version 1.9, 2011/04/03 10:05:37 |
version 1.27, 2012/02/05 16:46:15 |
|
|
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF |
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF |
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. |
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. |
*/ |
*/ |
#include <sys/stat.h> |
#ifdef HAVE_CONFIG_H |
#include <sys/mman.h> |
#include "config.h" |
|
#endif |
|
|
|
#ifdef HAVE_MMAP |
|
# include <sys/stat.h> |
|
# include <sys/mman.h> |
|
#endif |
|
|
#include <assert.h> |
#include <assert.h> |
#include <ctype.h> |
#include <ctype.h> |
#include <fcntl.h> |
#include <fcntl.h> |
|
|
struct man *man; /* man parser */ |
struct man *man; /* man parser */ |
struct mdoc *mdoc; /* mdoc parser */ |
struct mdoc *mdoc; /* mdoc parser */ |
struct roff *roff; /* roff parser (!NULL) */ |
struct roff *roff; /* roff parser (!NULL) */ |
struct regset regs; /* roff registers */ |
|
int reparse_count; /* finite interp. stack */ |
int reparse_count; /* finite interp. stack */ |
mandocmsg mmsg; /* warning/error message handler */ |
mandocmsg mmsg; /* warning/error message handler */ |
void *arg; /* argument to mmsg */ |
void *arg; /* argument to mmsg */ |
const char *file; |
const char *file; |
|
struct buf *secondary; |
}; |
}; |
|
|
static void resize_buf(struct buf *, size_t); |
static void resize_buf(struct buf *, size_t); |
Line 138 static const char * const mandocerrs[MANDOCERR_MAX] = |
|
Line 144 static const char * const mandocerrs[MANDOCERR_MAX] = |
|
"tab in non-literal context", |
"tab in non-literal context", |
"end of line whitespace", |
"end of line whitespace", |
"bad comment style", |
"bad comment style", |
"unknown escape sequence", |
"bad escape sequence", |
"unterminated quoted string", |
"unterminated quoted string", |
|
|
|
/* related to equations */ |
|
"unexpected literal in equation", |
|
|
"generic error", |
"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 */ |
/* related to tables */ |
"bad table syntax", |
"bad table syntax", |
"bad table option", |
"bad table option", |
Line 176 static const char * const mandocerrs[MANDOCERR_MAX] = |
|
Line 192 static const char * const mandocerrs[MANDOCERR_MAX] = |
|
"not a manual", |
"not a manual", |
"column syntax is inconsistent", |
"column syntax is inconsistent", |
"NOT IMPLEMENTED: .Bd -file", |
"NOT IMPLEMENTED: .Bd -file", |
"line scope broken, syntax violated", |
|
"argument count wrong, violates syntax", |
"argument count wrong, violates syntax", |
"child violates parent syntax", |
"child violates parent syntax", |
"argument count wrong, violates syntax", |
"argument count wrong, violates syntax", |
Line 231 pset(const char *buf, int pos, struct mparse *curp) |
|
Line 246 pset(const char *buf, int pos, struct mparse *curp) |
|
switch (curp->inttype) { |
switch (curp->inttype) { |
case (MPARSE_MDOC): |
case (MPARSE_MDOC): |
if (NULL == curp->pmdoc) |
if (NULL == curp->pmdoc) |
curp->pmdoc = mdoc_alloc(&curp->regs, curp); |
curp->pmdoc = mdoc_alloc(curp->roff, curp); |
assert(curp->pmdoc); |
assert(curp->pmdoc); |
curp->mdoc = curp->pmdoc; |
curp->mdoc = curp->pmdoc; |
return; |
return; |
case (MPARSE_MAN): |
case (MPARSE_MAN): |
if (NULL == curp->pman) |
if (NULL == curp->pman) |
curp->pman = man_alloc(&curp->regs, curp); |
curp->pman = man_alloc(curp->roff, curp); |
assert(curp->pman); |
assert(curp->pman); |
curp->man = curp->pman; |
curp->man = curp->pman; |
return; |
return; |
Line 247 pset(const char *buf, int pos, struct mparse *curp) |
|
Line 262 pset(const char *buf, int pos, struct mparse *curp) |
|
|
|
if (pos >= 3 && 0 == memcmp(buf, ".Dd", 3)) { |
if (pos >= 3 && 0 == memcmp(buf, ".Dd", 3)) { |
if (NULL == curp->pmdoc) |
if (NULL == curp->pmdoc) |
curp->pmdoc = mdoc_alloc(&curp->regs, curp); |
curp->pmdoc = mdoc_alloc(curp->roff, curp); |
assert(curp->pmdoc); |
assert(curp->pmdoc); |
curp->mdoc = curp->pmdoc; |
curp->mdoc = curp->pmdoc; |
return; |
return; |
} |
} |
|
|
if (NULL == curp->pman) |
if (NULL == curp->pman) |
curp->pman = man_alloc(&curp->regs, curp); |
curp->pman = man_alloc(curp->roff, curp); |
assert(curp->pman); |
assert(curp->pman); |
curp->man = curp->pman; |
curp->man = curp->pman; |
} |
} |
Line 310 mparse_buf_r(struct mparse *curp, struct buf blk, int |
|
Line 325 mparse_buf_r(struct mparse *curp, struct buf blk, int |
|
* Warn about bogus characters. If you're using |
* Warn about bogus characters. If you're using |
* non-ASCII encoding, you're screwing your |
* non-ASCII encoding, you're screwing your |
* readers. Since I'd rather this not happen, |
* readers. Since I'd rather this not happen, |
* I'll be helpful and drop these characters so |
* I'll be helpful and replace these characters |
* we don't display gibberish. Note to manual |
* with "?", so we don't display gibberish. |
* writers: use special characters. |
* Note to manual writers: use special characters. |
*/ |
*/ |
|
|
c = (unsigned char) blk.buf[i]; |
c = (unsigned char) blk.buf[i]; |
Line 320 mparse_buf_r(struct mparse *curp, struct buf blk, int |
|
Line 335 mparse_buf_r(struct mparse *curp, struct buf blk, int |
|
if ( ! (isascii(c) && |
if ( ! (isascii(c) && |
(isgraph(c) || isblank(c)))) { |
(isgraph(c) || isblank(c)))) { |
mandoc_msg(MANDOCERR_BADCHAR, curp, |
mandoc_msg(MANDOCERR_BADCHAR, curp, |
curp->line, pos, "ignoring byte"); |
curp->line, pos, NULL); |
i++; |
i++; |
|
if (pos >= (int)ln.sz) |
|
resize_buf(&ln, 256); |
|
ln.buf[pos++] = '?'; |
continue; |
continue; |
} |
} |
|
|
Line 350 mparse_buf_r(struct mparse *curp, struct buf blk, int |
|
Line 368 mparse_buf_r(struct mparse *curp, struct buf blk, int |
|
continue; |
continue; |
} |
} |
|
|
if ('"' == blk.buf[i + 1]) { |
if ('"' == blk.buf[i + 1] || '#' == blk.buf[i + 1]) { |
i += 2; |
i += 2; |
/* Comment, skip to end of line */ |
/* Comment, skip to end of line */ |
for (; i < (int)blk.sz; ++i) { |
for (; i < (int)blk.sz; ++i) { |
Line 396 mparse_buf_r(struct mparse *curp, struct buf blk, int |
|
Line 414 mparse_buf_r(struct mparse *curp, struct buf blk, int |
|
|
|
of = 0; |
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: |
rerun: |
rr = roff_parseln |
rr = roff_parseln |
(curp->roff, curp->line, |
(curp->roff, curp->line, |
|
|
assert(MANDOCLEVEL_FATAL <= curp->file_status); |
assert(MANDOCLEVEL_FATAL <= curp->file_status); |
break; |
break; |
case (ROFF_SO): |
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); |
mparse_readfd_r(curp, -1, ln.buf + of, 1); |
if (MANDOCLEVEL_FATAL <= curp->file_status) |
if (MANDOCLEVEL_FATAL <= curp->file_status) |
break; |
break; |
|
|
|
|
/* |
/* |
* If input parsers have not been allocated, do so now. |
* 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 |
* locally per parse routine since we can use different |
* parsers with each one. |
* parsers with each one. |
*/ |
*/ |
Line 525 pdesc(struct mparse *curp, const char *file, int fd) |
|
Line 571 pdesc(struct mparse *curp, const char *file, int fd) |
|
|
|
mparse_buf_r(curp, blk, 1); |
mparse_buf_r(curp, blk, 1); |
|
|
|
#ifdef HAVE_MMAP |
if (with_mmap) |
if (with_mmap) |
munmap(blk.buf, blk.sz); |
munmap(blk.buf, blk.sz); |
else |
else |
|
#endif |
free(blk.buf); |
free(blk.buf); |
} |
} |
|
|
static int |
static int |
read_whole_file(const char *file, int fd, struct buf *fb, int *with_mmap) |
read_whole_file(const char *file, int fd, struct buf *fb, int *with_mmap) |
{ |
{ |
struct stat st; |
|
size_t off; |
size_t off; |
ssize_t ssz; |
ssize_t ssz; |
|
|
|
#ifdef HAVE_MMAP |
|
struct stat st; |
if (-1 == fstat(fd, &st)) { |
if (-1 == fstat(fd, &st)) { |
perror(file); |
perror(file); |
return(0); |
return(0); |
Line 562 read_whole_file(const char *file, int fd, struct buf * |
|
Line 611 read_whole_file(const char *file, int fd, struct buf * |
|
if (fb->buf != MAP_FAILED) |
if (fb->buf != MAP_FAILED) |
return(1); |
return(1); |
} |
} |
|
#endif |
|
|
/* |
/* |
* If this isn't a regular file (like, say, stdin), then we must |
* If this isn't a regular file (like, say, stdin), then we must |
Line 662 mparse_alloc(enum mparset inttype, enum mandoclevel wl |
|
Line 712 mparse_alloc(enum mparset inttype, enum mandoclevel wl |
|
{ |
{ |
struct mparse *curp; |
struct mparse *curp; |
|
|
|
assert(wlevel <= MANDOCLEVEL_FATAL); |
|
|
curp = mandoc_calloc(1, sizeof(struct mparse)); |
curp = mandoc_calloc(1, sizeof(struct mparse)); |
|
|
curp->wlevel = wlevel; |
curp->wlevel = wlevel; |
Line 669 mparse_alloc(enum mparset inttype, enum mandoclevel wl |
|
Line 721 mparse_alloc(enum mparset inttype, enum mandoclevel wl |
|
curp->arg = arg; |
curp->arg = arg; |
curp->inttype = inttype; |
curp->inttype = inttype; |
|
|
curp->roff = roff_alloc(&curp->regs, curp); |
curp->roff = roff_alloc(curp); |
return(curp); |
return(curp); |
} |
} |
|
|
|
|
mparse_reset(struct mparse *curp) |
mparse_reset(struct mparse *curp) |
{ |
{ |
|
|
memset(&curp->regs, 0, sizeof(struct regset)); |
|
|
|
roff_reset(curp->roff); |
roff_reset(curp->roff); |
|
|
if (curp->mdoc) |
if (curp->mdoc) |
mdoc_reset(curp->mdoc); |
mdoc_reset(curp->mdoc); |
if (curp->man) |
if (curp->man) |
man_reset(curp->man); |
man_reset(curp->man); |
|
if (curp->secondary) |
|
curp->secondary->sz = 0; |
|
|
curp->file_status = MANDOCLEVEL_OK; |
curp->file_status = MANDOCLEVEL_OK; |
curp->mdoc = NULL; |
curp->mdoc = NULL; |
Line 701 mparse_free(struct mparse *curp) |
|
Line 753 mparse_free(struct mparse *curp) |
|
man_free(curp->pman); |
man_free(curp->pman); |
if (curp->roff) |
if (curp->roff) |
roff_free(curp->roff); |
roff_free(curp->roff); |
|
if (curp->secondary) |
|
free(curp->secondary->buf); |
|
|
|
free(curp->secondary); |
free(curp); |
free(curp); |
} |
} |
|
|
|
|
mparse_strlevel(enum mandoclevel lvl) |
mparse_strlevel(enum mandoclevel lvl) |
{ |
{ |
return(mandoclevels[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); |
} |
} |