Support for maxrmargin-breaking with TERMP_NOBREAK.

[mandoc.git] / mdocterm.c

diff --git a/mdocterm.c b/mdocterm.c
index 5e5e751d571f8bbef92df77cf431c82e4301de23..7a0e1230af49bd0223eb1b85e09061e9f712bb5e 100644 (file)
--- a/mdocterm.c
+++ b/mdocterm.c
@@ -1,4 +1,4 @@
-/* $Id: mdocterm.c,v 1.16 2009/02/27 08:20:15 kristaps Exp $ */
+/* $Id: mdocterm.c,v 1.24 2009/03/02 12:09:32 kristaps Exp $ */

 /*
  * Copyright (c) 2008 Kristaps Dzonsons <kristaps@kth.se>
  *
 /*
  * Copyright (c) 2008 Kristaps Dzonsons <kristaps@kth.se>
  *


@@ -16,6 +16,8 @@
  * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
  * PERFORMANCE OF THIS SOFTWARE.
  */
  * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
  * PERFORMANCE OF THIS SOFTWARE.
  */

+#include <sys/utsname.h>
+

 #include <assert.h>
 #include <ctype.h>
 #include <err.h>
 #include <assert.h>
 #include <ctype.h>
 #include <err.h>


@@ -36,6 +38,7 @@
 #define        TERMSYM_LARROW          "<-"
 #define        TERMSYM_RARROW          "->"
 #define        TERMSYM_UARROW          "^"
 #define        TERMSYM_LARROW          "<-"
 #define        TERMSYM_RARROW          "->"
 #define        TERMSYM_UARROW          "^"

+#define        TERMSYM_DARROW          "v"

 #define        TERMSYM_LSQUOTE         "`"
 #define        TERMSYM_RSQUOTE         "\'"
 #define        TERMSYM_SQUOTE          "\'"
 #define        TERMSYM_LSQUOTE         "`"
 #define        TERMSYM_RSQUOTE         "\'"
 #define        TERMSYM_SQUOTE          "\'"


@@ -52,7 +55,8 @@
 #define        TERMSYM_GRAVE           "`"
 #define        TERMSYM_PI              "pi"
 #define        TERMSYM_PLUSMINUS       "+="
 #define        TERMSYM_GRAVE           "`"
 #define        TERMSYM_PI              "pi"
 #define        TERMSYM_PLUSMINUS       "+="

-#define        TERMSYM_INFINITY        "infinity"
+#define        TERMSYM_INF             "oo"
+#define        TERMSYM_INF2            "infinity"

 #define        TERMSYM_NAN             "NaN"
 #define        TERMSYM_BAR             "|"
 #define        TERMSYM_BULLET          "o"
 #define        TERMSYM_NAN             "NaN"
 #define        TERMSYM_BAR             "|"
 #define        TERMSYM_BULLET          "o"


@@ -81,6 +85,7 @@ static        void              footer(struct termp *,
 static void              pword(struct termp *, const char *, size_t);
 static void              pescape(struct termp *, 
                                const char *, size_t *, size_t);
 static void              pword(struct termp *, const char *, size_t);
 static void              pescape(struct termp *, 
                                const char *, size_t *, size_t);

+static void              pgraph(struct termp *, char);

 static void              nescape(struct termp *, 
                                const char *, size_t);
 static void              chara(struct termp *, char);
 static void              nescape(struct termp *, 
                                const char *, size_t);
 static void              chara(struct termp *, char);


@@ -108,7 +113,7 @@ main(int argc, char *argv[])
        if (NULL == (mdoc = mmain_mdoc(p)))
                mmain_exit(p, 1);
 
        if (NULL == (mdoc = mmain_mdoc(p)))
                mmain_exit(p, 1);
 

-       termp.maxrmargin = 80; /* XXX */
+       termp.maxrmargin = 78; /* XXX */

        termp.rmargin = termp.maxrmargin;
        termp.maxcols = 1024;
        termp.offset = termp.col = 0;
        termp.rmargin = termp.maxrmargin;
        termp.maxcols = 1024;
        termp.offset = termp.col = 0;


@@ -169,9 +174,10 @@ flushln(struct termp *p)
        for (i = 0; i < p->col; i++) {
                /*
                 * Count up visible word characters.  Control sequences
        for (i = 0; i < p->col; i++) {
                /*
                 * Count up visible word characters.  Control sequences

-                * (starting with the CSI) aren't counted. 
+                * (starting with the CSI) aren't counted.  A space
+                * generates a non-printing word, which is valid (the
+                * space is printed according to regular spacing rules).

                 */
                 */

-               assert( ! xisspace(p->buf[i]));

 
                /* LINTED */
                for (j = i, vsz = 0; j < p->col; j++) {
 
                /* LINTED */
                for (j = i, vsz = 0; j < p->col; j++) {


@@ -183,27 +189,35 @@ flushln(struct termp *p)
                        } else
                                vsz++;
                }
                        } else
                                vsz++;
                }

-               assert(vsz > 0);

 
                /*
 
                /*

+                * If we're breaking normally...
+                *

                 * If a word is too long and we're within a line, put it
                 * on the next line.  Puke if we're being asked to write
                 * something that will exceed the right margin (i.e.,
                 * If a word is too long and we're within a line, put it
                 * on the next line.  Puke if we're being asked to write
                 * something that will exceed the right margin (i.e.,

-                * from a fresh line or when we're not allowed to break
-                * the line with TERMP_NOBREAK).
+                * from a fresh line).
+                *
+                * If we're not breaking...
+                *
+                * Don't let the visible size exceed the full right
+                * margin.

                 */
 
                 */
 

-               /* FIXME: allow selective right-margin breaking. */
-
-               if (vis && vis + vsz > maxvis) {
-                       if (p->flags & TERMP_NOBREAK)
+               if ( ! (TERMP_NOBREAK & p->flags)) {
+                       if (vis && vis + vsz > maxvis) {
+                               putchar('\n');
+                               for (j = 0; j < p->offset; j++)
+                                       putchar(' ');
+                               vis = 0;
+                       } else if (vis + vsz > maxvis)

                                errx(1, "word breaks right margin");
                                errx(1, "word breaks right margin");

+               } else if (vis + vsz > p->maxrmargin - p->offset) {

                        putchar('\n');
                        putchar('\n');

-                       for (j = 0; j < p->offset; j++)
+                       for (j = 0; j < p->rmargin; j++)

                                putchar(' ');
                                putchar(' ');

-                       vis = 0;
-               } else if (vis + vsz > maxvis)
-                       errx(1, "word breaks right margin");
+                       vis = p->rmargin;
+               }

 
                /* 
                 * Write out the word and a trailing space.  Omit the
 
                /* 
                 * Write out the word and a trailing space.  Omit the


@@ -222,15 +236,22 @@ flushln(struct termp *p)
                }
        }
 
                }
        }
 

+       if ((TERMP_NOBREAK & p->flags) && vis >= maxvis) {
+               putchar('\n');
+               for (i = 0; i < p->rmargin; i++)
+                       putchar(' ');
+               p->col = 0;
+               return;
+       }
+

        /*
         * If we're not to right-marginalise it (newline), then instead
         * pad to the right margin and stay off.
         */
 
        if (p->flags & TERMP_NOBREAK) {
        /*
         * If we're not to right-marginalise it (newline), then instead
         * pad to the right margin and stay off.
         */
 
        if (p->flags & TERMP_NOBREAK) {

-               if ( ! (p->flags & TERMP_NORPAD))
-                       for ( ; vis < maxvis; vis++)
-                               putchar(' ');
+               for ( ; vis < maxvis; vis++)
+                       putchar(' ');

        } else
                putchar('\n');
 
        } else
                putchar('\n');
 


@@ -338,50 +359,70 @@ nescape(struct termp *p, const char *word, size_t len)
                        stringa(p, TERMSYM_RBRACK);
                else if ('l' == word[0] && 'B' == word[1])
                        stringa(p, TERMSYM_LBRACK);
                        stringa(p, TERMSYM_RBRACK);
                else if ('l' == word[0] && 'B' == word[1])
                        stringa(p, TERMSYM_LBRACK);

+               else if ('l' == word[0] && 'q' == word[1])
+                       stringa(p, TERMSYM_LDQUOTE);
+               else if ('r' == word[0] && 'q' == word[1])
+                       stringa(p, TERMSYM_RDQUOTE);
+               else if ('o' == word[0] && 'q' == word[1])
+                       stringa(p, TERMSYM_LSQUOTE);
+               else if ('a' == word[0] && 'q' == word[1])
+                       stringa(p, TERMSYM_RSQUOTE);

                else if ('<' == word[0] && '-' == word[1])
                        stringa(p, TERMSYM_LARROW);
                else if ('-' == word[0] && '>' == word[1])
                        stringa(p, TERMSYM_RARROW);
                else if ('<' == word[0] && '-' == word[1])
                        stringa(p, TERMSYM_LARROW);
                else if ('-' == word[0] && '>' == word[1])
                        stringa(p, TERMSYM_RARROW);

-               else if ('l' == word[0] && 'q' == word[1])
-                       stringa(p, TERMSYM_DQUOTE);
-               else if ('r' == word[0] && 'q' == word[1])
-                       stringa(p, TERMSYM_DQUOTE);

                else if ('b' == word[0] && 'u' == word[1])
                        stringa(p, TERMSYM_BULLET);
                else if ('b' == word[0] && 'u' == word[1])
                        stringa(p, TERMSYM_BULLET);

-               else if ('L' == word[0] && 'e' == word[1])
-                       stringa(p, TERMSYM_LE);

                else if ('<' == word[0] && '=' == word[1])
                        stringa(p, TERMSYM_LE);
                else if ('<' == word[0] && '=' == word[1])
                        stringa(p, TERMSYM_LE);

-               else if ('G' == word[0] && 'e' == word[1])
-                       stringa(p, TERMSYM_GE);

                else if ('>' == word[0] && '=' == word[1])
                        stringa(p, TERMSYM_GE);
                else if ('>' == word[0] && '=' == word[1])
                        stringa(p, TERMSYM_GE);

-               else if ('R' == word[0] && 'q' == word[1])
-                       stringa(p, TERMSYM_RDQUOTE);
-               else if ('L' == word[0] && 'q' == word[1])
-                       stringa(p, TERMSYM_LDQUOTE);
+               else if ('=' == word[0] && '=' == word[1])
+                       stringa(p, TERMSYM_EQ);
+               else if ('+' == word[0] && '-' == word[1])
+                       stringa(p, TERMSYM_PLUSMINUS);

                else if ('u' == word[0] && 'a' == word[1])
                        stringa(p, TERMSYM_UARROW);
                else if ('u' == word[0] && 'a' == word[1])
                        stringa(p, TERMSYM_UARROW);

+               else if ('d' == word[0] && 'a' == word[1])
+                       stringa(p, TERMSYM_DARROW);

                else if ('a' == word[0] && 'a' == word[1])
                        stringa(p, TERMSYM_ACUTE);
                else if ('g' == word[0] && 'a' == word[1])
                        stringa(p, TERMSYM_GRAVE);
                else if ('a' == word[0] && 'a' == word[1])
                        stringa(p, TERMSYM_ACUTE);
                else if ('g' == word[0] && 'a' == word[1])
                        stringa(p, TERMSYM_GRAVE);

-               else if ('P' == word[0] && 'i' == word[1])
-                       stringa(p, TERMSYM_PI);
-               else if ('N' == word[0] && 'e' == word[1])
+               else if ('!' == word[0] && '=' == word[1])

                        stringa(p, TERMSYM_NEQ);
                        stringa(p, TERMSYM_NEQ);

-               else if ('L' == word[0] && 't' == word[1])
-                       stringa(p, TERMSYM_LT);
+               else if ('i' == word[0] && 'f' == word[1])
+                       stringa(p, TERMSYM_INF);
+               else if ('n' == word[0] && 'a' == word[1])
+                       stringa(p, TERMSYM_NAN);
+               else if ('b' == word[0] && 'a' == word[1])
+                       stringa(p, TERMSYM_BAR);
+
+               /* Deprecated forms. */
+               else if ('B' == word[0] && 'a' == word[1])
+                       stringa(p, TERMSYM_BAR);
+               else if ('I' == word[0] && 'f' == word[1])
+                       stringa(p, TERMSYM_INF2);
+               else if ('G' == word[0] && 'e' == word[1])
+                       stringa(p, TERMSYM_GE);

                else if ('G' == word[0] && 't' == word[1])
                        stringa(p, TERMSYM_GT);
                else if ('G' == word[0] && 't' == word[1])
                        stringa(p, TERMSYM_GT);

-               else if ('P' == word[0] && 'm' == word[1])
-                       stringa(p, TERMSYM_PLUSMINUS);
-               else if ('I' == word[0] && 'f' == word[1])
-                       stringa(p, TERMSYM_INFINITY);
+               else if ('L' == word[0] && 'e' == word[1])
+                       stringa(p, TERMSYM_LE);
+               else if ('L' == word[0] && 'q' == word[1])
+                       stringa(p, TERMSYM_LDQUOTE);
+               else if ('L' == word[0] && 't' == word[1])
+                       stringa(p, TERMSYM_LT);

                else if ('N' == word[0] && 'a' == word[1])
                        stringa(p, TERMSYM_NAN);
                else if ('N' == word[0] && 'a' == word[1])
                        stringa(p, TERMSYM_NAN);

-               else if ('B' == word[0] && 'a' == word[1])
-                       stringa(p, TERMSYM_BAR);
+               else if ('N' == word[0] && 'e' == word[1])
+                       stringa(p, TERMSYM_NEQ);
+               else if ('P' == word[0] && 'i' == word[1])
+                       stringa(p, TERMSYM_PI);
+               else if ('P' == word[0] && 'm' == word[1])
+                       stringa(p, TERMSYM_PLUSMINUS);
+               else if ('R' == word[0] && 'q' == word[1])
+                       stringa(p, TERMSYM_RDQUOTE);

                break;
        default:
                break;
                break;
        default:
                break;


@@ -389,6 +430,26 @@ nescape(struct termp *p, const char *word, size_t len)
 }
 
 
 }
 
 

+static void
+pgraph(struct termp *p, char byte)
+{
+       int              i;
+
+       switch (byte) {
+       case (' '):
+               chara(p, ' ');
+               break;
+       case ('\t'):
+               for (i = 0; i < INDENT; i++)
+                       chara(p, ' ');
+               break;
+       default:
+               warnx("unknown non-graphing character");
+               break;
+       }
+}
+
+

 static void
 pescape(struct termp *p, const char *word, size_t *i, size_t len)
 {
 static void
 pescape(struct termp *p, const char *word, size_t *i, size_t len)
 {


@@ -441,6 +502,10 @@ pescape(struct termp *p, const char *word, size_t *i, size_t len)
                        /* FALLTHROUGH */
                case ('.'):
                        chara(p, word[*i]);
                        /* FALLTHROUGH */
                case ('.'):
                        chara(p, word[*i]);

+                       break;
+               case ('e'):
+                       chara(p, '\\');
+                       break;

                default:
                        break;
                }
                default:
                        break;
                }


@@ -460,8 +525,6 @@ pword(struct termp *p, const char *word, size_t len)
 {
        size_t           i;
 
 {
        size_t           i;
 

-       /*assert(len > 0);*/ /* Can be, if literal. */
-

        /*
         * Handle pwords, partial words, which may be either a single
         * word or a phrase that cannot be broken down (such as a
        /*
         * Handle pwords, partial words, which may be either a single
         * word or a phrase that cannot be broken down (such as a


@@ -490,6 +553,10 @@ pword(struct termp *p, const char *word, size_t len)
                        pescape(p, word, &i, len);
                        continue;
                }
                        pescape(p, word, &i, len);
                        continue;
                }

+               if ( ! isgraph((int)word[i])) {
+                       pgraph(p, word[i]);
+                       continue;
+               }

                chara(p, word[i]);
        }
 
                chara(p, word[i]);
        }
 


@@ -530,7 +597,14 @@ word(struct termp *p, const char *word)
                if ( ! xisspace(word[i])) {
                        j++;
                        continue;
                if ( ! xisspace(word[i])) {
                        j++;
                        continue;

+               } 
+               
+               /* Escaped spaces don't delimit... */
+               if (i > 0 && xisspace(word[i]) && '\\' == word[i - 1]) {
+                       j++;
+                       continue;

                }
                }

+

                if (0 == j)
                        continue;
                assert(i >= j);
                if (0 == j)
                        continue;
                assert(i >= j);


@@ -652,15 +726,21 @@ footer(struct termp *p, const struct mdoc_meta *meta)
 static void
 header(struct termp *p, const struct mdoc_meta *meta)
 {
 static void
 header(struct termp *p, const struct mdoc_meta *meta)
 {

-       char            *buf, *title, *bufp;
+       char            *buf, *title, *bufp, *vbuf;

        const char      *pp;
        const char      *pp;

+       struct utsname   uts;
+
+       p->rmargin = p->maxrmargin;
+       p->offset = 0;

 
        if (NULL == (buf = malloc(p->rmargin)))
                err(1, "malloc");
        if (NULL == (title = malloc(p->rmargin)))
                err(1, "malloc");
 
        if (NULL == (buf = malloc(p->rmargin)))
                err(1, "malloc");
        if (NULL == (title = malloc(p->rmargin)))
                err(1, "malloc");

+       if (NULL == (vbuf = malloc(p->rmargin)))
+               err(1, "malloc");

 
 

-       if (NULL == (pp = mdoc_vol2a(meta->vol)))
+       if (NULL == (pp = mdoc_vol2a(meta->vol))) {

                switch (meta->msec) {
                case (MSEC_1):
                        /* FALLTHROUGH */
                switch (meta->msec) {
                case (MSEC_1):
                        /* FALLTHROUGH */


@@ -685,11 +765,20 @@ header(struct termp *p, const struct mdoc_meta *meta)
                        pp = mdoc_vol2a(VOL_KM);
                        break;
                default:
                        pp = mdoc_vol2a(VOL_KM);
                        break;
                default:

-                       /* FIXME: capitalise. */
-                       if (NULL == (pp = mdoc_msec2a(meta->msec)))
-                               pp = mdoc_msec2a(MSEC_local);

                        break;
                }
                        break;
                }

+       }
+       vbuf[0] = 0;
+
+       if (pp) {
+               if (-1 == uname(&uts)) 
+                       err(1, "uname");
+               (void)strlcat(vbuf, uts.sysname, p->rmargin);
+               (void)strlcat(vbuf, " ", p->rmargin);
+       } else if (NULL == (pp = mdoc_msec2a(meta->msec)))
+               pp = mdoc_msec2a(MSEC_local);
+
+       (void)strlcat(vbuf, pp, p->rmargin);

 
        /*
         * The header is strange.  It has three components, which are
 
        /*
         * The header is strange.  It has three components, which are


@@ -706,9 +795,9 @@ header(struct termp *p, const struct mdoc_meta *meta)
 
        if (mdoc_arch2a(meta->arch))
                (void)snprintf(buf, p->rmargin, "%s (%s)",
 
        if (mdoc_arch2a(meta->arch))
                (void)snprintf(buf, p->rmargin, "%s (%s)",

-                               pp, mdoc_arch2a(meta->arch));
+                               vbuf, mdoc_arch2a(meta->arch));

        else
        else

-               (void)strlcpy(buf, pp, p->rmargin);
+               (void)strlcpy(buf, vbuf, p->rmargin);

 
        pp = mdoc_msec2a(meta->msec);
 
 
        pp = mdoc_msec2a(meta->msec);
 


@@ -745,5 +834,6 @@ header(struct termp *p, const struct mdoc_meta *meta)
        p->flags &= ~TERMP_NOSPACE;
 
        free(title);
        p->flags &= ~TERMP_NOSPACE;
 
        free(title);

+       free(vbuf);

        free(buf);
 }
        free(buf);
 }