Small tweaks for release.

[mandoc.git] / read.c

diff --git a/read.c b/read.c
index 8c1da44a079f3bf386ad7870ffa2096115d897a5..5b14e357d2bd8bfafe8f3a63f9ae8dfe3a19a976 100644 (file)
--- a/read.c
+++ b/read.c
@@ -1,4 +1,4 @@
-/*     $Id: read.c,v 1.24 2011/10/06 22:29:12 kristaps Exp $ */
+/*     $Id: read.c,v 1.28 2012/02/16 20:51:31 joerg Exp $ */
 /*
  * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2010, 2011 Ingo Schwarze <schwarze@openbsd.org>
@@ -28,6 +28,7 @@
 #include <ctype.h>
 #include <fcntl.h>
 #include <stdarg.h>
+#include <stdint.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
@@ -37,6 +38,7 @@
 #include "libmandoc.h"
 #include "mdoc.h"
 #include "man.h"
+#include "main.h"
 
 #ifndef MAP_FILE
 #define        MAP_FILE        0
@@ -70,7 +72,6 @@ static        void      resize_buf(struct buf *, size_t);
 static void      mparse_buf_r(struct mparse *, struct buf, int);
 static void      mparse_readfd_r(struct mparse *, int, const char *, int);
 static void      pset(const char *, int, struct mparse *);
-static void      pdesc(struct mparse *, const char *, int);
 static int       read_whole_file(const char *, int, struct buf *, int *);
 static void      mparse_end(struct mparse *);
 
@@ -192,7 +193,6 @@ static      const char * const      mandocerrs[MANDOCERR_MAX] = {
        "not a manual",
        "column syntax is inconsistent",
        "NOT IMPLEMENTED: .Bd -file",
-       "line scope broken, syntax violated",
        "argument count wrong, violates syntax",
        "child violates parent syntax",
        "argument count wrong, violates syntax",
@@ -326,9 +326,9 @@ mparse_buf_r(struct mparse *curp, struct buf blk, int start)
                         * Warn about bogus characters.  If you're using
                         * non-ASCII encoding, you're screwing your
                         * readers.  Since I'd rather this not happen,
-                        * I'll be helpful and drop these characters so
-                        * we don't display gibberish.  Note to manual
-                        * writers: use special characters.
+                        * I'll be helpful and replace these characters
+                        * with "?", so we don't display gibberish.
+                        * Note to manual writers: use special characters.
                         */
 
                        c = (unsigned char) blk.buf[i];
@@ -336,8 +336,11 @@ mparse_buf_r(struct mparse *curp, struct buf blk, int start)
                        if ( ! (isascii(c) && 
                                        (isgraph(c) || isblank(c)))) {
                                mandoc_msg(MANDOCERR_BADCHAR, curp,
-                                               curp->line, pos, "ignoring byte");
+                                               curp->line, pos, NULL);
                                i++;
+                               if (pos >= (int)ln.sz)
+                                       resize_buf(&ln, 256);
+                               ln.buf[pos++] = '?';
                                continue;
                        }
 
@@ -464,7 +467,8 @@ rerun:
                         * buffer because we're going to descend into
                         * the file recursively.
                         */
-                       curp->secondary->sz -= pos + 1;
+                       if (curp->secondary) 
+                               curp->secondary->sz -= pos + 1;
                        mparse_readfd_r(curp, -1, ln.buf + of, 1);
                        if (MANDOCLEVEL_FATAL <= curp->file_status)
                                break;
@@ -544,38 +548,6 @@ rerun:
        free(ln.buf);
 }
 
-static void
-pdesc(struct mparse *curp, const char *file, int fd)
-{
-       struct buf       blk;
-       int              with_mmap;
-
-       /*
-        * Run for each opened file; may be called more than once for
-        * each full parse sequence if the opened file is nested (i.e.,
-        * from `so').  Simply sucks in the whole file and moves into
-        * the parse phase for the file.
-        */
-
-       if ( ! read_whole_file(file, fd, &blk, &with_mmap)) {
-               curp->file_status = MANDOCLEVEL_SYSERR;
-               return;
-       }
-
-       /* Line number is per-file. */
-
-       curp->line = 1;
-
-       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)
 {
@@ -671,29 +643,72 @@ mparse_end(struct mparse *curp)
 }
 
 static void
-mparse_readfd_r(struct mparse *curp, int fd, const char *file, int re)
+mparse_parse_buffer(struct mparse *curp, struct buf blk, const char *file,
+               int re)
 {
        const char      *svfile;
 
+       /* Line number is per-file. */
+       svfile = curp->file;
+       curp->file = file;
+       curp->line = 1;
+
+       mparse_buf_r(curp, blk, 1);
+
+       if (0 == re && MANDOCLEVEL_FATAL > curp->file_status)
+               mparse_end(curp);
+
+       curp->file = svfile;
+}
+
+enum mandoclevel
+mparse_readmem(struct mparse *curp, const void *buf, size_t len,
+               const char *file)
+{
+       struct buf blk;
+
+       blk.buf = UNCONST(buf);
+       blk.sz = len;
+
+       mparse_parse_buffer(curp, blk, file, 0);
+       return(curp->file_status);
+}
+
+static void
+mparse_readfd_r(struct mparse *curp, int fd, const char *file, int re)
+{
+       struct buf       blk;
+       int              with_mmap;
+
        if (-1 == fd)
                if (-1 == (fd = open(file, O_RDONLY, 0))) {
                        perror(file);
                        curp->file_status = MANDOCLEVEL_SYSERR;
                        return;
                }
+       /*
+        * Run for each opened file; may be called more than once for
+        * each full parse sequence if the opened file is nested (i.e.,
+        * from `so').  Simply sucks in the whole file and moves into
+        * the parse phase for the file.
+        */
 
-       svfile = curp->file;
-       curp->file = file;
+       if ( ! read_whole_file(file, fd, &blk, &with_mmap)) {
+               curp->file_status = MANDOCLEVEL_SYSERR;
+               return;
+       }
 
-       pdesc(curp, file, fd);
+       mparse_parse_buffer(curp, blk, file, re);
 
-       if (0 == re && MANDOCLEVEL_FATAL > curp->file_status)
-               mparse_end(curp);
+#ifdef HAVE_MMAP
+       if (with_mmap)
+               munmap(blk.buf, blk.sz);
+       else
+#endif
+               free(blk.buf);
 
        if (STDIN_FILENO != fd && -1 == close(fd))
                perror(file);
-
-       curp->file = svfile;
 }
 
 enum mandoclevel