Re: PATCH: 3.2.0b5 config parser fixes
"Joe R. Jah" <[email protected]>
| Newsgroups | gmane.comp.web.htdig.devel |
|---|---|
| Message-ID | <[email protected]> |
On Fri, 23 Apr 2004, Gilles Detillieux wrote: > Date: Fri, 23 Apr 2004 14:35:53 -0500 (CDT) > From: Gilles Detillieux <[email protected]> > To: Joe R. Jah <[email protected]> > Cc: "ht://Dig developers list" <[email protected]> > Subject: Re: [htdig-dev] PATCH: 3.2.0b5 config parser fixes > > According to Joe R. Jah: > > patch -p0 < ../patch/config_parser.0 > > patching file `htcommon/conf_lexer.lxx' > > patching file `htcommon/conf_parser.yxx' > > Hunk #4 FAILED at 254. > > 1 out of 4 hunks FAILED -- saving rejects to htcommon/conf_parser.yxx.rej > > That hunk failed because I built the patch against the CVS version of > conf_parser.yxx, not the version in 3.2.0b5. Back in November, Lachlan > changed the printf in yyerror() from > > printf ("%s\nIn line %d\n",s,yylineno); > > to > > fprintf (stderr, "%s\nIn line %d\n",s,yylineno); > > so the patch program choked on the hunk that patches yyerror(). To make > the config_parser.0 patch compatible with 3.2.0b5, you'd need to make > the reverse change to that line (prefixed by "-") in the patch. I believe we should make the patches against virgin 3.2.0b5; alternatively we should indicate to what snapshot they apply. Many people do not download weekly snapshots, and use the patches to upgrade 3.2.0b5. I plead guilty, until a few minutes ago;) I see that except one, the patches in the patch site have not been committed to 3.2.0b5-20040418; when will they be committed? > > Manually applied the rejected hunk, but: > > > > conf_parser.cxx: In function `int yy_do_parse(yystack *)': > > conf_parser.cxx:287: warning: suggest parentheses around assignment used > > as truth value > > conf_parser.cxx:403: `aConf' undeclared (first use this function) > > conf_parser.cxx:403: (Each undeclared identifier is reported only once > > conf_parser.cxx:403: for each function it appears in.) > > conf_parser.cxx:334: warning: label `yyerrlab' defined but not used > > conf_parser.cxx:329: warning: label `yynewerror' defined but not used > > conf_parser.cxx: At top level: > > conf_parser.cxx:2: warning: `char yysccsid[43]' defined but not used > > gmake[1]: *** [conf_parser.lo] Error 1 > > That's odd. Did your system use yacc or bison to build conf_parser.cxx > from the .yxx source? Which version of this tool do you have? The config > parser relies on the "#define YYPARSE_PARAM aConf" statement to get bison > to use aConf as the argument to yyparse(), so the declaration is supposed > to be auto-generated. I guess for completeness, the patch would need to > include the autogenerated *.h and *.cxx files as well. My system used yacc; it's time stamp is relatively recent: -r-xr-xr-x 1 bin bin 54144 Jul 17 2003 /usr/bin/yacc I also reconfigured with GNU Bison version 1.28; this is what I got:( sed '/^#/ s|y\.tab\.c|conf_parser.cxx|' y.tab.c >conf_parser.cxxt && mv conf_parser.cxxt conf_parser.cxx sed: y.tab.c: No such file or directory gmake[1]: *** [conf_parser.cxx] Error 1 gmake[1]: Leaving directory `/usr/tmp/htdig/htdig-3.2.0b5/htcommon' gmake: *** [all-recursive] Error 1 Does it require a more recent version of Bison? Regards, Joe -- _/ _/_/_/ _/ ____________ __o _/ _/ _/ _/ ______________ _-\<,_ _/ _/ _/_/_/ _/ _/ ......(_)/ (_) _/_/ oe _/ _/. _/_/ ah [email protected] ------------------------------------------------------- This SF.net email is sponsored by: The Robotic Monkeys at ThinkGeek For a limited time only, get FREE Ground shipping on all orders of $35 or more. Hurry up and shop folks, this offer expires April 30th! http://www.thinkgeek.com/freeshipping/?cpg=12297 _______________________________________________ ht://Dig Developer mailing list: [email protected] List information (subscribe/unsubscribe, etc.) https://lists.sourceforge.net/lists/listinfo/htdig-dev