doc updates

c/mapping/rval_func_evaluators.c

@@ -69,7 +69,7 @@ function_lookup_t FUNCTION_LOOKUP_TABLE[] = {
 	{FUNC_CLASS_CONVERSION, "hexfmt",    1, "Convert int to string, e.g. 255 to \"0xff\"."},
 	{FUNC_CLASS_CONVERSION, "int",       1, "Convert int/float/bool/string to int."},
 	{FUNC_CLASS_CONVERSION, "string",    1, "Convert int/float/bool/string to string."},
-	{FUNC_CLASS_CONVERSION, "typeof",    1, "Convert argument to type of argument (e.g. string). For debug."},
+	{FUNC_CLASS_CONVERSION, "typeof",    1, "Convert argument to type of argument (e.g. MT_STRING). For debug."},
 
 	{FUNC_CLASS_STRING, ".",        2, "String concatenation."},
 	{FUNC_CLASS_STRING, "gsub",     3, "Example: '$name=gsub($name, \"old\", \"new\")'\n(replace all)."},

c/todo.txt

@@ -8,15 +8,14 @@ TOP-OF-LIST SUMMARY
 
 * rh/fedora/centos mlr-3.4.0
 
-* tri-split null. then RT @sum[$group] = @sum[$group] + $value etc. be sure $nosuch sums to MT_ABSENT.
+* tri-split null. then RT @sum[$group] += $value etc. be sure $nosuch sums to MT_ABSENT.
 ? @x = $0 ?  $0 = @x ?
 * parameterized-emit () vs. no-() unification
 * parameterized-emit neatens
 * += et al. UTs
+? unset/clear? for srec as well as oosvar. either rhs="" or remove key. latter since ass't does former. cf. cut -x.
 * allow begin/end srec assignments in the grammar, to allow better error messages than 'syntax error' in the caller
 * double-check oosvar+emit+tri-split-null by re-implementing as many mappers as possible using ossvars + emit
-? unset/clear ... ?
-? rethink filter ... ?
 
 * 64-bit lengths for containers. test with 5-billion-integer-seq data.
 
@@ -52,7 +51,8 @@ PARAMETERIZED EMIT
 * handle too-short, too-long, just-right cases
 * rename functions
 * rename variables as needed for clarity
-* needs significant comment-by-example
+* rename 'temp' functions
+* needs significant comment-by-example in the code
 * emit w/ and w/o parentheses in the grammar -- try to get just one
 * mld section, w/ olh brief example & xref to it.
 
@@ -62,13 +62,15 @@ TRI-SPLIT NULL, AND += ET AL.
 * to mld:ref
 
 ! UTs
-  - 
+
+----------------------------------------------------------------
   mlr put '@sum += $x; filter false; end{emit @sum}' ../data/small
   mlr put '@sum += $x; filter false; end{emit(@sum)}' ../data/small
   mlr put '@sum[$a] += $x; filter false; end{emit(@sum, "a")}' ../data/small
 
   mlr put '$nonesuch = @nonesuch' ../data/small
 
+----------------------------------------------------------------
   echo x=1,y=2 | mlr put '$z = $x + $y'  # $y is MT_INT
   echo x=1,y=  | mlr put '$z = $x + $y'  # $y is MT_EMPTY:  key present but value empty
   echo x=1,y=  | mlr put '$z = $x + $u'  # $y is MT_ABSENT : key not present
@@ -85,13 +87,22 @@ x=1,y=2
 x=1,y=
 x=1,y=
 
-  echo x=1,y=2 | mlr put '$z = $x + $y; $x=typeof($x)'  # $y is MT_INT
-  echo x=1,y=  | mlr put '$z = $x + $y'  # $y is MT_EMPTY:  key present but value empty
-  echo x=1,y=  | mlr put '$z = $x + $u'  # $y is MT_ABSENT : key not present
+----------------------------------------------------------------
+  echo x=1,y=2 | mlr put '$z = $x + $y; $x=typeof($x);$y=typeof($y);$z=typeof($z)'
+  echo x=1,y=  | mlr put '$z = $x + $y; $x=typeof($x);$y=typeof($y);$z=typeof($z)'
+  echo x=1,y=  | mlr put '$z = $x + $u; $x=typeof($x);$y=typeof($y);$z=typeof($z)'
 
-  echo x=1,y=2 | mlr put '@s = @s + $y'  # @s is MT_UNINIT on first record, then MT_INT on second
-  echo x=1,y=  | mlr put '@s = @s + $y'  # MT_
-  echo x=1,y=  | mlr put '@s = @s + $u'  # MT_
+  echo x=1,y=2 | mlr put '@s = @s + $y; $x=typeof($x);$y=typeof($y);$z=typeof($z)'
+  echo x=1,y=  | mlr put '@s = @s + $y; $x=typeof($x);$y=typeof($y);$z=typeof($z)'
+  echo x=1,y=  | mlr put '@s = @s + $u; $x=typeof($x);$y=typeof($y);$z=typeof($z)'
+
+x=MT_INT,y=MT_INT,z=MT_INT
+x=MT_INT,y=MT_VOID,z=MT_VOID
+x=MT_INT,y=MT_VOID,z=MT_VOID
+
+x=MT_INT,y=MT_INT,z=MT_ABSENT
+x=MT_INT,y=MT_VOID,z=MT_ABSENT
+x=MT_INT,y=MT_VOID,z=MT_ABSENT
 
 ================================================================
 64-BIT CONTAINER SIZES
@@ -132,12 +143,6 @@ OOSVARS:
     > top?
     - uniq?
 
-  o DSL more:
-    - people will ask for an unset/clear syntax
-    - people will ask for +=, -=, etc.
-    - if / else ... not yet (if ever) but think carefully about how not to break the (possible) future while
-      implementing the present
-
   o perf notes:
 
     $ time mlr stats1 -a mean -f x ../data/big.dkvp

doc/mlr.1

@@ -2,12 +2,12 @@
 .\"     Title: mlr
 .\"    Author: [see the "AUTHOR" section]
 .\" Generator: ./mkman.rb
-.\"      Date: 2016-02-26
+.\"      Date: 2016-02-27
 .\"    Manual: \ \&
 .\"    Source: \ \&
 .\"  Language: English
 .\"
-.TH "MILLER" "1" "2016-02-26" "\ \&" "\ \&"
+.TH "MILLER" "1" "2016-02-27" "\ \&" "\ \&"
 .\" -----------------------------------------------------------------
 .\" * Portability definitions
 .\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -156,12 +156,12 @@ output separator to the given value.
 .\}
 .nf
  + + - - * / // % ** | ^ & ~ << >> == != =~ !=~ > >= < <= && || ^^ ! ? : isnull
- isnotnull boolean float fmtnum hexfmt int string . gsub strlen sub tolower
- toupper abs acos acosh asin asinh atan atan2 atanh cbrt ceil cos cosh erf erfc
- exp expm1 floor invqnorm log log10 log1p logifit madd max mexp min mmul msub
- pow qnorm round roundm sgn sin sinh sqrt tan tanh urand urand32 urandint
- dhms2fsec dhms2sec fsec2dhms fsec2hms gmt2sec hms2fsec hms2sec sec2dhms
- sec2gmt sec2hms strftime strptime systime
+ isnotnull boolean float fmtnum hexfmt int string typeof . gsub strlen sub
+ tolower toupper abs acos acosh asin asinh atan atan2 atanh cbrt ceil cos cosh
+ erf erfc exp expm1 floor invqnorm log log10 log1p logifit madd max mexp min
+ mmul msub pow qnorm round roundm sgn sin sinh sqrt tan tanh urand urand32
+ urandint dhms2fsec dhms2sec fsec2dhms fsec2hms gmt2sec hms2fsec hms2sec
+ sec2dhms sec2gmt sec2hms strftime strptime systime
 Please use "mlr --help-function {function name}" for function-specific help.
 Please use "mlr --help-all-functions" or "mlr -f" for help on all functions.
 .fi
@@ -1459,6 +1459,15 @@ printf-style format string, e.g. "%06lld".
 .fi
 .if n \{\
 .RE
+.SS "typeof"
+.if n \{\
+.RS 0
+.\}
+.nf
+(class=conversion #args=1): Convert argument to type of argument (e.g. MT_STRING). For debug.
+.fi
+.if n \{\
+.RE
 .SS "."
 .if n \{\
 .RS 0

doc/reference.html

@@ -372,8 +372,8 @@ Verbs:
 
 Functions for the filter and put verbs:
    + + - - * / // % ** | ^ & ~ << >> == != =~ !=~ > >= < <= && || ^^ ! ? :
-   isnull isnotnull boolean float fmtnum hexfmt int string . gsub strlen sub
-   tolower toupper abs acos acosh asin asinh atan atan2 atanh cbrt ceil cos
+   isnull isnotnull boolean float fmtnum hexfmt int string typeof . gsub strlen
+   sub tolower toupper abs acos acosh asin asinh atan atan2 atanh cbrt ceil cos
    cosh erf erfc exp expm1 floor invqnorm log log10 log1p logifit madd max mexp
    min mmul msub pow qnorm round roundm sgn sin sinh sqrt tan tanh urand
    urand32 urandint dhms2fsec dhms2sec fsec2dhms fsec2hms gmt2sec hms2fsec
@@ -3714,6 +3714,8 @@ int (class=conversion #args=1): Convert int/float/bool/string to int.
 
 string (class=conversion #args=1): Convert int/float/bool/string to string.
 
+typeof (class=conversion #args=1): Convert argument to type of argument (e.g. MT_STRING). For debug.
+
 . (class=string #args=2): String concatenation.
 
 gsub (class=string #args=3): Example: '$name=gsub($name, "old", "new")'