=================================================================== RCS file: /cvs/pod2mdoc/pod2mdoc.c,v retrieving revision 1.10 retrieving revision 1.12 diff -u -p -r1.10 -r1.12 --- pod2mdoc/pod2mdoc.c 2014/03/24 01:43:42 1.10 +++ pod2mdoc/pod2mdoc.c 2014/04/01 11:58:32 1.12 @@ -1,4 +1,4 @@ -/* $Id: pod2mdoc.c,v 1.10 2014/03/24 01:43:42 kristaps Exp $ */ +/* $Id: pod2mdoc.c,v 1.12 2014/04/01 11:58:32 kristaps Exp $ */ /* * Copyright (c) 2014 Kristaps Dzonsons * @@ -44,11 +44,17 @@ enum list { LIST__MAX }; +enum sect { + SECT_NONE = 0, + SECT_NAME, /* NAME section */ + SECT_SYNOPSIS, /* SYNOPSIS section */ +}; + struct state { int parsing; /* after =cut of before command */ int paused; /* in =begin and before =end */ int haspar; /* in paragraph: do we need Pp? */ - int isname; /* are we the NAME section? */ + enum sect sect; /* which section are we in? */ const char *fname; /* file being parsed */ #define LIST_STACKSZ 128 enum list lstack[LIST_STACKSZ]; /* open lists */ @@ -247,8 +253,8 @@ trylink(const char *buf, size_t *start, size_t end, si * reentrant). */ static int -formatcode(const char *buf, size_t *start, - size_t end, int reentrant, int nomacro) +formatcode(struct state *st, const char *buf, + size_t *start, size_t end, int reentrant, int nomacro) { enum fmt fmt; size_t i, j, dsz; @@ -354,6 +360,28 @@ formatcode(const char *buf, size_t *start, printf("Em "); break; case (FMT_BOLD): + /* + * Doclifting: if we're a bold "-xx" and we're + * in the SYNOPSIS section, then it's likely + * that we're a flag. + * Be really strict: only do this when the dash + * is followed by alnums til the end marker, + * which mustn't be a custom. + */ + if (SECT_SYNOPSIS == st->sect && + end - *start > 1 && + '-' == buf[*start] && + (isalnum((int)buf[*start + 1]) || + '?' == buf[*start + 1])) { + for (i = *start + 1; i < end; i++) + if ( ! isalnum((int)buf[i])) + break; + if (i < end && '>' == buf[i]) { + (*start)++; + printf("Fl "); + break; + } + } printf("Sy "); break; case (FMT_CODE): @@ -402,7 +430,7 @@ formatcode(const char *buf, size_t *start, } } if (*start + 1 < end && '<' == buf[*start + 1]) { - formatcode(buf, start, end, 1, nomacro); + formatcode(st, buf, start, end, 1, nomacro); continue; } @@ -457,13 +485,14 @@ formatcode(const char *buf, size_t *start, * Calls formatcode() til the end of a paragraph. */ static void -formatcodeln(const char *buf, size_t *start, size_t end, int nomacro) +formatcodeln(struct state *st, const char *buf, + size_t *start, size_t end, int nomacro) { last = ' '; while (*start < end) { if (*start + 1 < end && '<' == buf[*start + 1]) { - formatcode(buf, start, end, 1, nomacro); + formatcode(st, buf, start, end, 1, nomacro); continue; } /* @@ -563,24 +592,28 @@ command(struct state *st, const char *buf, size_t star * how pod2man handles it. */ printf(".Sh "); - st->isname = 0; - if (end - start == 4) + st->sect = SECT_NONE; + if (end - start == 4) { if (0 == memcmp(&buf[start], "NAME", 4)) - st->isname = 1; - formatcodeln(buf, &start, end, 1); + st->sect = SECT_NAME; + } else if (end - start == 8) { + if (0 == memcmp(&buf[start], "SYNOPSIS", 8)) + st->sect = SECT_SYNOPSIS; + } + formatcodeln(st, buf, &start, end, 1); putchar('\n'); st->haspar = 1; break; case (CMD_HEAD2): printf(".Ss "); - formatcodeln(buf, &start, end, 1); + formatcodeln(st, buf, &start, end, 1); putchar('\n'); st->haspar = 1; break; case (CMD_HEAD3): puts(".Pp"); printf(".Em "); - formatcodeln(buf, &start, end, 0); + formatcodeln(st, buf, &start, end, 0); putchar('\n'); puts(".Pp"); st->haspar = 1; @@ -588,7 +621,7 @@ command(struct state *st, const char *buf, size_t star case (CMD_HEAD4): puts(".Pp"); printf(".No "); - formatcodeln(buf, &start, end, 0); + formatcodeln(st, buf, &start, end, 0); putchar('\n'); puts(".Pp"); st->haspar = 1; @@ -642,7 +675,7 @@ command(struct state *st, const char *buf, size_t star switch (st->lstack[st->lpos - 1]) { case (LIST_TAG): printf(".It "); - formatcodeln(buf, &start, end, 0); + formatcodeln(st, buf, &start, end, 0); putchar('\n'); break; case (LIST_ENUM): @@ -743,7 +776,7 @@ ordinary(struct state *st, const char *buf, size_t sta * we're in "name - description" format. * To wit, print out a "Nm" and "Nd" in that format. */ - if (st->isname) { + if (SECT_NAME == st->sect) { for (i = end - 1; i > start; i--) if ('-' == buf[i]) break; @@ -754,11 +787,11 @@ ordinary(struct state *st, const char *buf, size_t sta if ('-' != buf[i]) break; printf(".Nm "); - formatcodeln(buf, &start, i + 1, 1); + formatcodeln(st, buf, &start, i + 1, 1); putchar('\n'); start = j + 1; printf(".Nd "); - formatcodeln(buf, &start, end, 1); + formatcodeln(st, buf, &start, end, 1); putchar('\n'); return; } @@ -784,6 +817,26 @@ ordinary(struct state *st, const char *buf, size_t sta printf("\\&"); else if ('\n' == last && '\'' == buf[start]) printf("\\&"); +#if notyet + /* + * If we're in the SYNOPSIS, have square + * brackets indicate that we're opening and + * closing an optional context. + */ + if (SECT_SYNOPSIS == st->sect) { + if ('[' == buf[start] || + ']' == buf[start]) { + if (last != '\n') + putchar('\n'); + if ('[' == buf[start]) + printf(".Oo\n"); + else + printf(".Oc\n"); + start++; + continue; + } + } +#endif putchar(last = buf[start++]); if ('\\' == last) putchar('e'); @@ -799,7 +852,7 @@ ordinary(struct state *st, const char *buf, size_t sta * Consume all whitespace so we don't * accidentally start an implicit literal line. */ - if (formatcode(buf, &start, end, 0, 0)) { + if (formatcode(st, buf, &start, end, 0, 0)) { putchar(last = '\n'); while (start < end && ' ' == buf[start]) start++; @@ -828,19 +881,7 @@ ordinary(struct state *st, const char *buf, size_t sta for ( ; start < end; start++) if (' ' != buf[start] && '\t' != buf[start]) break; - } else if (start < end) { - /* - * Default: print the character. - * Escape initial control characters. - */ - if ('\n' == last && '.' == buf[start]) - printf("\\&"); - else if ('\n' == last && '\'' == buf[start]) - printf("\\&"); - putchar(last = buf[start++]); - if ('\\' == last) - putchar('e'); - } + } } if (last != '\n')