=================================================================== RCS file: /cvs/mandoc/main.c,v retrieving revision 1.118 retrieving revision 1.143 diff -u -p -r1.118 -r1.143 --- mandoc/main.c 2010/12/05 15:55:01 1.118 +++ mandoc/main.c 2011/02/06 20:36:36 1.143 @@ -1,7 +1,7 @@ -/* $Id: main.c,v 1.118 2010/12/05 15:55:01 kristaps Exp $ */ +/* $Id: main.c,v 1.143 2011/02/06 20:36:36 kristaps Exp $ */ /* - * Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons - * Copyright (c) 2010 Ingo Schwarze + * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons + * Copyright (c) 2010, 2011 Ingo Schwarze * * Permission to use, copy, modify, and distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -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 */ @@ -122,71 +124,97 @@ 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", - "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", + "skipping no-space 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", + "skipping end of block that is not open", - "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", - "NOT IMPLEMENTED: skipping request", + "unknown escape sequence", + "unterminated quoted string", + + "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", + "data block still open", + "ignoring extra data cells", + + "input stack limit exceeded, infinite loop?", + "skipping bad character", + "escaped character not allowed in a name", + "skipping text before the first section header", + "skipping unknown macro", + "NOT IMPLEMENTED, please use groff: 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", @@ -375,7 +403,7 @@ 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); @@ -508,10 +536,7 @@ 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 @@ -632,6 +657,7 @@ pdesc(struct curparse *curp) static void parsebuf(struct curparse *curp, struct buf blk, int start) { + const struct tbl_span *span; struct buf ln; enum rofferr rr; int i, of, rc; @@ -654,10 +680,22 @@ 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])) { + + /* + * When finding an unescaped newline character, + * leave the character loop to process the line. + * Skip a preceding carriage return, if any. + */ + + if ('\r' == blk.buf[i] && i + 1 < (int)blk.sz && + '\n' == blk.buf[i + 1]) + ++i; if ('\n' == blk.buf[i]) { ++i; ++lnn; @@ -692,11 +730,18 @@ parsebuf(struct curparse *curp, struct buf blk, int st continue; } - /* Found escape & at least one other char. */ + /* + * Found escape and at least one other character. + * When it's a newline character, skip it. + * When there is a carriage return in between, + * skip that one as well. + */ + if ('\r' == blk.buf[i + 1] && i + 2 < (int)blk.sz && + '\n' == blk.buf[i + 2]) + ++i; if ('\n' == blk.buf[i + 1]) { i += 2; - /* Escaped newlines are skipped over */ ++lnn; continue; } @@ -754,7 +799,11 @@ 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): @@ -774,11 +823,19 @@ 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 @@ -792,9 +849,23 @@ 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); + while (NULL != (span = roff_span(curp->roff))) { + if (curp->man) + man_addspan(curp->man, span); + else + mdoc_addspan(curp->mdoc, span); + } + } else if (ROFF_EQN == rr) { + assert(curp->man || curp->mdoc); + } else if (curp->man || curp->mdoc) { rc = curp->man ? man_parseln(curp->man, curp->line, ln.buf, of) :