=================================================================== RCS file: /cvs/mandoc/read.c,v retrieving revision 1.5 retrieving revision 1.26 diff -u -p -r1.5 -r1.26 --- mandoc/read.c 2011/03/22 09:50:11 1.5 +++ mandoc/read.c 2011/11/07 01:24:40 1.26 @@ -1,4 +1,4 @@ -/* $Id: read.c,v 1.5 2011/03/22 09:50:11 kristaps Exp $ */ +/* $Id: read.c,v 1.26 2011/11/07 01:24:40 schwarze Exp $ */ /* * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons * Copyright (c) 2010, 2011 Ingo Schwarze @@ -15,9 +15,15 @@ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. */ -#include -#include +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif +#ifdef HAVE_MMAP +# include +# include +#endif + #include #include #include @@ -53,11 +59,11 @@ struct mparse { struct man *man; /* man parser */ struct mdoc *mdoc; /* mdoc parser */ struct roff *roff; /* roff parser (!NULL) */ - struct regset regs; /* roff registers */ int reparse_count; /* finite interp. stack */ mandocmsg mmsg; /* warning/error message handler */ void *arg; /* argument to mmsg */ const char *file; + struct buf *secondary; }; static void resize_buf(struct buf *, size_t); @@ -78,6 +84,133 @@ static const enum mandocerr mandoclimits[MANDOCLEVEL_M MANDOCERR_MAX }; +static const char * const mandocerrs[MANDOCERR_MAX] = { + "ok", + + "generic warning", + + /* related to the prologue */ + "no title in document", + "document title should be all caps", + "unknown manual section", + "date missing, using today's date", + "cannot parse date, using it verbatim", + "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)", + "NAME section must come first", + "bad NAME section contents", + "manual name not yet set", + "sections out of conventional order", + "duplicate section name", + "section not in conventional manual section", + + /* 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", + "line scope broken", + + /* 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", + + /* related to bad macro arguments */ + "skipping argument", + "duplicate argument", + "duplicate display type", + "duplicate list type", + "unknown AT&T UNIX version", + "bad Boolean value", + "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", + "bad escape sequence", + "unterminated quoted string", + + /* related to equations */ + "unexpected literal in equation", + + "generic error", + + /* related to equations */ + "unexpected equation scope closure", + "equation scope open on exit", + "overlapping equation scopes", + "unexpected end of equation", + "equation syntax 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", + "argument count wrong", + "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)", + "missing list type", + "line argument(s) will be lost", + "body argument(s) will be lost", + + "generic fatal error", + + "not a manual", + "column syntax is inconsistent", + "NOT IMPLEMENTED: .Bd -file", + "argument count wrong, violates syntax", + "child violates parent syntax", + "argument count wrong, violates syntax", + "NOT IMPLEMENTED: .so with absolute path or \"..\"", + "no document body", + "no document prologue", + "static buffer exhausted", +}; + +static const char * const mandoclevels[MANDOCLEVEL_MAX] = { + "SUCCESS", + "RESERVED", + "WARNING", + "ERROR", + "FATAL", + "BADARG", + "SYSERR" +}; + static void resize_buf(struct buf *buf, size_t initial) { @@ -113,13 +246,13 @@ pset(const char *buf, int pos, struct mparse *curp) switch (curp->inttype) { case (MPARSE_MDOC): if (NULL == curp->pmdoc) - curp->pmdoc = mdoc_alloc(&curp->regs, curp); + curp->pmdoc = mdoc_alloc(curp->roff, curp); assert(curp->pmdoc); curp->mdoc = curp->pmdoc; return; case (MPARSE_MAN): if (NULL == curp->pman) - curp->pman = man_alloc(&curp->regs, curp); + curp->pman = man_alloc(curp->roff, curp); assert(curp->pman); curp->man = curp->pman; return; @@ -129,14 +262,14 @@ pset(const char *buf, int pos, struct mparse *curp) if (pos >= 3 && 0 == memcmp(buf, ".Dd", 3)) { if (NULL == curp->pmdoc) - curp->pmdoc = mdoc_alloc(&curp->regs, curp); + curp->pmdoc = mdoc_alloc(curp->roff, curp); assert(curp->pmdoc); curp->mdoc = curp->pmdoc; return; } if (NULL == curp->pman) - curp->pman = man_alloc(&curp->regs, curp); + curp->pman = man_alloc(curp->roff, curp); assert(curp->pman); curp->man = curp->pman; } @@ -232,7 +365,7 @@ mparse_buf_r(struct mparse *curp, struct buf blk, int continue; } - if ('"' == blk.buf[i + 1]) { + if ('"' == blk.buf[i + 1] || '#' == blk.buf[i + 1]) { i += 2; /* Comment, skip to end of line */ for (; i < (int)blk.sz; ++i) { @@ -278,6 +411,27 @@ mparse_buf_r(struct mparse *curp, struct buf blk, int of = 0; + /* + * Maintain a lookaside buffer of all parsed lines. We + * only do this if mparse_keep() has been invoked (the + * buffer may be accessed with mparse_getkeep()). + */ + + if (curp->secondary) { + curp->secondary->buf = + mandoc_realloc + (curp->secondary->buf, + curp->secondary->sz + pos + 2); + memcpy(curp->secondary->buf + + curp->secondary->sz, + ln.buf, pos); + curp->secondary->sz += pos; + curp->secondary->buf + [curp->secondary->sz] = '\n'; + curp->secondary->sz++; + curp->secondary->buf + [curp->secondary->sz] = '\0'; + } rerun: rr = roff_parseln (curp->roff, curp->line, @@ -304,6 +458,13 @@ rerun: assert(MANDOCLEVEL_FATAL <= curp->file_status); break; case (ROFF_SO): + /* + * We remove `so' clauses from our lookaside + * buffer because we're going to descend into + * the file recursively. + */ + if (curp->secondary) + curp->secondary->sz -= pos + 1; mparse_readfd_r(curp, -1, ln.buf + of, 1); if (MANDOCLEVEL_FATAL <= curp->file_status) break; @@ -323,7 +484,7 @@ rerun: /* * If input parsers have not been allocated, do so now. - * We keep these instanced betwen parsers, but set them + * We keep these instanced between parsers, but set them * locally per parse routine since we can use different * parsers with each one. */ @@ -407,19 +568,22 @@ pdesc(struct mparse *curp, const char *file, int fd) mparse_buf_r(curp, blk, 1); +#ifdef HAVE_MMAP if (with_mmap) munmap(blk.buf, blk.sz); else +#endif free(blk.buf); } static int read_whole_file(const char *file, int fd, struct buf *fb, int *with_mmap) { - struct stat st; size_t off; ssize_t ssz; +#ifdef HAVE_MMAP + struct stat st; if (-1 == fstat(fd, &st)) { perror(file); return(0); @@ -444,6 +608,7 @@ read_whole_file(const char *file, int fd, struct buf * if (fb->buf != MAP_FAILED) return(1); } +#endif /* * If this isn't a regular file (like, say, stdin), then we must @@ -496,16 +661,11 @@ mparse_end(struct mparse *curp) return; } -#if 0 - /* FIXME: NOTE a parser may not have been assigned, yet. */ - if ( ! (curp->man || curp->mdoc)) { - /* FIXME: make into an mandoc.h error. */ - fprintf(stderr, "%s: Not a manual\n", curp->file); + mandoc_msg(MANDOCERR_NOTMANUAL, curp, 1, 0, NULL); curp->file_status = MANDOCLEVEL_FATAL; - goto cleanup; + return; } -#endif roff_endparse(curp->roff); } @@ -549,6 +709,8 @@ mparse_alloc(enum mparset inttype, enum mandoclevel wl { struct mparse *curp; + assert(wlevel <= MANDOCLEVEL_FATAL); + curp = mandoc_calloc(1, sizeof(struct mparse)); curp->wlevel = wlevel; @@ -556,7 +718,7 @@ mparse_alloc(enum mparset inttype, enum mandoclevel wl curp->arg = arg; curp->inttype = inttype; - curp->roff = roff_alloc(&curp->regs, curp); + curp->roff = roff_alloc(curp); return(curp); } @@ -564,14 +726,14 @@ void mparse_reset(struct mparse *curp) { - memset(&curp->regs, 0, sizeof(struct regset)); - roff_reset(curp->roff); if (curp->mdoc) mdoc_reset(curp->mdoc); if (curp->man) man_reset(curp->man); + if (curp->secondary) + curp->secondary->sz = 0; curp->file_status = MANDOCLEVEL_OK; curp->mdoc = NULL; @@ -588,7 +750,10 @@ mparse_free(struct mparse *curp) man_free(curp->pman); if (curp->roff) roff_free(curp->roff); + if (curp->secondary) + free(curp->secondary->buf); + free(curp->secondary); free(curp); } @@ -596,8 +761,10 @@ void mparse_result(struct mparse *curp, struct mdoc **mdoc, struct man **man) { - *mdoc = curp->mdoc; - *man = curp->man; + if (mdoc) + *mdoc = curp->mdoc; + if (man) + *man = curp->man; } void @@ -627,8 +794,38 @@ mandoc_msg(enum mandocerr er, struct mparse *m, if (level < m->wlevel) return; - (*m->mmsg)(er, level, m->file, ln, col, msg); + if (m->mmsg) + (*m->mmsg)(er, level, m->file, ln, col, msg); if (m->file_status < level) m->file_status = level; +} + +const char * +mparse_strerror(enum mandocerr er) +{ + + return(mandocerrs[er]); +} + +const char * +mparse_strlevel(enum mandoclevel lvl) +{ + return(mandoclevels[lvl]); +} + +void +mparse_keep(struct mparse *p) +{ + + assert(NULL == p->secondary); + p->secondary = mandoc_calloc(1, sizeof(struct buf)); +} + +const char * +mparse_getkeep(const struct mparse *p) +{ + + assert(p->secondary); + return(p->secondary->sz ? p->secondary->buf : NULL); }