From 1e2770f31a622a24c0d49363a1c1646040da1ca2 Mon Sep 17 00:00:00 2001 From: John Kerl Date: Sat, 25 Dec 2021 12:20:19 -0500 Subject: [PATCH] Update perf-docs to include XTAB I/O improvement on #799 --- docs/src/file-formats.md | 2 +- docs/src/new-in-miller-6.md | 5 ++--- docs/src/new-in-miller-6.md.in | 5 ++--- docs/src/reference-main-number-formatting.md | 2 +- todo.txt | 2 -- 5 files changed, 6 insertions(+), 10 deletions(-) diff --git a/docs/src/file-formats.md b/docs/src/file-formats.md index a962c2ef3..df48e3ffc 100644 --- a/docs/src/file-formats.md +++ b/docs/src/file-formats.md @@ -470,7 +470,7 @@ See the [seprators page](reference-main-separators.md) regarding how to specify ## NIDX: Index-numbered (toolkit style) -With `--inidx --ifs ' ' --repifs`, Miller splits lines on whitespace and assigns integer field names starting with 1. +With `--inidx --ifs ' ' --repifs`, Miller splits lines on spaces and assigns integer field names starting with 1. This recapitulates Unix-toolkit behavior. diff --git a/docs/src/new-in-miller-6.md b/docs/src/new-in-miller-6.md index 49c0407ba..22e3a9974 100644 --- a/docs/src/new-in-miller-6.md +++ b/docs/src/new-in-miller-6.md @@ -280,7 +280,7 @@ For the [first benchmark](https://github.com/johnkerl/miller/blob/main/scripts/t | CSV-lite | 1.671 | 1.428 | 1.17x | | DKVP | 2.485 | 2.040 | 1.22x | | NIDX | 1.638 | 1.468 | 1.12x | -| XTAB | 5.147 | 6.252 | 0.82x | +| XTAB | 5.147 | 2.184 | 2.35x | | JSON | 12.457 | 12.416 | 1.00x | For the [second benchmark](https://github.com/johnkerl/miller/blob/main/scripts/chain-cmps.sh), the operations are varied: @@ -308,7 +308,6 @@ For the [third benchmark](https://github.com/johnkerl/miller/blob/main/scripts/c Notes: * CSV processing is particularly improved in Miller 6. -* Record I/O is improved for all but XTAB. -* JSON continues to be a CPU-intensive format. +* Record I/O is improved across the board, except that JSON continues to be a CPU-intensive format. * Miller 6's `sort` merits more performance analysis. * Longer then-chains benefit from Miller 6's [multicore approach](cpu.md). diff --git a/docs/src/new-in-miller-6.md.in b/docs/src/new-in-miller-6.md.in index af19ed47d..000763f52 100644 --- a/docs/src/new-in-miller-6.md.in +++ b/docs/src/new-in-miller-6.md.in @@ -238,7 +238,7 @@ For the [first benchmark](https://github.com/johnkerl/miller/blob/main/scripts/t | CSV-lite | 1.671 | 1.428 | 1.17x | | DKVP | 2.485 | 2.040 | 1.22x | | NIDX | 1.638 | 1.468 | 1.12x | -| XTAB | 5.147 | 6.252 | 0.82x | +| XTAB | 5.147 | 2.184 | 2.35x | | JSON | 12.457 | 12.416 | 1.00x | For the [second benchmark](https://github.com/johnkerl/miller/blob/main/scripts/chain-cmps.sh), the operations are varied: @@ -266,7 +266,6 @@ For the [third benchmark](https://github.com/johnkerl/miller/blob/main/scripts/c Notes: * CSV processing is particularly improved in Miller 6. -* Record I/O is improved for all but XTAB. -* JSON continues to be a CPU-intensive format. +* Record I/O is improved across the board, except that JSON continues to be a CPU-intensive format. * Miller 6's `sort` merits more performance analysis. * Longer then-chains benefit from Miller 6's [multicore approach](cpu.md). diff --git a/docs/src/reference-main-number-formatting.md b/docs/src/reference-main-number-formatting.md index 12a975605..da86fdb21 100644 --- a/docs/src/reference-main-number-formatting.md +++ b/docs/src/reference-main-number-formatting.md @@ -24,7 +24,7 @@ The command-line option `--ofmt {format string}` is the global number format for --ofmt %.9e --ofmt %.6f --ofmt %.0f -These are just familiar `printf` formats. (TODO: write about type-checking once that's implemented.) Additionally, if you use leading width (e.g. `%18.12f`) then the output will contain embedded whitespace, which may not be what you want if you pipe the output to something else, particularly CSV. I use Miller's pretty-print format (`mlr --opprint`) to column-align numerical data. +These are just familiar `printf` formats. Additionally, if you use leading width (e.g. `%18.12f`) then the output will contain embedded whitespace, which may not be what you want if you pipe the output to something else, particularly CSV. I use Miller's pretty-print format (`mlr --opprint`) to column-align numerical data. 
 echo 'x=3.1,y=4.3' | mlr --ofmt '%8.3f' cat
diff --git a/todo.txt b/todo.txt
index ac7dc44db..7ee0c5277 100644
--- a/todo.txt
+++ b/todo.txt
@@ -10,8 +10,6 @@ PUNCHDOWN LIST
   - 0b1011 olh/webdoc
   ? array/map fields: marshal as JSON_SINGLE_LINE
 
-* red-PR cmd/mprofs
-
 * shebang
   - olh -s & --norc; --version & --bare-version; --
   - maybe multiple -s -- ? []args would continue growing during parse ...