process no input files, nor standard input either.
+                                Useful for `mlr put` with `begin`/`end` statements
+                                only. (Same as `--from /dev/null`.) Also useful in
+                                `mlr -n put -v '...'` for analyzing abstract syntax
+                                trees (if that's your thing).
 
-       Lastly, note that if --prepipe or --prepipex is specified, it replaces any
-       decisions that might have been made based on the file suffix. Also,
-       --gzin/--bz2in/--zin are ignored if --prepipe is also specified.
+OUTPUT-COLORIZATION FLAGS
+       Miller uses colors to highlight outputs. You can specify color preferences.
+       Note: output colorization does not work on Windows.
 
-   COMMENTS IN DATA
-       --skip-comments                 Ignore commented lines (prefixed by "#")
-                                       within the input.
-       --skip-comments-with {string}   Ignore commented lines within input, with
-                                       specified prefix.
-       --pass-comments                 Immediately print commented lines (prefixed by "#")
-                                       within the input.
-       --pass-comments-with {string}   Immediately print commented lines within input, with
-                                       specified prefix.
+       Things having colors:
 
-       Notes:
-       * Comments are only honored at the start of a line.
-       * In the absence of any of the above four options, comments are data like
-         any other text.
-       * When pass-comments is used, comment lines are written to standard output
-         immediately upon being read; they are not part of the record stream.  Results
-         may be counterintuitive. A suggestion is to place comments at the start of
-         data files.
+       * Keys in CSV header lines, JSON keys, etc
+       * Values in CSV data lines, JSON scalar values, etc in regression-test output
+       * Some online-help strings
 
-   CSV-SPECIFIC OPTIONS
-       --implicit-csv-header Use 1,2,3,... as field labels, rather than from line 1
-                          of input files. Tip: combine with "label" to recreate
-                          missing headers.
-       --no-implicit-csv-header Do not use --implicit-csv-header. This is the default
-                          anyway -- the main use is for the flags to 'mlr join' if you have
-                          main file(s) which are headerless but you want to join in on
-                          a file which does have a CSV header. Then you could use
-                          'mlr --csv --implicit-csv-header join --no-implicit-csv-header
-                          -l your-join-in-with-header.csv ... your-headerless.csv'
-       --allow-ragged-csv-input|--ragged If a data line has fewer fields than the header line,
-                          fill remaining keys with empty string. If a data line has more
-                          fields than the header line, use integer field labels as in
-                          the implicit-header case.
-       --headerless-csv-output   Print only CSV data lines.
-       -N                 Keystroke-saver for --implicit-csv-header --headerless-csv-output.
+       Rules for coloring:
 
-   DOUBLE-QUOTING FOR CSV/CSVLITE OUTPUT
-       THIS IS STILL WIP FOR MILLER 6
-       --quote-all        Wrap all fields in double quotes
-       --quote-none       Do not wrap any fields in double quotes, even if they have
-                          OFS or ORS in them
-       --quote-minimal    Wrap fields in double quotes only if they have OFS or ORS
-                          in them (default)
-       --quote-numeric    Wrap fields in double quotes only if they have numbers
-                          in them
-       --quote-original   Wrap fields in double quotes if and only if they were
-                          quoted on input. This isn't sticky for computed fields:
-                          e.g. if fields a and b were quoted on input and you do
-                          "put '$c = $a . $b'" then field c won't inherit a or b's
-                          was-quoted-on-input flag.
+       * By default, colorize output only if writing to stdout and stdout is a TTY.
+           * Example: color: `mlr --csv cat foo.csv`
+           * Example: no color: `mlr --csv cat foo.csv > bar.csv`
+           * Example: no color: `mlr --csv cat foo.csv | less`
+       * The default colors were chosen since they look OK with white or black terminal background,
+         and are differentiable with common varieties of human color vision.
 
-   NUMBER FORMATTING
-       THIS IS STILL WIP FOR MILLER 6
-         --ofmt {format}    E.g. %.18f, %.0f, %9.6e. Please use sprintf-style codes for
-                            floating-point nummbers. If not specified, default formatting is used.
-                            See also the fmtnum function within mlr put (mlr --help-all-functions);
-                            see also the format-values function.
+       Mechanisms for coloring:
 
-   OTHER OPTIONS
-       --seed {n} with n of the form 12345678 or 0xcafefeed. For put/filter
-                          urand()/urandint()/urand32().
-       --nr-progress-mod {m}, with m a positive integer: print filename and record
-                          count to os.Stderr every m input records.
-       --from {filename}  Use this to specify an input file before the verb(s),
-                          rather than after. May be used more than once. Example:
-                          "mlr --from a.dat --from b.dat cat" is the same as
-                          "mlr cat a.dat b.dat".
-       --mfrom {filenames} --  Use this to specify one of more input files before the verb(s),
-                          rather than after. May be used more than once.
-                          The list of filename must end with "--". This is useful
-                          for example since "--from *.csv" doesn't do what you might
-                          hope but "--mfrom *.csv --" does.
-       --load {filename}  Load DSL script file for all put/filter operations on the command line.
-                          If the name following --load is a directory, load all "*.mlr" files
-                          in that directory. This is just like "put -f" and "filter -f"
-                          except it's up-front on the command line, so you can do something like
-                          alias mlr='mlr --load ~/myscripts' if you like.
-       --mload {names} -- Like --load but works with more than one filename,
-                          e.g. '--mload *.mlr --'.
-       -n                 Process no input files, nor standard input either. Useful
-                          for mlr put with begin/end statements only. (Same as --from
-                          /dev/null.) Also useful in "mlr -n put -v '...'" for
-                          analyzing abstract syntax trees (if that's your thing).
-       -I                 Process files in-place. For each file name on the command
-                          line, output is written to a temp file in the same
-                          directory, which is then renamed over the original. Each
-                          file is processed in isolation: if the output format is
-                          CSV, CSV headers will be present in each output file
-                          statistics are only over each file's own records; and so on.
+       * Miller uses ANSI escape sequences only. This does not work on Windows except within Cygwin.
+       * Requires `TERM` environment variable to be set to non-empty string.
+       * Doesn't try to check to see whether the terminal is capable of 256-color
+         ANSI vs 16-color ANSI. Note that if colors are in the range 0..15
+         then 16-color ANSI escapes are used, so this is in the user's control.
+
+       How you can control colorization:
+
+       * Suppression/unsuppression:
+           * Environment variable `export MLR_NO_COLOR=true` means don't color even if stdout+TTY.
+           * Environment variable `export MLR_ALWAYS_COLOR=true` means do color even if not stdout+TTY.
+             For example, you might want to use this when piping mlr output to `less -r`.
+           * Command-line flags `--no-color` or `-M`, `--always-color` or `-C`.
+
+       * Color choices can be specified by using environment variables, or command-line flags,
+         with values 0..255:
+           * `export MLR_KEY_COLOR=208`, `MLR_VALUE_COLOR=33`, etc.:
+               `MLR_KEY_COLOR` `MLR_VALUE_COLOR` `MLR_PASS_COLOR` `MLR_FAIL_COLOR`
+               `MLR_REPL_PS1_COLOR` `MLR_REPL_PS2_COLOR` `MLR_HELP_COLOR`
+           * Command-line flags `--key-color 208`, `--value-color 33`, etc.:
+               `--key-color` `--value-color` `--pass-color` `--fail-color`
+               `--repl-ps1-color` `--repl-ps2-color` `--help-color`
+           * This is particularly useful if your terminal's background color clashes with current settings.
+
+       If environment-variable settings and command-line flags are both provided, the latter take precedence.
+
+       Please do mlr `--list-color-codes` to see the available color codes (like 170), and
+       `mlr --list-color-names` to see available names (like `orchid`).
+
+       --always-color or -C
+       --fail-color
+       --help-color
+       --key-color
+       --list-color-codes
+       --list-color-names
+       --no-color or -M
+       --pass-color
+       --value-color
+
+PPRINT-ONLY FLAGS
+       These are flags which are applicable to PPRINT output format.
+
+       --barred                 Prints a border around PPRINT output (not available
+                                for input).
+       --right                  Right-justifies all fields for PPRINT output.
+
+SEPARATOR FLAGS
+       Separator options:
+
+           --rs     --irs     --ors              Record separators, e.g. 'lf' or '\\r\\n'
+           --fs     --ifs     --ofs  --repifs    Field separators, e.g. comma
+           --ps     --ips     --ops              Pair separators, e.g. equals sign
+
+       TODO: auto-detect is still TBD for Miller 6
+
+       Notes about line endings:
+
+       * Default line endings (`--irs` and `--ors`) are "auto" which means autodetect from
+         the input file format, as long as the input file(s) have lines ending in either
+         LF (also known as linefeed, `\n`, `0x0a`, or Unix-style) or CRLF (also known as
+         carriage-return/linefeed pairs, `\r\n`, `0x0d 0x0a`, or Windows-style).
+       * If both `irs` and `ors` are `auto` (which is the default) then LF input will lead to LF
+         output and CRLF input will lead to CRLF output, regardless of the platform you're
+         running on.
+       * The line-ending autodetector triggers on the first line ending detected in the input
+         stream. E.g. if you specify a CRLF-terminated file on the command line followed by an
+         LF-terminated file then autodetected line endings will be CRLF.
+       * If you use `--ors {something else}` with (default or explicitly specified) `--irs auto`
+         then line endings are autodetected on input and set to what you specify on output.
+       * If you use `--irs {something else}` with (default or explicitly specified) `--ors auto`
+         then the output line endings used are LF on Unix/Linux/BSD/MacOSX, and CRLF on Windows.
+
+       Notes about all other separators:
+
+       * IPS/OPS are only used for DKVP and XTAB formats, since only in these formats
+         do key-value pairs appear juxtaposed.
+       * IRS/ORS are ignored for XTAB format. Nominally IFS and OFS are newlines;
+         XTAB records are separated by two or more consecutive IFS/OFS -- i.e.
+         a blank line. Everything above about `--irs/--ors/--rs auto` becomes `--ifs/--ofs/--fs`
+         auto for XTAB format. (XTAB's default IFS/OFS are "auto".)
+       * OFS must be single-character for PPRINT format. This is because it is used
+         with repetition for alignment; multi-character separators would make
+         alignment impossible.
+       * OPS may be multi-character for XTAB format, in which case alignment is
+         disabled.
+       * TSV is simply CSV using tab as field separator (`--fs tab`).
+       * FS/PS are ignored for markdown format; RS is used.
+       * All FS and PS options are ignored for JSON format, since they are not relevant
+         to the JSON format.
+       * You can specify separators in any of the following ways, shown by example:
+         - Type them out, quoting as necessary for shell escapes, e.g.
+           `--fs '|' --ips :`
+         - C-style escape sequences, e.g. `--rs '\r\n' --fs '\t'`.
+         - To avoid backslashing, you can use any of the following names:
+              TODO desc-to-chars map
+
+       * Default separators by format:
+            TODO default_xses
+
+       --fs {string}            Specify FS for input and output.
+       --ifs {string}           Specify FS for input.
+       --ips {string}           Specify PS for input.
+       --irs {string}           Specify RS for input.
+       --ofs {string}           Specify FS for output.
+       --ops {string}           Specify PS for output.
+       --ors {string}           Specify RS for output.
+       --ps {string}            Specify PS for input and output.
+       --repifs                 Let IFS be repeated: e.g. for splitting on multiple
+                                spaces.
+       --rs {string}            Specify RS for input and output.
 
 AUXILIARY COMMANDS
        Available subcommands:
@@ -386,80 +617,6 @@ AUXILIARY COMMANDS
          repl
        For more information, please invoke mlr {subcommand} --help.
 
-REPL
-       Usage: mlr repl [options] {zero or more data-file names}
-       -v Prints the expressions's AST (abstract syntax tree), which gives
-          full transparency on the precedence and associativity rules of
-          Miller's grammar, to stdout.
-
-       -d Like -v but uses a parenthesized-expression format for the AST.
-
-       -D Like -d but with output all on one line.
-
-       -w Show warnings about uninitialized variables
-
-       -q Don't show startup banner
-
-       -s Don't show prompts
-
-       --load {DSL script file} Load script file before presenting the prompt.
-          If the name following --load is a directory, load all "*.mlr" files
-          in that directory.
-
-       --mload {DSL script files} -- Like --load but works with more than one filename,
-          e.g. '--mload *.mlr --'.
-
-       -h|--help Show this message.
-
-       Or any --icsv, --ojson, etc. reader/writer options as for the main Miller command line.
-
-       Any data-file names are opened just as if you had waited and typed :open {filenames}
-       at the Miller REPL prompt.
-
-OUTPUT COLORIZATION
-       Things having colors:
-       * Keys in CSV header lines, JSON keys, etc
-       * Values in CSV data lines, JSON scalar values, etc
-        in regression-test output
-       * Some online-help strings
-
-       Rules for coloring:
-       * By default, colorize output only if writing to stdout and stdout is a TTY.
-         * Example: color: mlr --csv cat foo.csv
-         * Example: no color: mlr --csv cat foo.csv > bar.csv
-         * Example: no color: mlr --csv cat foo.csv | less
-       * The default colors were chosen since they look OK with white or black terminal background,
-         and are differentiable with common varieties of human color vision.
-
-       Mechanisms for coloring:
-       * Miller uses ANSI escape sequences only. This does not work on Windows except on Cygwin.
-       * Requires TERM environment variable to be set to non-empty string.
-       * Doesn't try to check to see whether the terminal is capable of 256-color
-         ANSI vs 16-color ANSI. Note that if colors are in the range 0..15
-         then 16-color ANSI escapes are used, so this is in the user's control.
-
-       How you can control colorization:
-       * Suppression/unsuppression:
-         * Environment variable export MLR_NO_COLOR=true means don't color even if stdout+TTY.
-         * Environment variable export MLR_ALWAYS_COLOR=true means do color even if not stdout+TTY.
-           For example, you might want to use this when piping mlr output to less -r.
-         * Command-line flags --no-color or -M, --always-color or -C.
-
-       * Color choices can be specified by using environment variables, or command-line flags,
-         with values 0..255:
-         * export MLR_KEY_COLOR=208, MLR_VALUE_COLOR-33, etc.:
-           MLR_KEY_COLOR MLR_VALUE_COLOR MLR_PASS_COLOR MLR_FAIL_COLOR
-           MLR_REPL_PS1_COLOR MLR_REPL_PS2_COLOR MLR_HELP_COLOR
-         * Command-line flags --key-color 208, --value-color 33, etc.:
-           --key-color --value-color --pass-color --fail-color
-           --repl-ps1-color --repl-ps2-color --help-color
-         * This is particularly useful if your terminal's background color clashes with current settings.
-
-       If environment-variable settings and command-line flags are both provided,the latter take precedence.
-
-       Please do mlr --list-color-codes to see the available color codes (like 170), and
-       mlr --list-color-names to see available names (like orchid).
-
 MLRRC
        You can set up personal defaults via a $HOME/.mlrrc and/or ./.mlrrc.
        For example, if you usually process CSV, then you can put "--csv" in your .mlrrc file
@@ -493,6 +650,36 @@ MLRRC
        See also:
        https://miller.readthedocs.io/en/latest/customization.html
 
+REPL
+       Usage: mlr repl [options] {zero or more data-file names}
+       -v Prints the expressions's AST (abstract syntax tree), which gives
+          full transparency on the precedence and associativity rules of
+          Miller's grammar, to stdout.
+
+       -d Like -v but uses a parenthesized-expression format for the AST.
+
+       -D Like -d but with output all on one line.
+
+       -w Show warnings about uninitialized variables
+
+       -q Don't show startup banner
+
+       -s Don't show prompts
+
+       --load {DSL script file} Load script file before presenting the prompt.
+          If the name following --load is a directory, load all "*.mlr" files
+          in that directory.
+
+       --mload {DSL script files} -- Like --load but works with more than one filename,
+          e.g. '--mload *.mlr --'.
+
+       -h|--help Show this message.
+
+       Or any --icsv, --ojson, etc. reader/writer options as for the main Miller command line.
+
+       Any data-file names are opened just as if you had waited and typed :open {filenames}
+       at the Miller REPL prompt.
+
 VERBS
    altkv
        Usage: mlr altkv [options]
@@ -2526,4 +2713,4 @@ SEE ALSO
 
 
 
-                                  2021-09-05                         MILLER(1)
+                                  2021-09-08                         MILLER(1)
diff --git a/man6/mkman.rb b/man6/mkman.rb
index f522a8fd9..e40047afe 100755
--- a/man6/mkman.rb
+++ b/man6/mkman.rb
@@ -43,62 +43,30 @@ a special case.) This manpage documents #{`mlr --version`.chomp}."""
 	print make_section('DATA FORMATS', [])
 	print make_code_block(`mlr help data-formats`)
 
+	print make_section('HELP OPTIONS', [])
+	print make_code_block(`mlr help topics`)
+
 	print make_section('VERB LIST', [])
 	print make_code_block(`mlr help list-verbs-as-paragraph`)
 
 	print make_section('FUNCTION LIST', [])
 	print make_code_block(`mlr help list-functions-as-paragraph`)
 
-	print make_section('HELP OPTIONS', [])
-	print make_code_block(`mlr help topics`)
-
-  print make_section('OPTIONS', [
-"""In the following option flags, the version with \"i\" designates the input
-stream, \"o\" the output stream, and the version without prefix sets the option
-for both input and output stream. For example: --irs sets the input record
-separator, --ors the output record separator, and --rs sets both the input and
-output separator to the given value."""
-  ])
-
-	print make_subsection('DATA-FORMAT OPTIONS', [])
-	print make_code_block(`mlr help data-format-options`)
-
-	print make_subsection('FORMAT-CONVERSION KEYSTROKE-SAVERS', [])
-	print make_code_block(`mlr help format-conversion`)
-
-	print make_subsection('SEPARATORS', [])
-	print make_code_block(`mlr help separator-options`)
-
-	print make_subsection('COMPRESSED I/O', [])
-	print make_code_block(`mlr help compressed-data`)
-
-	print make_subsection('COMMENTS IN DATA', [])
-	print make_code_block(`mlr help comments-in-data`)
-
-	print make_subsection('CSV-SPECIFIC OPTIONS', [])
-	print make_code_block(`mlr help csv-options`)
-
-	print make_subsection('DOUBLE-QUOTING FOR CSV/CSVLITE OUTPUT', [])
-	print make_code_block(`mlr help double-quoting`)
-
-	print make_subsection('NUMBER FORMATTING', [])
-	print make_code_block(`mlr help number-formatting`)
-
-	print make_subsection('OTHER OPTIONS', [])
-	print make_code_block(`mlr help misc`)
+  section_names = `mlr help list-flag-sections`.split("\n")
+  for section_name in section_names
+	  print make_section(section_name.upcase, [""])
+	  print make_code_block(`mlr help show-help-for-section '#{section_name}'`)
+  end
 
 	print make_section('AUXILIARY COMMANDS', [])
 	print make_code_block(`mlr aux-list`)
 
-  print make_section('REPL', [])
-  print make_code_block(`mlr repl -h`)
-
-  print make_section('OUTPUT COLORIZATION', [])
-  print make_code_block(`mlr help output-colorization`)
-
   print make_section('MLRRC', [])
   print make_code_block(`mlr help mlrrc`)
 
+  print make_section('REPL', [])
+  print make_code_block(`mlr repl -h`)
+
   verbs = `mlr help list-verbs`
   print make_section('VERBS', [
     ""
diff --git a/man6/mlr6.1 b/man6/mlr6.1
index a5dee8e50..e059dfaae 100644
--- a/man6/mlr6.1
+++ b/man6/mlr6.1
@@ -2,12 +2,12 @@
 .\"     Title: mlr
 .\"    Author: [see the "AUTHOR" section]
 .\" Generator: ./mkman.rb
-.\"      Date: 2021-09-05
+.\"      Date: 2021-09-08
 .\"    Manual: \ \&
 .\"    Source: \ \&
 .\"  Language: English
 .\"
-.TH "MILLER" "1" "2021-09-05" "\ \&" "\ \&"
+.TH "MILLER" "1" "2021-09-08" "\ \&" "\ \&"
 .\" -----------------------------------------------------------------
 .\" * Portability definitions
 .\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -29,7 +29,7 @@
 miller \- like awk, sed, cut, join, and sort for name-indexed data such as CSV and tabular JSON.
 .SH "SYNOPSIS"
 .sp
-Usage: mlr [I/O options] {verb} [verb-dependent options ...] {zero or more file names}
+Usage: mlr [flags] {verb} [verb-dependent options ...] {zero or more file names}
 Output of one verb may be chained as input to another using "then", e.g.
   mlr stats1 -a min,mean,max -f flag,u,v -g color then sort -f color
 Please see 'mlr help topics' for more information.
@@ -128,6 +128,49 @@ NIDX: implicitly numerically indexed (Unix-toolkit style)
 .fi
 .if n \{\
 .RE
+.SH "HELP OPTIONS"
+.if n \{\
+.RS 0
+.\}
+.nf
+Type 'mlr help {topic}' for any of the following:
+Essentials:
+  mlr help topics
+  mlr help basic-examples
+  mlr help data-formats
+Flags:
+  mlr help flags
+Verbs:
+  mlr help list-verbs
+  mlr help usage-verbs
+  mlr help verb
+Functions:
+  mlr help list-functions
+  mlr help list-function-classes
+  mlr help list-functions-in-class
+  mlr help usage-functions
+  mlr help usage-functions-by-class
+  mlr help function
+Keywords:
+  mlr help list-keywords
+  mlr help usage-keywords
+  mlr help keyword
+Other:
+  mlr help auxents
+  mlr help mlrrc
+  mlr help output-colorization
+  mlr help type-arithmetic-info
+Shorthands:
+  mlr -g = mlr help flags
+  mlr -l = mlr help list-verbs
+  mlr -L = mlr help usage-verbs
+  mlr -f = mlr help list-functions
+  mlr -F = mlr help usage-functions
+  mlr -k = mlr help list-keywords
+  mlr -K = mlr help usage-keywords
+.fi
+.if n \{\
+.RE
 .SH "VERB LIST"
 .if n \{\
 .RS 0
@@ -173,133 +216,217 @@ version ! != !=~ % & && * ** + - . .* .+ .- ./ / // < << <= == =~ > >= >> >>>
 .fi
 .if n \{\
 .RE
-.SH "HELP OPTIONS"
+.SH "COMMENTS-IN-DATA FLAGS"
+.sp
+
 .if n \{\
 .RS 0
 .\}
 .nf
-Type 'mlr help {topic}' for any of the following:
-  mlr help topics
-  mlr help auxents
-  mlr help basic-examples
-  mlr help comments-in-data
-  mlr help compressed-data
-  mlr help csv-options
-  mlr help data-format-options
-  mlr help data-formats
-  mlr help double-quoting
-  mlr help format-conversion
-  mlr help function
-  mlr help keyword
-  mlr help list-functions
-  mlr help list-function-classes
-  mlr help list-functions-in-class
-  mlr help list-functions-as-paragraph
-  mlr help list-functions-as-table
-  mlr help list-keywords
-  mlr help list-keywords-as-paragraph
-  mlr help list-verbs
-  mlr help list-verbs-as-paragraph
-  mlr help misc
-  mlr help mlrrc
-  mlr help number-formatting
-  mlr help output-colorization
-  mlr help separator-options
-  mlr help type-arithmetic-info
-  mlr help usage-functions
-  mlr help usage-functions-by-class
-  mlr help usage-keywords
-  mlr help usage-verbs
-  mlr help verb
-Shorthands:
-  mlr -l = mlr help list-verbs
-  mlr -L = mlr help usage-verbs
-  mlr -f = mlr help list-functions
-  mlr -F = mlr help usage-functions
-  mlr -k = mlr help list-keywords
-  mlr -K = mlr help usage-keywords
+Miller lets you put comments in your data, such as
+
+    # This is a comment for a CSV file
+    a,b,c
+    1,2,3
+    4,5,6
+
+Notes:
+
+* Comments are only honored at the start of a line.
+* In the absence of any of the below four options, comments are data like
+  any other text. (The comments-in-data feature is opt-in.)
+* When `--pass-comments` is used, comment lines are written to standard output
+  immediately upon being read; they are not part of the record stream.  Results
+  may be counterintuitive. A suggestion is to place comments at the start of
+  data files.
+
+--pass-comments          Immediately print commented lines (prefixed by `#`)
+                         within the input.
+--pass-comments-with {string}
+                         Immediately print commented lines within input, with
+                         specified prefix.
+--skip-comments          Ignore commented lines (prefixed by `#`) within the
+                         input.
+--skip-comments-with {string}
+                         Ignore commented lines within input, with specified
+                         prefix.
 .fi
 .if n \{\
 .RE
-.SH "OPTIONS"
+.SH "COMPRESSED-DATA FLAGS"
 .sp
-In the following option flags, the version with "i" designates the input
-stream, "o" the output stream, and the version without prefix sets the option
-for both input and output stream. For example: --irs sets the input record
-separator, --ors the output record separator, and --rs sets both the input and
-output separator to the given value.
-.SS "DATA-FORMAT OPTIONS"
+
 .if n \{\
 .RS 0
 .\}
 .nf
---idkvp   --odkvp   --dkvp      Delimited key-value pairs, e.g "a=1,b=2"
-                                 (Miller's default format).
+Miller offers a few different ways to handle reading data files which have been compressed.
 
---inidx   --onidx   --nidx      Implicitly-integer-indexed fields (Unix-toolkit style).
--T                              Synonymous with "--nidx --fs tab".
+* Decompression done within the Miller process itself: `--bz2in` `--gzin` `--zin`
+* Decompression done outside the Miller process: `--prepipe` `--prepipex`
 
---icsv    --ocsv    --csv       Comma-separated value (or tab-separated with --fs tab, etc.)
+Using `--prepipe` and `--prepipex` you can specify an action to be
+taken on each input file.  The prepipe command must be able to read from
+standard input; it will be invoked with `{command} < {filename}`.  The
+prepipex command must take a filename as argument; it will be invoked with
+`{command} {filename}`.
 
---itsv    --otsv    --tsv       Keystroke-savers for "--icsv --ifs tab",
-                                "--ocsv --ofs tab", "--csv --fs tab".
---iasv    --oasv    --asv       Similar but using ASCII FS 0x1f and RS 0x1e\en",
---iusv    --ousv    --usv       Similar but using Unicode FS U+241F (UTF-8 0xe2909f)\en",
-                                and RS U+241E (UTF-8 0xe2909e)\en",
+Examples:
 
---icsvlite --ocsvlite --csvlite Comma-separated value (or tab-separated with --fs tab, etc.).
-							    The 'lite' CSV does not handle RFC-CSV double-quoting rules; is
-							    slightly faster and handles heterogeneity in the input stream via
-							    empty newline followed by new header line. See also
-								https://johnkerl.org/miller6/file-formats.html#csv-tsv-asv-usv-etc
+    mlr --prepipe gunzip
+    mlr --prepipe zcat -cf
+    mlr --prepipe xz -cd
+    mlr --prepipe cat
 
---itsvlite --otsvlite --tsvlite Keystroke-savers for "--icsvlite --ifs tab",
-                                "--ocsvlite --ofs tab", "--csvlite --fs tab".
--t                              Synonymous with --tsvlite.
---iasvlite --oasvlite --asvlite Similar to --itsvlite et al. but using ASCII FS 0x1f and RS 0x1e\en",
---iusvlite --ousvlite --usvlite Similar to --itsvlite et al. but using Unicode FS U+241F (UTF-8 0xe2909f)\en",
-                                and RS U+241E (UTF-8 0xe2909e)\en",
+Note that this feature is quite general and is not limited to decompression
+utilities. You can use it to apply per-file filters of your choice.  For output
+compression (or other) utilities, simply pipe the output:
+`mlr ... | {your compression command} > outputfilenamegoeshere`
 
---ipprint --opprint --pprint    Pretty-printed tabular (produces no
-                                output until all input is in).
-                    --right     Right-justifies all fields for PPRINT output.
-                    --barred    Prints a border around PPRINT output
-                                (only available for output).
+Lastly, note that if `--prepipe` or `--prepipex` is specified, it replaces any
+decisions that might have been made based on the file suffix. Likewise,
+`--gzin`/`--bz2in`/`--zin` are ignored if `--prepipe` is also specified.
 
-          --omd                 Markdown-tabular (only available for output).
+--bz2in                  Uncompress bzip2 within the Miller process. Done by
+                         default if file ends in `.bz2`.
+--gzin                   Uncompress gzip within the Miller process. Done by
+                         default if file ends in `.gz`.
+--prepipe {decompression command}
+                         You can, of course, already do without this for
+                         single input files, e.g. `gunzip < myfile.csv.gz |
+                         mlr ...`. Allowed at the command line, but not in
+                         `.mlrrc` to avoid unexpected code execution.
+--prepipe-bz2            Same as `--prepipe bz2`, except this is allowed in
+                         `.mlrrc`.
+--prepipe-gunzip         Same as `--prepipe gunzip`, except this is allowed in
+                         `.mlrrc`.
+--prepipe-zcat           Same as `--prepipe zcat`, except this is allowed in
+                         `.mlrrc`.
+--prepipex {decompression command}
+                         Like `--prepipe` with one exception: doesn't insert
+                         `<` between command and filename at runtime. Useful
+                         for some commands like `unzip -qc` which don't read
+                         standard input. Allowed at the command line, but not
+                         in `.mlrrc` to avoid unexpected code execution.
+--zin                    Uncompress zlib within the Miller process. Done by
+                         default if file ends in `.z`.
+.fi
+.if n \{\
+.RE
+.SH "CSV-ONLY FLAGS"
+.sp
 
---ixtab   --oxtab   --xtab      Pretty-printed vertical-tabular.
-                    --xvright   Right-justifies values for XTAB format.
+.if n \{\
+.RS 0
+.\}
+.nf
+--allow-ragged-csv-input or --ragged
+                         If a data line has fewer fields than the header line,
+                         fill remaining keys with empty string. If a data line
+                         has more fields than the header line, use integer
+                         field labels as in the implicit-header case.
+--headerless-csv-output  Print only CSV data lines; do not print CSV header
+                         lines.
+--implicit-csv-header    Use 1,2,3,... as field labels, rather than from line
+                         1 of input files. Tip: combine with `label` to
+                         recreate missing headers.
+--no-implicit-csv-header Opposite of `--implicit-csv-header`. This is the
+                         default anyway -- the main use is for the flags to
+                         `mlr join` if you have main file(s) which are
+                         headerless but you want to join in on a file which
+                         does have a CSV header. Then you could use `mlr --csv
+                         --implicit-csv-header join --no-implicit-csv-header
+                         -l your-join-in-with-header.csv ...
+                         your-headerless.csv`.
+-N                       Keystroke-saver for `--implicit-csv-header
+                         --headerless-csv-output`.
+.fi
+.if n \{\
+.RE
+.SH "FILE-FORMAT FLAGS"
+.sp
 
---ijson   --ojson   --json      JSON tabular: sequence or list of one-level
-                                maps: {...}{...} or [{...},{...}].
-                    --jvstack   Put one key-value pair per line for JSON output.
-                 --no-jvstack   Put objects/arrays all on one line for JSON output.
-              --jsonx --ojsonx  Keystroke-savers for --json --jvstack
-              --jsonx --ojsonx  and --ojson --jvstack, respectively.
-                    --jlistwrap Wrap JSON output in outermost [ ].
-            --flatsep {string} Separator for flattening multi-level JSON keys,
-                                e.g. '{"a":{"b":3}}' becomes a:b => 3 for
-                                non-JSON formats. Defaults to ..\en",
+.if n \{\
+.RS 0
+.\}
+.nf
+TO DO: brief list of formats w/ xref to m6 webdocs.
 
--p is a keystroke-saver for --nidx --fs space --repifs
-
-Examples: --csv for CSV-formatted input and output; --icsv --opprint for
+Examples: `--csv` for CSV-formatted input and output; `--icsv --opprint` for
 CSV-formatted input and pretty-printed output.
 
-Please use --iformat1 --oformat2 rather than --format1 --oformat2.
-The latter sets up input and output flags for format1, not all of which
-are overridden in all cases by setting output format to format2.
+Please use `--iformat1 --oformat2` rather than `--format1 --oformat2`.
+The latter sets up input and output flags for `format1`, not all of which
+are overridden in all cases by setting output format to `format2`.
+
+--asv or --asvlite       Use ASV format for input and output data.
+--csv or -c              Use CSV format for input and output data.
+--csvlite                Use CSV-lite format for input and output data.
+--dkvp                   Use DKVP format for input and output data.
+--iasv or --iasvlite     Use ASV format for input data.
+--icsv                   Use CSV format for input data.
+--icsvlite               Use CSV-lite format for input data.
+--idkvp                  Use DKVP format for input data.
+--ijson                  Use JSON format for input data.
+--inidx                  Use NIDX format for input data.
+--io {format name}       Use format name for input and output data. For
+                         example: `--io csv` is the same as `--csv`.
+--ipprint                Use PPRINT format for input data.
+--itsv                   Use TSV format for input data.
+--itsvlite               Use TSV-lite format for input data.
+--iusv or --iusvlite     Use USV format for input data.
+--ixtab                  Use XTAB format for input data.
+--json or -j             Use JSON format for input and output data.
+--nidx                   Use NIDX format for input and output data.
+--oasv or --oasvlite     Use ASV format for output data.
+--ocsv                   Use CSV format for output data.
+--ocsvlite               Use CSV-lite format for output data.
+--odkvp                  Use DKVP format for output data.
+--ojson                  Use JSON format for output data.
+--omd                    Use markdown-tabular format for output data.
+--onidx                  Use NIDX format for output data.
+--opprint                Use PPRINT format for output data.
+--otsv                   Use TSV format for output data.
+--otsvlite               Use TSV-lite format for output data.
+--ousv or --ousvlite     Use USV format for output data.
+--oxtab                  Use XTAB format for output data.
+--pprint                 Use PPRINT format for input and output data.
+--tsv                    Use TSV format for input and output data.
+--tsvlite or -t          Use TSV-lite format for input and output data.
+--usv or --usvlite       Use USV format for input and output data.
+--xtab                   Use XTAB format for input and output data.
+-i {format name}         Use format name for input data. For example: `-i csv`
+                         is the same as `--icsv`.
+-o {format name}         Use format name for output data. For example: `-o
+                         csv` is the same as `--ocsv`.
 .fi
 .if n \{\
 .RE
-.SS "FORMAT-CONVERSION KEYSTROKE-SAVERS"
+.SH "FLATTEN-UNFLATTEN FLAGS"
+.sp
+
+.if n \{\
+.RS 0
+.\}
+.nf
+--flatsep or --jflatsep or --oflatsep {string}
+                         Separator for flattening multi-level JSON keys, e.g.
+                         `{"a":{"b":3}}` becomes `a:b => 3` for non-JSON
+                         formats. Defaults to `.`.
+--no-auto-flatten        
+--no-auto-unflatten      
+.fi
+.if n \{\
+.RE
+.SH "FORMAT-CONVERSION KEYSTROKE-SAVER FLAGS"
+.sp
+
 .if n \{\
 .RS 0
 .\}
 .nf
 As keystroke-savers for format-conversion you may use the following:
---c2t --c2d --c2n --c2j --c2x --c2p --c2m
+      --c2t --c2d --c2n --c2j --c2x --c2p --c2m
 --t2c       --t2d --t2n --t2j --t2x --t2p --t2m
 --d2c --d2t       --d2n --d2j --d2x --d2p --d2m
 --n2c --n2t --n2d       --n2j --n2x --n2p --n2m
@@ -309,173 +436,319 @@ As keystroke-savers for format-conversion you may use the following:
 The letters c t d n j x p m refer to formats CSV, TSV, DKVP, NIDX, JSON, XTAB,
 PPRINT, and markdown, respectively. Note that markdown format is available for
 output only.
-.fi
-.if n \{\
-.RE
-.SS "SEPARATORS"
-.if n \{\
-.RS 0
-.\}
-.nf
-THIS IS STILL TBD FOR MILLER 6
-.fi
-.if n \{\
-.RE
-.SS "COMPRESSED I/O"
-.if n \{\
-.RS 0
-.\}
-.nf
-Decompression done within the Miller process itself:
---gzin  Uncompress gzip within the Miller process. Done by default if file ends in ".gz".
---bz2in Uncompress bz2ip within the Miller process. Done by default if file ends in ".bz2".
---zin   Uncompress zlib within the Miller process. Done by default if file ends in ".z".
 
-Decompression done outside the Miller process:
---prepipe {command} You can, of course, already do without this for single input files,
-  e.g. "gunzip < myfile.csv.gz | mlr ..."
---prepipex {command} Like --prepipe with one exception: doesn't insert '<' between
-  command and filename at runtime. Useful for some commands like 'unzip -qc'
-  which don't read standard input.
-
-Using --prepipe and --prepipex you can specify an action to be taken on each
-input file. This prepipe command must be able to read from standard input; it
-will be invoked with {command} < {filename}.
-
-Examples:
-  mlr --prepipe gunzip
-  mlr --prepipe zcat -cf
-  mlr --prepipe xz -cd
-  mlr --prepipe cat
-
-Note that this feature is quite general and is not limited to decompression
-utilities. You can use it to apply per-file filters of your choice.  For output
-compression (or other) utilities, simply pipe the output:
-mlr ... | {your compression command} > outputfilenamegoeshere
-
-Lastly, note that if --prepipe or --prepipex is specified, it replaces any
-decisions that might have been made based on the file suffix. Also,
---gzin/--bz2in/--zin are ignored if --prepipe is also specified.
+--c2b                    Use CSV for input, PPRINT with `--barred` for output.
+--c2d                    Use CSV for input, DKVP for output.
+--c2j                    Use CSV for input, JSON for output.
+--c2m                    Use CSV for input, markdown-tabular for output.
+--c2n                    Use CSV for input, NIDX for output.
+--c2p                    Use CSV for input, PPRINT for output.
+--c2t                    Use CSV for input, TSV for output.
+--c2x                    Use CSV for input, XTAB for output.
+--d2b                    Use DKVP for input, PPRINT with `--barred` for
+                         output.
+--d2c                    Use DKVP for input, CSV for output.
+--d2j                    Use DKVP for input, JSON for output.
+--d2m                    Use DKVP for input, markdown-tabular for output.
+--d2n                    Use DKVP for input, NIDX for output.
+--d2p                    Use DKVP for input, PPRINT for output.
+--d2t                    Use DKVP for input, TSV for output.
+--d2x                    Use DKVP for input, XTAB for output.
+--j2b                    Use JSON for input, PPRINT with --barred for output.
+--j2c                    Use JSON for input, CSV for output.
+--j2d                    Use JSON for input, DKVP for output.
+--j2m                    Use JSON for input, markdown-tabular for output.
+--j2n                    Use JSON for input, NIDX for output.
+--j2p                    Use JSON for input, PPRINT for output.
+--j2t                    Use JSON for input, TSV for output.
+--j2x                    Use JSON for input, XTAB for output.
+--n2b                    Use NIDX for input, PPRINT with `--barred` for
+                         output.
+--n2c                    Use NIDX for input, CSV for output.
+--n2d                    Use NIDX for input, DKVP for output.
+--n2j                    Use NIDX for input, JSON for output.
+--n2m                    Use NIDX for input, markdown-tabular for output.
+--n2p                    Use NIDX for input, PPRINT for output.
+--n2t                    Use NIDX for input, TSV for output.
+--n2x                    Use NIDX for input, XTAB for output.
+--p2c                    Use PPRINT for input, CSV for output.
+--p2d                    Use PPRINT for input, DKVP for output.
+--p2j                    Use PPRINT for input, JSON for output.
+--p2m                    Use PPRINT for input, markdown-tabular for output.
+--p2n                    Use PPRINT for input, NIDX for output.
+--p2t                    Use PPRINT for input, TSV for output.
+--p2x                    Use PPRINT for input, XTAB for output.
+--t2b                    Use TSV for input, PPRINT with `--barred` for output.
+--t2c                    Use TSV for input, CSV for output.
+--t2d                    Use TSV for input, DKVP for output.
+--t2j                    Use TSV for input, JSON for output.
+--t2m                    Use TSV for input, markdown-tabular for output.
+--t2n                    Use TSV for input, NIDX for output.
+--t2p                    Use TSV for input, PPRINT for output.
+--t2x                    Use TSV for input, XTAB for output.
+--x2b                    Use XTAB for input, PPRINT with `--barred` for
+                         output.
+--x2c                    Use XTAB for input, CSV for output.
+--x2d                    Use XTAB for input, DKVP for output.
+--x2j                    Use XTAB for input, JSON for output.
+--x2m                    Use XTAB for input, markdown-tabular for output.
+--x2n                    Use XTAB for input, NIDX for output.
+--x2p                    Use XTAB for input, PPRINT for output.
+--x2t                    Use XTAB for input, TSV for output.
+-p                       Keystroke-saver for `--nidx --fs space --repifs`.
+-T                       Keystroke-saver for `--nidx --fs tab`.
 .fi
 .if n \{\
 .RE
-.SS "COMMENTS IN DATA"
+.SH "JSON-ONLY FLAGS"
+.sp
+
 .if n \{\
 .RS 0
 .\}
 .nf
---skip-comments                 Ignore commented lines (prefixed by "#")
-                                within the input.
---skip-comments-with {string}   Ignore commented lines within input, with
-                                specified prefix.
---pass-comments                 Immediately print commented lines (prefixed by "#")
-                                within the input.
---pass-comments-with {string}   Immediately print commented lines within input, with
-                                specified prefix.
+These are flags which are applicable to JSON format.
 
-Notes:
-* Comments are only honored at the start of a line.
-* In the absence of any of the above four options, comments are data like
-  any other text.
-* When pass-comments is used, comment lines are written to standard output
-  immediately upon being read; they are not part of the record stream.  Results
-  may be counterintuitive. A suggestion is to place comments at the start of
-  data files.
+--jlistwrap or --jl      Wrap JSON output in outermost `[ ]`.
+--jvstack                Put one key-value pair per line for JSON output
+                         (multi-line output).
+--no-jvstack             Put objects/arrays all on one line for JSON output.
 .fi
 .if n \{\
 .RE
-.SS "CSV-SPECIFIC OPTIONS"
+.SH "LEGACY FLAGS"
+.sp
+
 .if n \{\
 .RS 0
 .\}
 .nf
---implicit-csv-header Use 1,2,3,... as field labels, rather than from line 1
-                   of input files. Tip: combine with "label" to recreate
-                   missing headers.
---no-implicit-csv-header Do not use --implicit-csv-header. This is the default
-                   anyway -- the main use is for the flags to 'mlr join' if you have
-                   main file(s) which are headerless but you want to join in on
-                   a file which does have a CSV header. Then you could use
-                   'mlr --csv --implicit-csv-header join --no-implicit-csv-header
-                   -l your-join-in-with-header.csv ... your-headerless.csv'
---allow-ragged-csv-input|--ragged If a data line has fewer fields than the header line,
-                   fill remaining keys with empty string. If a data line has more
-                   fields than the header line, use integer field labels as in
-                   the implicit-header case.
---headerless-csv-output   Print only CSV data lines.
--N                 Keystroke-saver for --implicit-csv-header --headerless-csv-output.
+These are flags which don't do anything in the current Miller version.
+They are accepted as no-op flags in order to keep old scripts from breaking.
+
+--jknquoteint            Type information from JSON input files is now
+                         preserved throughout the processing stream.
+--jquoteall              Type information from JSON input files is now
+                         preserved throughout the processing stream.
+--json-fatal-arrays-on-input
+                         Miller now supports arrays as of version 6.
+--json-map-arrays-on-input
+                         Miller now supports arrays as of version 6.
+--json-skip-arrays-on-input
+                         Miller now supports arrays as of version 6.
+--jsonx                  The `--jvstack` flag is now default true in Miller 6.
+--jvquoteall             Type information from JSON input files is now
+                         preserved throughout the processing stream.
+--mmap                   Miller no longer uses memory-mapping to access data
+                         files.
+--no-fflush              The current implementation of Miller does not use
+                         buffered output, so there is no longer anything to
+                         suppress here.
+--no-mmap                Miller no longer uses memory-mapping to access data
+                         files.
+--ojsonx                 The `--jvstack` flag is now default true in Miller 6.
 .fi
 .if n \{\
 .RE
-.SS "DOUBLE-QUOTING FOR CSV/CSVLITE OUTPUT"
+.SH "MISCELLANEOUS FLAGS"
+.sp
+
 .if n \{\
 .RS 0
 .\}
 .nf
-THIS IS STILL WIP FOR MILLER 6
---quote-all        Wrap all fields in double quotes
---quote-none       Do not wrap any fields in double quotes, even if they have
-                   OFS or ORS in them
---quote-minimal    Wrap fields in double quotes only if they have OFS or ORS
-                   in them (default)
---quote-numeric    Wrap fields in double quotes only if they have numbers
-                   in them
---quote-original   Wrap fields in double quotes if and only if they were
-                   quoted on input. This isn't sticky for computed fields:
-                   e.g. if fields a and b were quoted on input and you do
-                   "put '$c = $a . $b'" then field c won't inherit a or b's
-                   was-quoted-on-input flag.
+--from {filename}        Use this to specify an input file before the verb(s),
+                         rather than after. May be used more than once.
+                         Example: `mlr --from a.dat --from b.dat cat` is the
+                         same as `mlr cat a.dat b.dat`.
+--load {filename}        Load DSL script file for all put/filter operations on
+                         the command line. If the name following `--load` is a
+                         directory, load all `*.mlr` files in that directory.
+                         This is just like `put -f` and `filter -f` except
+                         it's up-front on the command line, so you can do
+                         something like `alias mlr='mlr --load ~/myscripts'`
+                         if you like.
+--mfrom {filenames}      Use this to specify one of more input files before
+                         the verb(s), rather than after. May be used more than
+                         once. The list of filename must end with `--`. This
+                         is useful for example since `--from *.csv` doesn't do
+                         what you might hope but `--mfrom *.csv --` does.
+--mload {filenames}      Like `--load` but works with more than one filename,
+                         e.g. `--mload *.mlr --`.
+--ofmt {format}          E.g. %.18f, %.0f, %9.6e. Please use sprintf-style
+                         codes for floating-point nummbers. If not specified,
+                         default formatting is used. See also the `fmtnum`
+                         function and the `format-values` verb.
+--seed {n}               with `n` of the form `12345678` or `0xcafefeed`. For
+                         `put`/`filter` `urand`, `urandint`, and `urand32`.
+-I                       Process files in-place. For each file name on the
+                         command line, output is written to a temp file in the
+                         same directory, which is then renamed over the
+                         original. Each file is processed in isolation: if the
+                         output format is CSV, CSV headers will be present in
+                         each output file, statistics are only over each
+                         file's own records; and so on.
+-n                       Process no input files, nor standard input either.
+                         Useful for `mlr put` with `begin`/`end` statements
+                         only. (Same as `--from /dev/null`.) Also useful in
+                         `mlr -n put -v '...'` for analyzing abstract syntax
+                         trees (if that's your thing).
 .fi
 .if n \{\
 .RE
-.SS "NUMBER FORMATTING"
+.SH "OUTPUT-COLORIZATION FLAGS"
+.sp
+
 .if n \{\
 .RS 0
 .\}
 .nf
-THIS IS STILL WIP FOR MILLER 6
-  --ofmt {format}    E.g. %.18f, %.0f, %9.6e. Please use sprintf-style codes for
-                     floating-point nummbers. If not specified, default formatting is used.
-                     See also the fmtnum function within mlr put (mlr --help-all-functions);
-                     see also the format-values function.
+Miller uses colors to highlight outputs. You can specify color preferences.
+Note: output colorization does not work on Windows.
+
+Things having colors:
+
+* Keys in CSV header lines, JSON keys, etc
+* Values in CSV data lines, JSON scalar values, etc in regression-test output
+* Some online-help strings
+
+Rules for coloring:
+
+* By default, colorize output only if writing to stdout and stdout is a TTY.
+    * Example: color: `mlr --csv cat foo.csv`
+    * Example: no color: `mlr --csv cat foo.csv > bar.csv`
+    * Example: no color: `mlr --csv cat foo.csv | less`
+* The default colors were chosen since they look OK with white or black terminal background,
+  and are differentiable with common varieties of human color vision.
+
+Mechanisms for coloring:
+
+* Miller uses ANSI escape sequences only. This does not work on Windows except within Cygwin.
+* Requires `TERM` environment variable to be set to non-empty string.
+* Doesn't try to check to see whether the terminal is capable of 256-color
+  ANSI vs 16-color ANSI. Note that if colors are in the range 0..15
+  then 16-color ANSI escapes are used, so this is in the user's control.
+
+How you can control colorization:
+
+* Suppression/unsuppression:
+    * Environment variable `export MLR_NO_COLOR=true` means don't color even if stdout+TTY.
+    * Environment variable `export MLR_ALWAYS_COLOR=true` means do color even if not stdout+TTY.
+      For example, you might want to use this when piping mlr output to `less -r`.
+    * Command-line flags `--no-color` or `-M`, `--always-color` or `-C`.
+
+* Color choices can be specified by using environment variables, or command-line flags,
+  with values 0..255:
+    * `export MLR_KEY_COLOR=208`, `MLR_VALUE_COLOR=33`, etc.:
+        `MLR_KEY_COLOR` `MLR_VALUE_COLOR` `MLR_PASS_COLOR` `MLR_FAIL_COLOR`
+        `MLR_REPL_PS1_COLOR` `MLR_REPL_PS2_COLOR` `MLR_HELP_COLOR`
+    * Command-line flags `--key-color 208`, `--value-color 33`, etc.:
+        `--key-color` `--value-color` `--pass-color` `--fail-color`
+        `--repl-ps1-color` `--repl-ps2-color` `--help-color`
+    * This is particularly useful if your terminal's background color clashes with current settings.
+
+If environment-variable settings and command-line flags are both provided, the latter take precedence.
+
+Please do mlr `--list-color-codes` to see the available color codes (like 170), and
+`mlr --list-color-names` to see available names (like `orchid`).
+
+--always-color or -C     
+--fail-color             
+--help-color             
+--key-color              
+--list-color-codes       
+--list-color-names       
+--no-color or -M         
+--pass-color             
+--value-color            
 .fi
 .if n \{\
 .RE
-.SS "OTHER OPTIONS"
+.SH "PPRINT-ONLY FLAGS"
+.sp
+
 .if n \{\
 .RS 0
 .\}
 .nf
---seed {n} with n of the form 12345678 or 0xcafefeed. For put/filter
-                   urand()/urandint()/urand32().
---nr-progress-mod {m}, with m a positive integer: print filename and record
-                   count to os.Stderr every m input records.
---from {filename}  Use this to specify an input file before the verb(s),
-                   rather than after. May be used more than once. Example:
-                   "mlr --from a.dat --from b.dat cat" is the same as
-                   "mlr cat a.dat b.dat".
---mfrom {filenames} --  Use this to specify one of more input files before the verb(s),
-                   rather than after. May be used more than once.
-                   The list of filename must end with "--". This is useful
-                   for example since "--from *.csv" doesn't do what you might
-                   hope but "--mfrom *.csv --" does.
---load {filename}  Load DSL script file for all put/filter operations on the command line.
-                   If the name following --load is a directory, load all "*.mlr" files
-                   in that directory. This is just like "put -f" and "filter -f"
-                   except it's up-front on the command line, so you can do something like
-                   alias mlr='mlr --load ~/myscripts' if you like.
---mload {names} -- Like --load but works with more than one filename,
-                   e.g. '--mload *.mlr --'.
--n                 Process no input files, nor standard input either. Useful
-                   for mlr put with begin/end statements only. (Same as --from
-                   /dev/null.) Also useful in "mlr -n put -v '...'" for
-                   analyzing abstract syntax trees (if that's your thing).
--I                 Process files in-place. For each file name on the command
-                   line, output is written to a temp file in the same
-                   directory, which is then renamed over the original. Each
-                   file is processed in isolation: if the output format is
-                   CSV, CSV headers will be present in each output file
-                   statistics are only over each file's own records; and so on.
+These are flags which are applicable to PPRINT output format.
+
+--barred                 Prints a border around PPRINT output (not available
+                         for input).
+--right                  Right-justifies all fields for PPRINT output.
+.fi
+.if n \{\
+.RE
+.SH "SEPARATOR FLAGS"
+.sp
+
+.if n \{\
+.RS 0
+.\}
+.nf
+Separator options:
+
+    --rs     --irs     --ors              Record separators, e.g. 'lf' or '\e\er\e\en'
+    --fs     --ifs     --ofs  --repifs    Field separators, e.g. comma
+    --ps     --ips     --ops              Pair separators, e.g. equals sign
+
+TODO: auto-detect is still TBD for Miller 6
+
+Notes about line endings:
+
+* Default line endings (`--irs` and `--ors`) are "auto" which means autodetect from
+  the input file format, as long as the input file(s) have lines ending in either
+  LF (also known as linefeed, `\en`, `0x0a`, or Unix-style) or CRLF (also known as
+  carriage-return/linefeed pairs, `\er\en`, `0x0d 0x0a`, or Windows-style).
+* If both `irs` and `ors` are `auto` (which is the default) then LF input will lead to LF
+  output and CRLF input will lead to CRLF output, regardless of the platform you're
+  running on.
+* The line-ending autodetector triggers on the first line ending detected in the input
+  stream. E.g. if you specify a CRLF-terminated file on the command line followed by an
+  LF-terminated file then autodetected line endings will be CRLF.
+* If you use `--ors {something else}` with (default or explicitly specified) `--irs auto`
+  then line endings are autodetected on input and set to what you specify on output.
+* If you use `--irs {something else}` with (default or explicitly specified) `--ors auto`
+  then the output line endings used are LF on Unix/Linux/BSD/MacOSX, and CRLF on Windows.
+
+Notes about all other separators:
+
+* IPS/OPS are only used for DKVP and XTAB formats, since only in these formats
+  do key-value pairs appear juxtaposed.
+* IRS/ORS are ignored for XTAB format. Nominally IFS and OFS are newlines;
+  XTAB records are separated by two or more consecutive IFS/OFS -- i.e.
+  a blank line. Everything above about `--irs/--ors/--rs auto` becomes `--ifs/--ofs/--fs`
+  auto for XTAB format. (XTAB's default IFS/OFS are "auto".)
+* OFS must be single-character for PPRINT format. This is because it is used
+  with repetition for alignment; multi-character separators would make
+  alignment impossible.
+* OPS may be multi-character for XTAB format, in which case alignment is
+  disabled.
+* TSV is simply CSV using tab as field separator (`--fs tab`).
+* FS/PS are ignored for markdown format; RS is used.
+* All FS and PS options are ignored for JSON format, since they are not relevant
+  to the JSON format.
+* You can specify separators in any of the following ways, shown by example:
+  - Type them out, quoting as necessary for shell escapes, e.g.
+    `--fs '|' --ips :`
+  - C-style escape sequences, e.g. `--rs '\er\en' --fs '\et'`.
+  - To avoid backslashing, you can use any of the following names:
+	  TODO desc-to-chars map
+
+* Default separators by format:
+	TODO default_xses
+
+--fs {string}            Specify FS for input and output.
+--ifs {string}           Specify FS for input.
+--ips {string}           Specify PS for input.
+--irs {string}           Specify RS for input.
+--ofs {string}           Specify FS for output.
+--ops {string}           Specify PS for output.
+--ors {string}           Specify RS for output.
+--ps {string}            Specify PS for input and output.
+--repifs                 Let IFS be repeated: e.g. for splitting on multiple
+                         spaces.
+--rs {string}            Specify RS for input and output.
 .fi
 .if n \{\
 .RE
@@ -497,92 +770,6 @@ For more information, please invoke mlr {subcommand} --help.
 .fi
 .if n \{\
 .RE
-.SH "REPL"
-.if n \{\
-.RS 0
-.\}
-.nf
-Usage: mlr repl [options] {zero or more data-file names}
--v Prints the expressions's AST (abstract syntax tree), which gives
-   full transparency on the precedence and associativity rules of
-   Miller's grammar, to stdout.
-
--d Like -v but uses a parenthesized-expression format for the AST.
-
--D Like -d but with output all on one line.
-
--w Show warnings about uninitialized variables
-
--q Don't show startup banner
-
--s Don't show prompts
-
---load {DSL script file} Load script file before presenting the prompt.
-   If the name following --load is a directory, load all "*.mlr" files
-   in that directory.
-
---mload {DSL script files} -- Like --load but works with more than one filename,
-   e.g. '--mload *.mlr --'.
-
--h|--help Show this message.
-
-Or any --icsv, --ojson, etc. reader/writer options as for the main Miller command line.
-
-Any data-file names are opened just as if you had waited and typed :open {filenames}
-at the Miller REPL prompt.
-.fi
-.if n \{\
-.RE
-.SH "OUTPUT COLORIZATION"
-.if n \{\
-.RS 0
-.\}
-.nf
-Things having colors:
-* Keys in CSV header lines, JSON keys, etc
-* Values in CSV data lines, JSON scalar values, etc
- in regression-test output
-* Some online-help strings
-
-Rules for coloring:
-* By default, colorize output only if writing to stdout and stdout is a TTY.
-  * Example: color: mlr --csv cat foo.csv
-  * Example: no color: mlr --csv cat foo.csv > bar.csv
-  * Example: no color: mlr --csv cat foo.csv | less
-* The default colors were chosen since they look OK with white or black terminal background,
-  and are differentiable with common varieties of human color vision.
-
-Mechanisms for coloring:
-* Miller uses ANSI escape sequences only. This does not work on Windows except on Cygwin.
-* Requires TERM environment variable to be set to non-empty string.
-* Doesn't try to check to see whether the terminal is capable of 256-color
-  ANSI vs 16-color ANSI. Note that if colors are in the range 0..15
-  then 16-color ANSI escapes are used, so this is in the user's control.
-
-How you can control colorization:
-* Suppression/unsuppression:
-  * Environment variable export MLR_NO_COLOR=true means don't color even if stdout+TTY.
-  * Environment variable export MLR_ALWAYS_COLOR=true means do color even if not stdout+TTY.
-    For example, you might want to use this when piping mlr output to less -r.
-  * Command-line flags --no-color or -M, --always-color or -C.
-
-* Color choices can be specified by using environment variables, or command-line flags,
-  with values 0..255:
-  * export MLR_KEY_COLOR=208, MLR_VALUE_COLOR-33, etc.:
-    MLR_KEY_COLOR MLR_VALUE_COLOR MLR_PASS_COLOR MLR_FAIL_COLOR
-    MLR_REPL_PS1_COLOR MLR_REPL_PS2_COLOR MLR_HELP_COLOR
-  * Command-line flags --key-color 208, --value-color 33, etc.:
-    --key-color --value-color --pass-color --fail-color
-    --repl-ps1-color --repl-ps2-color --help-color
-  * This is particularly useful if your terminal's background color clashes with current settings.
-
-If environment-variable settings and command-line flags are both provided,the latter take precedence.
-
-Please do mlr --list-color-codes to see the available color codes (like 170), and
-mlr --list-color-names to see available names (like orchid).
-.fi
-.if n \{\
-.RE
 .SH "MLRRC"
 .if n \{\
 .RS 0
@@ -622,6 +809,42 @@ https://miller.readthedocs.io/en/latest/customization.html
 .fi
 .if n \{\
 .RE
+.SH "REPL"
+.if n \{\
+.RS 0
+.\}
+.nf
+Usage: mlr repl [options] {zero or more data-file names}
+-v Prints the expressions's AST (abstract syntax tree), which gives
+   full transparency on the precedence and associativity rules of
+   Miller's grammar, to stdout.
+
+-d Like -v but uses a parenthesized-expression format for the AST.
+
+-D Like -d but with output all on one line.
+
+-w Show warnings about uninitialized variables
+
+-q Don't show startup banner
+
+-s Don't show prompts
+
+--load {DSL script file} Load script file before presenting the prompt.
+   If the name following --load is a directory, load all "*.mlr" files
+   in that directory.
+
+--mload {DSL script files} -- Like --load but works with more than one filename,
+   e.g. '--mload *.mlr --'.
+
+-h|--help Show this message.
+
+Or any --icsv, --ojson, etc. reader/writer options as for the main Miller command line.
+
+Any data-file names are opened just as if you had waited and typed :open {filenames}
+at the Miller REPL prompt.
+.fi
+.if n \{\
+.RE
 .SH "VERBS"
 .sp