Merging patch by Ingo Schwarze.

[mandoc.git] / mdoc.c

diff --git a/mdoc.c b/mdoc.c
index 558782e3efb6b2bbca58dc93715404423d7a43bb..7e1d0f4b7a0ed1eafe4af7806c3ad1146c1a5797 100644 (file)
--- a/mdoc.c
+++ b/mdoc.c
@@ -1,4 +1,4 @@
-/*     $Id: mdoc.c,v 1.113 2009/10/30 05:58:38 kristaps Exp $ */
+/*     $Id: mdoc.c,v 1.119 2010/04/03 13:02:35 kristaps Exp $ */
 /*
  * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -14,6 +14,10 @@
  * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
  */
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
 #include <sys/types.h>
 
 #include <assert.h>
@@ -93,11 +97,11 @@ const       char *const __mdoc_macronames[MDOC_MAX] = {
        "Nm",           "Op",           "Ot",           "Pa",
        "Rv",           "St",           "Va",           "Vt",
        /* LINTED */
-       "Xr",           "\%A",          "\%B",          "\%D",
+       "Xr",           "%A",           "%B",           "%D",
        /* LINTED */
-       "\%I",          "\%J",          "\%N",          "\%O",
+       "%I",           "%J",           "%N",           "%O",
        /* LINTED */
-       "\%P",          "\%R",          "\%T",          "\%V",
+       "%P",           "%R",           "%T",           "%V",
        "Ac",           "Ao",           "Aq",           "At",
        "Bc",           "Bf",           "Bo",           "Bq",
        "Bsx",          "Bx",           "Db",           "Dc",
@@ -114,11 +118,11 @@ const     char *const __mdoc_macronames[MDOC_MAX] = {
        "Fr",           "Ud",           "Lb",           "Lp",
        "Lk",           "Mt",           "Brq",          "Bro",
        /* LINTED */
-       "Brc",          "\%C",          "Es",           "En",
+       "Brc",          "%C",           "Es",           "En",
        /* LINTED */
-       "Dx",           "\%Q",          "br",           "sp",
+       "Dx",           "%Q",           "br",           "sp",
        /* LINTED */
-       "\%U"
+       "%U"
        };
 
 const  char *const __mdoc_argnames[MDOC_ARG_MAX] = {            
@@ -139,7 +143,7 @@ const       char * const *mdoc_argnames = __mdoc_argnames;
 static void              mdoc_free1(struct mdoc *);
 static void              mdoc_alloc1(struct mdoc *);
 static struct mdoc_node *node_alloc(struct mdoc *, int, int, 
-                               int, enum mdoc_type);
+                               enum mdoct, enum mdoc_type);
 static int               node_append(struct mdoc *, 
                                struct mdoc_node *);
 static int               parsetext(struct mdoc *, int, char *);
@@ -148,11 +152,6 @@ static     int               macrowarn(struct mdoc *, int, const char *);
 static int               pstring(struct mdoc *, int, int, 
                                const char *, size_t);
 
-#ifdef __linux__
-extern size_t            strlcpy(char *, const char *, size_t);
-#endif
-
-
 const struct mdoc_node *
 mdoc_node(const struct mdoc *m)
 {
@@ -342,9 +341,11 @@ mdoc_err(struct mdoc *m, int line, int pos, int iserr, enum merr type)
 
 
 int
-mdoc_macro(struct mdoc *m, int tok, 
+mdoc_macro(struct mdoc *m, enum mdoct tok, 
                int ln, int pp, int *pos, char *buf)
 {
+
+       assert(tok < MDOC_MAX);
        /*
         * If we're in the prologue, deny "body" macros.  Similarly, if
         * we're in the body, deny prologue calls.
@@ -425,8 +426,8 @@ node_append(struct mdoc *mdoc, struct mdoc_node *p)
 
 
 static struct mdoc_node *
-node_alloc(struct mdoc *m, int line, 
-               int pos, int tok, enum mdoc_type type)
+node_alloc(struct mdoc *m, int line, int pos, 
+               enum mdoct tok, enum mdoc_type type)
 {
        struct mdoc_node *p;
 
@@ -435,15 +436,14 @@ node_alloc(struct mdoc *m, int line,
        p->line = line;
        p->pos = pos;
        p->tok = tok;
-       if (MDOC_TEXT != (p->type = type))
-               assert(p->tok >= 0);
+       p->type = type;
 
        return(p);
 }
 
 
 int
-mdoc_tail_alloc(struct mdoc *m, int line, int pos, int tok)
+mdoc_tail_alloc(struct mdoc *m, int line, int pos, enum mdoct tok)
 {
        struct mdoc_node *p;
 
@@ -456,7 +456,7 @@ mdoc_tail_alloc(struct mdoc *m, int line, int pos, int tok)
 
 
 int
-mdoc_head_alloc(struct mdoc *m, int line, int pos, int tok)
+mdoc_head_alloc(struct mdoc *m, int line, int pos, enum mdoct tok)
 {
        struct mdoc_node *p;
 
@@ -472,7 +472,7 @@ mdoc_head_alloc(struct mdoc *m, int line, int pos, int tok)
 
 
 int
-mdoc_body_alloc(struct mdoc *m, int line, int pos, int tok)
+mdoc_body_alloc(struct mdoc *m, int line, int pos, enum mdoct tok)
 {
        struct mdoc_node *p;
 
@@ -486,7 +486,7 @@ mdoc_body_alloc(struct mdoc *m, int line, int pos, int tok)
 
 int
 mdoc_block_alloc(struct mdoc *m, int line, int pos, 
-               int tok, struct mdoc_arg *args)
+               enum mdoct tok, struct mdoc_arg *args)
 {
        struct mdoc_node *p;
 
@@ -503,7 +503,7 @@ mdoc_block_alloc(struct mdoc *m, int line, int pos,
 
 int
 mdoc_elem_alloc(struct mdoc *m, int line, int pos, 
-               int tok, struct mdoc_arg *args)
+               enum mdoct tok, struct mdoc_arg *args)
 {
        struct mdoc_node *p;
 
@@ -582,6 +582,7 @@ static int
 parsetext(struct mdoc *m, int line, char *buf)
 {
        int              i, j;
+       char             sv;
 
        if (SEC_NONE == m->lastnamed)
                return(mdoc_perr(m, line, 0, ETEXTPROL));
@@ -598,8 +599,19 @@ parsetext(struct mdoc *m, int line, char *buf)
 
        for (i = 0; ' ' == buf[i]; i++)
                /* Skip leading whitespace. */ ;
-       if (0 == buf[i])
-               return(mdoc_perr(m, line, 0, ENOBLANK));
+
+       if ('\0' == buf[i]) {
+               if ( ! mdoc_pwarn(m, line, 0, ENOBLANK))
+                       return(0);
+               /*
+                * Assume that a `Pp' should be inserted in the case of
+                * a blank line.  Technically, blank lines aren't
+                * allowed, but enough manuals assume this behaviour
+                * that we want to work around it.
+                */
+               if ( ! mdoc_elem_alloc(m, line, 0, MDOC_Pp, NULL))
+                       return(0);
+       }
 
        /*
         * Break apart a free-form line into tokens.  Spaces are
@@ -614,15 +626,30 @@ parsetext(struct mdoc *m, int line, char *buf)
                if (i && ' ' == buf[i] && '\\' == buf[i - 1])
                        continue;
 
-               buf[i++] = 0;
+               sv = buf[i];
+               buf[i++] = '\0';
+
                if ( ! pstring(m, line, j, &buf[j], (size_t)(i - j)))
                        return(0);
 
+               /* Trailing whitespace?  Check at overwritten byte. */
+
+               if (' ' == sv && '\0' == buf[i])
+                       if ( ! mdoc_pwarn(m, line, i - 1, ETAILWS))
+                               return(0);
+
                for ( ; ' ' == buf[i]; i++)
                        /* Skip trailing whitespace. */ ;
 
                j = i;
-               if (0 == buf[i])
+
+               /* Trailing whitespace? */
+
+               if (' ' == buf[i - 1] && '\0' == buf[i])
+                       if ( ! mdoc_pwarn(m, line, i - 1, ETAILWS))
+                               return(0);
+
+               if ('\0' == buf[i])
                        break;
        }
 
@@ -659,7 +686,7 @@ parsemacro(struct mdoc *m, int ln, char *buf)
 
        /* Empty lines are ignored. */
 
-       if (0 == buf[1])
+       if ('\0' == buf[1])
                return(1);
 
        i = 1;
@@ -670,14 +697,14 @@ parsemacro(struct mdoc *m, int ln, char *buf)
                i++;
                while (buf[i] && ' ' == buf[i])
                        i++;
-               if (0 == buf[i])
+               if ('\0' == buf[i])
                        return(1);
        }
 
        /* Copy the first word into a nil-terminated buffer. */
 
        for (j = 0; j < 4; j++, i++) {
-               if (0 == (mac[j] = buf[i]))
+               if ('\0' == (mac[j] = buf[i]))
                        break;
                else if (' ' == buf[i])
                        break;
@@ -708,6 +735,12 @@ parsemacro(struct mdoc *m, int ln, char *buf)
        while (buf[i] && ' ' == buf[i])
                i++;
 
+       /* Trailing whitespace? */
+
+       if ('\0' == buf[i] && ' ' == buf[i - 1])
+               if ( ! mdoc_pwarn(m, ln, i - 1, ETAILWS))
+                       goto err;
+
        /* 
         * Begin recursive parse sequence.  Since we're at the start of
         * the line, we don't need to do callable/parseable checks.