mdocml: When parsing catpages, read from the first section (NAME, we

mdocml: When parsing catpages, read from the first section (NAME, we

[kristaps]

Log Message:
-----------
When parsing catpages, read from the first section (NAME, we hope) until
the next section.  Also, remove the limit of 72 characters and enforce
this, instead in the apropos frontend.

Modified Files:
--------------
    mdocml:
        apropos.c
        mandocdb.c

Revision Data
-------------
Index: mandocdb.c
===================================================================
RCS file: /usr/vhosts/mdocml.bsd.lv/cvs/mdocml/mandocdb.c,v
retrieving revision 1.42
retrieving revision 1.43
diff -Lmandocdb.c -Lmandocdb.c -u -p -r1.42 -r1.43
--- mandocdb.c
+++ mandocdb.c
@@ -23,6 +23,7 @@
 #include <sys/types.h>
 
 #include <assert.h>
+#include <ctype.h>
 #include <dirent.h>
 #include <fcntl.h>
 #include <getopt.h>
@@ -129,8 +130,8 @@ static	void		  ofile_argbuild(int, char 
 static	void		  ofile_dirbuild(const char *, const char *,
 				const char *, int, struct of **);
 static	void		  ofile_free(struct of *);
-static	void		  pformatted(DB *, struct buf *, struct buf *,
-				const struct of *);
+static	void		  pformatted(DB *, struct buf *, 
+				struct buf *, const struct of *);
 static	int		  pman_node(MAN_ARGS);
 static	void		  pmdoc_node(MDOC_ARGS);
 static	int		  pmdoc_head(MDOC_ARGS);
@@ -1334,6 +1335,8 @@ pman_node(MAN_ARGS)
 
 			if (0 == strncmp(start, "-", 1))
 				start += 1;
+			else if (0 == strncmp(start, "\\-\\-", 4))
+				start += 4;
 			else if (0 == strncmp(start, "\\-", 2))
 				start += 2;
 			else if (0 == strncmp(start, "\\(en", 4))
@@ -1364,12 +1367,12 @@ pman_node(MAN_ARGS)
  * By necessity, this involves rather crude guesswork.
  */
 static void
-pformatted(DB *hash, struct buf *buf, struct buf *dbuf,
-		 const struct of *of)
+pformatted(DB *hash, struct buf *buf, 
+		struct buf *dbuf, const struct of *of)
 {
 	FILE		*stream;
-	char		*line, *p;
-	size_t		 len, plen;
+	char		*line, *p, *title;
+	size_t		 len, plen, titlesz;
 
 	if (NULL == (stream = fopen(of->fname, "r"))) {
 		if (warnings)
@@ -1402,6 +1405,32 @@ pformatted(DB *hash, struct buf *buf, st
 	while (NULL != (line = fgetln(stream, &len)))
 		if ('\n' != *line && ' ' != *line)
 			break;
+	
+	/*
+	 * Read up until the next section into a buffer.
+	 * Strip the leading and trailing newline from each read line,
+	 * appending a trailing space.
+	 * Ignore empty (whitespace-only) lines.
+	 */
+
+	titlesz = 0;
+	title = NULL;
+
+	while (NULL != (line = fgetln(stream, &len))) {
+		if (' ' != *line || '\n' != line[(int)len - 1])
+			break;
+		while (len > 0 && isspace((unsigned char)*line)) {
+			line++;
+			len--;
+		}
+		if (1 == len)
+			continue;
+		title = mandoc_realloc(title, titlesz + len);
+		memcpy(title + titlesz, line, len);
+		titlesz += len;
+		title[(int)titlesz - 1] = ' ';
+	}
+
 
 	/*
 	 * If no page content can be found, or the input line
@@ -1410,18 +1439,19 @@ pformatted(DB *hash, struct buf *buf, st
 	 * description.
 	 */
 
-	line = fgetln(stream, &len);
-	if (NULL == line || ' ' != *line || '\n' != line[(int)len - 1]) {
+	if (NULL == title || '\0' == *title) {
 		if (warnings)
 			fprintf(stderr, "%s: cannot find NAME section\n",
 					of->fname);
 		buf_appendb(dbuf, buf->cp, buf->size);
 		hash_put(hash, buf, TYPE_Nd);
 		fclose(stream);
+		free(title);
 		return;
 	}
 
-	line[(int)--len] = '\0';
+	title = mandoc_realloc(title, titlesz + 1);
+	title[(int)titlesz] = '\0';
 
 	/*
 	 * Skip to the first dash.
@@ -1429,20 +1459,17 @@ pformatted(DB *hash, struct buf *buf, st
 	 * bytes).
 	 */
 
-	if (NULL != (p = strstr(line, "- "))) {
+	if (NULL != (p = strstr(title, "- "))) {
 		for (p += 2; ' ' == *p || '\b' == *p; p++)
 			/* Skip to next word. */ ;
 	} else {
 		if (warnings)
 			fprintf(stderr, "%s: no dash in title line\n",
 					of->fname);
-		p = line;
+		p = title;
 	}
 
-	if ((plen = strlen(p)) > 70) {
-		plen = 70;
-		p[plen] = '\0';
-	}
+	plen = strlen(p);
 
 	/* Strip backspace-encoding from line. */
 
@@ -1461,6 +1488,7 @@ pformatted(DB *hash, struct buf *buf, st
 	buf_appendb(buf, p, plen + 1);
 	hash_put(hash, buf, TYPE_Nd);
 	fclose(stream);
+	free(title);
 }
 
 static void
Index: apropos.c
===================================================================
RCS file: /usr/vhosts/mdocml.bsd.lv/cvs/mdocml/apropos.c,v
retrieving revision 1.24
retrieving revision 1.25
diff -Lapropos.c -Lapropos.c -u -p -r1.24 -r1.25
--- apropos.c
+++ apropos.c
@@ -127,7 +127,7 @@ list(struct res *res, size_t sz, void *a
 	qsort(res, sz, sizeof(struct res), cmp);
 
 	for (i = 0; i < (int)sz; i++)
-		printf("%s(%s%s%s) - %s\n", res[i].title,
+		printf("%s(%s%s%s) - %.70s\n", res[i].title,
 				res[i].cat,
 				*res[i].arch ? "/" : "",
 				*res[i].arch ? res[i].arch : "",
--
 To unsubscribe send an email to source+unsubscribe@mdocml.bsd.lv