summaryrefslogtreecommitdiff
path: root/usr.bin/m4
AgeCommit message (Collapse)Author
2002-04-28Implement -s.Marc Espie
Triggered by recent FreeBSD changes. - emits #line directives at every file change (like FreeBSD) - maintains a synch_lineno variable to verify when the output gets out of synch with the input, so that it can emit #line to re-synch as well (unlike FreeBSD) To do: either handle \end-of-line, or recognize when a macro expansion is in progress, so that line synch don't perturb cpp on multi-line expansions. With this, we should have a fully POSIX-compliant m4. ok miod@
2002-04-26use ansi function declarations. ok millert@Marc Espie
2002-04-26prototype mismatch, ansi conversion rules...Marc Espie
add -pedantic to flags.
2002-04-26fix indentation errorMarc Espie
2002-04-26rename `bool' to something else, to avoid C99 surprises.Marc Espie
2002-04-22Do not hardcode mvme88k-specific compilation flags here, especially sinceMiod Vallat
the new flags happen to be no different from those set in <sys.mk>
2002-04-18correct description of undefine, both according to the code, and toMarc Espie
single unix.
2002-02-16Part one of userland __P removal. Done with a simple regexp with some minor ↵Todd C. Miller
hand editing to make comments line up correctly. Another pass is forthcoming that handles the cases that could not be done automatically.
2001-12-28Kill yet another hardcoded size. Ok fries@Marc Espie
and flush(stdout) on debugging mode, as a vicious coredump can leave you without info otherwise.
2001-11-16incorrect writeTheo de Raadt
2001-10-13Resize buffer correctly, I'm an idiot.Marc Espie
Found by fries@, ok deraadt@
2001-10-10Handle macro expansion even if argc==2, which can happen in indir() calls.Marc Espie
Problem encountered by obecian@.
2001-10-10Prototypes shouldn't mention argument name.Marc Espie
2001-10-10Quote filename properly.Marc Espie
2001-10-10Fix indentMarc Espie
2001-10-06This PUTBACK must be protected against EOF. The others are fine.Marc Espie
2001-09-29show level of macro expansion, not dummy 1 (finally found some gnu-m4Marc Espie
scripts where that level != 1, hard to test otherwise...) Put a space after comma. Both these make trace output much closer to gnu-m4 output for debugging purposes.
2001-09-29Synchronize comment with reality.Marc Espie
In gnu-m4 mode, pass the real number of arguments to changequote/changecom, since those distinguish changequote() from changequote.
2001-09-29document tracing and compatibility with other m4.Marc Espie
2001-09-28Remove debugging scaffoldingMarc Espie
2001-09-28more emacs regexps in -g mode. Ok millert@Marc Espie
With this, some autoconf 2.52 scripts are now working completely.
2001-09-27pass all properties of builtins when defn...Marc Espie
2001-09-27Needed logic to handle newlines correctly.Marc Espie
2001-09-27traceon/traceoff built-ins.Marc Espie
2001-09-19inline some very common putback/chrsave. Worth roughly 10%Marc Espie
2001-09-18forgot to add -t handling.Marc Espie
2001-09-18better end-condition, fix tracing of dnl.Marc Espie
2001-09-18Tracing facilities. Same style as gnu-m4, because it's mostly used forMarc Espie
autoconf right now.
2001-09-18instrumentation for tracing mode: macro expansionMarc Espie
2001-09-18better gnu m4 emulation for dumpdefMarc Espie
2001-09-18let dumpdef go thru dump_one_def for each def.Marc Espie
2001-09-18let defn(builtin) work enough so thatMarc Espie
define(`newmacro', defn(builtin)) will work, as it should.
2001-09-18One single point for all macros/builtin expansion.Marc Espie
2001-09-188 bit clean look-ahead.Marc Espie
2001-09-18Fix regexp substitution.Marc Espie
- `vi rule' needs a pointer, because we must distinguish matches as the string position changes. - ^ should match only at beginning of line. - ^ should match at all beginnings of line.
2001-09-18slightly buggy macrosMarc Espie
2001-09-17Remove STATIC memory optimization. It's buggy (see regression test trip)Marc Espie
2001-09-16special-case MACRO (without parens) recognition, so that $# is set to 0Marc Espie
as it should.
2001-09-16Fix \\ in recognition in patsubst: must advance beyond it as well.Marc Espie
2001-07-28Wall cleanupPeter Valchev
2001-07-18Kill signal race. Reviewed by jjbg@Marc Espie
2001-07-18CDIAGFLAGSMarc Espie
2001-06-15Kill MAXSTR, it's no longer used at all.Marc Espie
Define MAXTOK directly.
2001-06-13Fix PR 1868. MAXSTR has no place there.Marc Espie
MAXSTR is still a hard limit, but only for token length, where 512 characters seems reasonable. Ok millert@
2001-03-05inspect MUST take an int to handle EOF issues correctly on unsigned charMarc Espie
machines. Problem reported by naddy.
2001-03-02Mark -o option with XXX (we may want to implement it someday, but currently itAaron Campbell
exists only for compatibility reasons); espie@ ok. This issue was brought to my attention by mpech@prosoft.org.lv
2001-02-05please alpha gccTheo de Raadt
2001-01-02m88k doesn't like optimization on m4.Steve Murphree
2000-11-10- Section shuffling: comply to the section ordering outlined in mdoc(7).Aaron Campbell
- Some .Nm trimming. - .Sh AUTHOR -> .Sh AUTHORS - Other miscellaneous fixes here and there.
2000-11-09Change all option list specifications to ".Bl -tag -width Ds". Most manAaron Campbell
pages just needed their -width parameter tweaked to "Ds", which provides a nice width of 6 constant characters. For consistency more than anything.
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-25 01:09:26 +0000