Commit graph

35 commits

Author SHA1 Message Date
Peter Stephenson 099e717c15 35359: Improved math parsing and errors.
Check for bogus trailing ")" at end of top-level parse.

Extend some math error messages to indicate they are math errors.
2015-06-02 10:21:55 +01:00
Peter Stephenson 0da0a0b9c7 35153: nested math substitution 2015-05-15 10:19:53 +01:00
Peter Stephenson 2ef4b38461 34892 (slightly tweaked): math evaluation fix
An empty expression resulting from substitution includes a
Nularg, which needs handling the same as an empty string.
2015-04-15 10:20:06 +01:00
Peter Stephenson 6fa63a0ce2 34606: fix up nested arithmetic substitution
Arithmetic within a parameter substitution is a special
case that needs fixing with the introduction of the new
Inparmath token.

Add test.
2015-02-22 21:32:08 +00:00
Peter Stephenson 52e938bac9 34570: Another nasty command / math substituion thing.
Mark arithmetic substitutions with tokens to make sure the substitution
go knows what to do.  Before it was guessing by counting the
parentheses at the end.
2015-02-19 10:22:40 +00:00
Peter Stephenson 126fb61c7c 34560: Fix $(( that's actually a multiline cmd subst. 2015-02-16 17:16:57 +00:00
Peter Stephenson 3a99ef322d 34280: more widespread use of FORCE_FLOAT.
Add the case of variables read for use in arithmetic expressions.
2015-01-15 13:52:40 +00:00
Peter Stephenson 5f4325a0a4 Propagate float/integer type in arithmetic assignment.
Add test.

Mention this and also floating point mod change in README.
2015-01-12 16:38:00 +00:00
Barton E. Schaefer 2753d49003 34247: regression test for 34230. 2015-01-11 18:50:17 -08:00
Peter Stephenson 62e4db10fe unposted: test binary input using 0b 2014-11-26 20:58:36 +00:00
Peter Stephenson fb707c0acb 33332: test safer numeric import from environment 2014-10-02 17:17:47 +01:00
Peter Stephenson 22b8fd6da9 32299: add use of underscores on arithmetic output for spacing 2014-01-23 10:32:59 +00:00
Peter Stephenson 4c611f9323 31982: detection of floating point constants was problematic.
Remove the cause and fix the original problem with floating point
numbers with leading zeros (users/17445) a different way.
2013-11-14 11:49:22 +00:00
Peter Stephenson f77a7a5b18 31937: zle -Fw uses widget semantics for file descriptor handler 2013-11-07 15:19:07 +00:00
Peter Stephenson 3def943d04 users/17665: add FORCE_FLOAT option 2013-03-05 20:04:53 +00:00
Bart Schaefer 431c22024d users/17445: fix handling of leading zeroes in floating point 2012-12-08 19:50:20 +00:00
Peter Stephenson e550c98d69 30647, 30649: allow underscores in numeric constants 2012-09-11 16:02:41 +00:00
Peter Stephenson f02778f83c 27611: cache parameter values in math eval so subscripts are eval'd once 2010-01-20 17:17:45 +00:00
Peter Stephenson 3c248a0de4 25905: fix and test doubled-hash radix output 2008-10-16 09:34:51 +00:00
Peter Stephenson 8a4af211c3 25149: add C_PRECEDENCES option to make arithmetic have C/Perl precedence 2008-06-12 13:45:04 +00:00
Peter Stephenson fe3884ba08 unposted addition to 25144 for tests 2008-06-10 09:13:01 +00:00
Peter Stephenson 6176062298 24271: handle parse failures in command substitution better 2007-12-16 22:20:31 +00:00
Peter Stephenson 5d6e1ce9c4 not 23562 but should have been: test correction 2007-06-18 13:39:56 +00:00
Peter Stephenson abae4fe16e 23562: add KSH_ZERO_SUBSCRIPT option and leave off by default 2007-06-18 13:25:03 +00:00
Peter Stephenson 2c92d005d7 23165: fix problems with bases: error if over 36 and don't interpret octal 2007-02-12 16:43:40 +00:00
Peter Stephenson b262d310b2 22686: unset array if assigning numeric parameter 2006-09-11 11:09:15 +00:00
Peter Stephenson 1bba34ced5 20599: test for $((##)) should now check for error 2004-11-30 14:14:51 +00:00
Peter Stephenson d3b14e887e 19941: fix (( 3, )) 2004-05-13 20:04:10 +00:00
Oliver Kiddle 76bd48d674 19308: zsh crashes with $((##)) 2003-12-17 20:47:39 +00:00
Oliver Kiddle 244f2a14fc 18391: add tests for read builtin a few glob qualifiers, 17678 and 18015 2003-03-26 17:33:04 +00:00
Peter Stephenson 8840ce7144 16489: Test/C01arith.ztst: default output for whole numbers
as floats now has `.' at the end, c.f. 16423.
2002-01-23 16:28:16 +00:00
Peter Stephenson 7060b34457 15334: use string "(eval)" as script name in debugging messages in eval's 2001-07-09 18:31:24 +00:00
Peter Stephenson cef000cc62 15329: test script changes to go with 15327 LINENO change 2001-07-09 16:41:19 +00:00
Peter Stephenson 2c75f5b0da 15291, 15292: fix bug setting scalar in math mode 2001-07-06 18:33:59 +00:00
Sven Wischnowsky 231f9625af moved from ./Test/06arith.ztst 2001-04-02 12:32:11 +00:00