X-Git-Url: https://git.saurik.com/bison.git/blobdiff_plain/20c37f213a14c112251907734402cbf80f44d780..8b9f2372c70b91359921b4967275b02b83cdc110:/TODO?ds=sidebyside diff --git a/TODO b/TODO index c87808a3..14bd676f 100644 --- a/TODO +++ b/TODO @@ -1,5 +1,59 @@ -*- outline -*- +* URGENT: Documenting C++ output +Write a first documentation for C++ output. + +* value_components_used +Was defined but not used: where was it coming from? It can't be to +check if %union is used, since the user is free to $n on her +union, doesn't she? + +* yyerror, yyprint interface +It should be improved, in particular when using Bison features such as +locations, and YYPARSE_PARAMS. For the time being, it is recommended +to #define yyerror and yyprint to steal internal variables... + +* documentation +Explain $axiom (and maybe change its name: BTYacc names it `goal', +byacc `$accept', probably based on AT&T Yacc). Complete the glossary +(item, axiom, ?). + +* report documentation +Extend with error. The hard part will probably be finding the right +rule so that a single state does not exhibit to many yet undocumented +``features''. Maybe an empty action ought to be presented too. Shall +we try to make a single grammar with all these features, or should we +have several very small grammars? + +* documentation +Some history of Bison and some bibliography would be most welcome. +Are there any Texinfo standards for bibliography? + +* Several %unions +I think this is a pleasant (but useless currently) feature, but in the +future, I want a means to %include other bits of grammars, and _then_ +it will be important for the various bits to define their needs in +%union. + +When implementing multiple-%union support, bare the following in mind: + +- when --yacc, this must be flagged as an error. Don't make it fatal + though. + +- The #line must now appear *inside* the definition of yystype. + Something like + + { + #line 12 "foo.y" + int ival; + #line 23 "foo.y" + char *sval; + } + +* --report=conflict-path +Provide better assistance for understanding the conflicts by providing +a sample text exhibiting the (LALR) ambiguity. + * Coding system independence Paul notes: @@ -13,36 +67,6 @@ Paul notes: PDP-10 ports :-) but they should probably be documented somewhere. -* Using enums instead of int for tokens. -Paul suggests: - - #ifndef YYTOKENTYPE - # if defined (__STDC__) || defined (__cplusplus) - /* Put the tokens into the symbol table, so that GDB and other debuggers - know about them. */ - enum yytokentype { - FOO = 256, - BAR, - ... - }; - /* POSIX requires `int' for tokens in interfaces. */ - # define YYTOKENTYPE int - # endif - #endif - #define FOO 256 - #define BAR 257 - ... - -> I'm in favor of -> -> %token FOO 256 -> %token BAR 257 -> -> and Bison moves error into 258. - -Yes, I think that's a valid extension too, if the user doesn't define -the token number for error. - * Output directory Akim: @@ -152,11 +176,6 @@ src/bison/tests % cd ./testsuite.dir/51 tests/testsuite.dir/51 % echo "()" | ./calc 1.2-1.3: parse error, unexpected ')', expecting error or "number" or '-' or '(' -* yyerror, yyprint interface -It should be improved, in particular when using Bison features such as -locations, and YYPARSE_PARAMS. For the time being, it is recommended -to #define yyerror and yyprint to steal internal variables... - * read_pipe.c This is not portable to DOS for instance. Implement a more portable scheme. Sources of inspiration include GNU diff, and Free Recode. @@ -171,40 +190,6 @@ critical for user data: when aborting a parsing, when handling the error token etc., we often throw away yylval without giving a chance of cleaning it up to the user. -* NEWS -Sort from 1.31 NEWS. - -* Prologue -The %union is declared after the user C declarations. It can be -a problem if YYSTYPE is declared after the user part. [] - -Actually, the real problem seems that the %union ought to be output -where it was defined. For instance, in gettext/intl/plural.y, we -have: - - %{ - ... - #include "gettextP.h" - ... - %} - - %union { - unsigned long int num; - enum operator op; - struct expression *exp; - } - - %{ - ... - static int yylex PARAMS ((YYSTYPE *lval, const char **pexp)); - ... - %} - -Where the first part defines struct expression, the second uses it to -define YYSTYPE, and the last uses YYSTYPE. Only this order is valid. - -Note that we have the same problem with GCC. - * --graph Show reductions. [] @@ -212,7 +197,6 @@ Show reductions. [] ** %no-lines [ok] ** %no-parser [] ** %pure-parser [] -** %semantic-parser [] ** %token-table [] ** Options which could use parse_dquoted_param (). Maybe transfered in lex.c. @@ -313,55 +297,75 @@ It is unfortunate that there is a total order for precedence. It makes it impossible to have modular precedence information. We should move to partial orders. -* Parsing grammars -Rewrite the reader in Bison. - -* Problems with aliases -From: "Baum, Nathan I" -Subject: Token Alias Bug -To: "'bug-bison@gnu.org'" - -I've noticed a bug in bison. Sadly, our eternally wise sysadmins won't let -us use CVS, so I can't find out if it's been fixed already... - -Basically, I made a program (in flex) that went through a .y file looking -for "..."-tokens, and then outputed a %token -line for it. For single-character ""-tokens, I reasoned, I could just use -[%token 'A' "A"]. However, this causes Bison to output a [#define 'A' 65], -which cppp chokes on, not unreasonably. (And even if cppp didn't choke, I -obviously wouldn't want (char)'A' to be replaced with (int)65 throughout my -code. - -Bison normally forgoes outputing a #define for a character token. However, -it always outputs an aliased token -- even if the token is an alias for a -character token. We don't want that. The problem is in /output.c/, as I -recall. When it outputs the token definitions, it checks for a character -token, and then checks for an alias token. If the character token check is -placed after the alias check, then it works correctly. - -Alias tokens seem to be something of a kludge. What about an [%alias "..."] -command... - - %alias T_IF "IF" - -Hmm. I can't help thinking... What about a --generate-lex option that -creates an .l file for the alias tokens used... (Or an option to make a -gperf file, etc...) - -* Presentation of the report file -From: "Baum, Nathan I" -Subject: Token Alias Bug -To: "'bug-bison@gnu.org'" - -I've also noticed something, that whilst not *wrong*, is inconvienient: I -use the verbose mode to help find the causes of unresolved shift/reduce -conflicts. However, this mode insists on starting the .output file with a -list of *resolved* conflicts, something I find quite useless. Might it be -possible to define a -v mode, and a -vv mode -- Where the -vv mode shows -everything, but the -v mode only tells you what you need for examining -conflicts? (Or, perhaps, a "*** This state has N conflicts ***" marker above -each state with conflicts.) +This will be possible with a Bison parser for the grammar, as it will +make it much easier to extend the grammar. +* Parsing grammars +Rewrite the reader in Flex/Bison. There will be delicate parts, in +particular, expect the scanner to be hard to write. Many interesting +features cannot be implemented without such a new reader. + +* $undefined +From Hans: +- If the Bison generated parser experiences an undefined number in the +character range, that character is written out in diagnostic messages, an +addition to the $undefined value. + +Suggest: Change the name $undefined to undefined; looks better in outputs. + +* Default Action +From Hans: +- For use with my C++ parser, I transported the "switch (yyn)" statement +that Bison writes to the bison.simple skeleton file. This way, I can remove +the current default rule $$ = $1 implementation, which causes a double +assignment to $$ which may not be OK under C++, replacing it with a +"default:" part within the switch statement. + +Note that the default rule $$ = $1, when typed, is perfectly OK under C, +but in the C++ implementation I made, this rule is different from +$$ = $1. I therefore think that one should implement +a Bison option where every typed default rule is explicitly written out +(same typed ruled can of course be grouped together). + +Note: Robert Anisko handles this. He knows how to do it. + +* Warnings +It would be nice to have warning support. See how Autoconf handles +them, it is fairly well described there. It would be very nice to +implement this in such a way that other programs could use +lib/warnings.[ch]. + +Don't work on this without first announcing you do, as I already have +thought about it, and know many of the components that can be used to +implement it. + +* Pre and post actions. +From: Florian Krohm +Subject: YYACT_EPILOGUE +To: bug-bison@gnu.org +X-Sent: 1 week, 4 days, 14 hours, 38 minutes, 11 seconds ago + +The other day I had the need for explicitly building the parse tree. I +used %locations for that and defined YYLLOC_DEFAULT to call a function +that returns the tree node for the production. Easy. But I also needed +to assign the S-attribute to the tree node. That cannot be done in +YYLLOC_DEFAULT, because it is invoked before the action is executed. +The way I solved this was to define a macro YYACT_EPILOGUE that would +be invoked after the action. For reasons of symmetry I also added +YYACT_PROLOGUE. Although I had no use for that I can envision how it +might come in handy for debugging purposes. +All is needed is to add + +#if YYLSP_NEEDED + YYACT_EPILOGUE (yyval, (yyvsp - yylen), yylen, yyloc, (yylsp - yylen)); +#else + YYACT_EPILOGUE (yyval, (yyvsp - yylen), yylen); +#endif + +at the proper place to bison.simple. Ditto for YYACT_PROLOGUE. + +I was wondering what you think about adding YYACT_PROLOGUE/EPILOGUE +to bison. If you're interested, I'll work on a patch. -----