=================================================================== RCS file: /cvs/mandoc/main.c,v retrieving revision 1.117 retrieving revision 1.134 diff -u -p -r1.117 -r1.134 --- mandoc/main.c 2010/12/05 15:49:37 1.117 +++ mandoc/main.c 2011/01/04 12:06:21 1.134 @@ -1,6 +1,6 @@ -/* $Id: main.c,v 1.117 2010/12/05 15:49:37 kristaps Exp $ */ +/* $Id: main.c,v 1.134 2011/01/04 12:06:21 kristaps Exp $ */ /* - * Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons + * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons * Copyright (c) 2010 Ingo Schwarze * * Permission to use, copy, modify, and distribute this software for any @@ -41,6 +41,7 @@ #define MAP_FILE 0 #endif +#define REPARSE_LIMIT 1000 #define UNCONST(a) ((void *)(uintptr_t)(const void *)(a)) /* FIXME: Intel's compiler? LLVM? pcc? */ @@ -89,6 +90,7 @@ struct curparse { struct mdoc *mdoc; /* mdoc parser */ struct roff *roff; /* roff parser (!NULL) */ struct regset regs; /* roff registers */ + int reparse_count; /* finite interpolation stack */ enum outt outtype; /* which output to use */ out_mdoc outmdoc; /* mdoc output ptr */ out_man outman; /* man output ptr */ @@ -122,72 +124,95 @@ static const char * const mandocerrs[MANDOCERR_MAX] = "generic warning", + /* related to the prologue */ + "no title in document", + "document title should be all caps", + "unknown manual section", + "cannot parse date argument", + "prologue macros out of order", + "duplicate prologue macro", + "macro not allowed in prologue", + "macro not allowed in body", + + /* related to document structure */ ".so is fragile, better use ln(1)", - "text should be uppercase", + "NAME section must come first", + "bad NAME section contents", + "manual name not yet set", "sections out of conventional order", - "section name repeats", - "out of order prologue", - "repeated prologue entry", - "list type must come first", - "tab in non-literal context", - "bad escape sequence", - "unterminated quoted string", - "argument requires the width argument", - "superfluous width argument", - "bad date argument", - "bad width argument", - "unknown manual section", + "duplicate section name", "section not in conventional manual section", - "end of line whitespace", + + /* related to macros and nesting */ + "skipping obsolete macro", + "skipping paragraph macro", "blocks badly nested", + "child violates parent syntax", + "nested displays are not portable", + "already in literal mode", - "generic error", + /* related to missing macro arguments */ + "skipping empty macro", + "argument count wrong", + "missing display type", + "list type must come first", + "tag lists require a width argument", + "missing font type", - "NAME section must come first", + /* related to bad macro arguments */ + "skipping argument", + "duplicate argument", + "duplicate display type", + "duplicate list type", + "unknown AT&T UNIX version", "bad Boolean value", - "child violates parent syntax", - "displays may not be nested", - "bad AT&T symbol", - "bad standard", - "list type repeated", - "display type repeated", - "argument repeated", - "ignoring argument", - "manual name not yet set", - "obsolete macro ignored", - "empty macro ignored", - "macro not allowed in body", - "macro not allowed in prologue", - "bad character", - "bad NAME section contents", - "no blank lines", - "no text in this context", + "unknown font", + "unknown standard specifier", + "bad width argument", + + /* related to plain text */ + "blank line in non-literal context", + "tab in non-literal context", + "end of line whitespace", "bad comment style", - "unknown macro will be lost", + "unknown escape sequence", + "unterminated quoted string", + + /* related to tables */ + "extra data cells", + + "generic error", + + /* related to tables */ + "bad table syntax", + "bad table option", + "bad table layout", + "no table layout cells specified", + "no table data cells specified", + "ignore data in cell", + + "input stack limit exceeded, infinite loop?", + "skipping bad character", + "skipping text before the first section header", + "skipping unknown macro", "NOT IMPLEMENTED: skipping request", "line scope broken", "argument count wrong", - "request scope close w/none open", - "scope already open", + "skipping end of block that is not open", + "missing end of block", "scope open on exit", "uname(3) system call failed", "macro requires line argument(s)", "macro requires body argument(s)", "macro requires argument(s)", - "no title in document", "missing list type", - "missing display type", - "missing font type", "line argument(s) will be lost", "body argument(s) will be lost", - "paragraph macro ignored", "generic fatal error", "column syntax is inconsistent", - "unsupported display type", - "blocks badly nested", - "no such block is open", + "NOT IMPLEMENTED: .Bd -file", "line scope broken, syntax violated", "argument count wrong, violates syntax", "child violates parent syntax", @@ -376,7 +401,7 @@ static void resize_buf(struct buf *buf, size_t initial) { - buf->sz = buf->sz ? 2 * buf->sz : initial; + buf->sz = buf->sz > initial/2 ? 2 * buf->sz : initial; buf->buf = realloc(buf->buf, buf->sz); if (NULL == buf->buf) { perror(NULL); @@ -509,10 +534,7 @@ fdesc(struct curparse *curp) } assert(curp->roff); - if ( ! roff_endparse(curp->roff)) { - assert(MANDOCLEVEL_FATAL <= file_status); - goto cleanup; - } + roff_endparse(curp->roff); /* * With -Wstop and warnings or errors of at least @@ -655,8 +677,10 @@ parsebuf(struct curparse *curp, struct buf blk, int st if (0 == pos && '\0' == blk.buf[i]) break; - if (start) + if (start) { curp->line = lnn; + curp->reparse_count = 0; + } while (i < (int)blk.sz && (start || '\0' != blk.buf[i])) { if ('\n' == blk.buf[i]) { @@ -755,7 +779,11 @@ rerun: switch (rr) { case (ROFF_REPARSE): - parsebuf(curp, ln, 0); + if (REPARSE_LIMIT >= ++curp->reparse_count) + parsebuf(curp, ln, 0); + else + mmsg(MANDOCERR_ROFFLOOP, curp, + curp->line, pos, NULL); pos = 0; continue; case (ROFF_APPEND): @@ -775,11 +803,19 @@ rerun: continue; } else break; - case (ROFF_CONT): + default: break; } /* + * If we encounter errors in the recursive parsebuf() + * call, make sure we don't continue parsing. + */ + + if (MANDOCLEVEL_FATAL <= file_status) + break; + + /* * If input parsers have not been allocated, do so now. * We keep these instanced betwen parsers, but set them * locally per parse routine since we can use different @@ -793,9 +829,20 @@ rerun: * Lastly, push down into the parsers themselves. One * of these will have already been set in the pset() * routine. + * If libroff returns ROFF_TBL, then add it to the + * currently open parse. Since we only get here if + * there does exist data (see tbl_data.c), we're + * guaranteed that something's been allocated. */ - if (curp->man || curp->mdoc) { + if (ROFF_TBL == rr) { + assert(curp->man || curp->mdoc); + if (curp->man) + man_addspan(curp->man, roff_span(curp->roff)); + else + mdoc_addspan(curp->mdoc, roff_span(curp->roff)); + + } else if (curp->man || curp->mdoc) { rc = curp->man ? man_parseln(curp->man, curp->line, ln.buf, of) :