From 792720a705b86669c5d8a25a3250aa5619d8e1b1 Mon Sep 17 00:00:00 2001 From: Kristaps Dzonsons Date: Mon, 31 May 2010 10:28:04 +0000 Subject: Remove enum mdocargerr from phrase() (unused). Add `Ta' macro, which is basically a NULL case everywhere but in mdoc_macro.c, where it closes out an existing `It' body scope and opens a new one, then continues parsing as in phrase() (TODO: merge these two?). Fix where scope-breaking was silently just dying instead of printing an error. Fix where trailing `Ta' or tab weren't creating a new MDOC_BODY context. We now support arbitrarily complex `It' contents for `Bl -column'. --- mdoc.c | 2 +- mdoc.h | 1 + mdoc_action.c | 1 + mdoc_argv.c | 10 ++++- mdoc_html.c | 1 + mdoc_macro.c | 118 +++++++++++++++++++++++++++++++++++++++++++++----------- mdoc_term.c | 1 + mdoc_validate.c | 1 + 8 files changed, 111 insertions(+), 24 deletions(-) diff --git a/mdoc.c b/mdoc.c index f5373f75..8aeec090 100644 --- a/mdoc.c +++ b/mdoc.c @@ -69,7 +69,7 @@ const char *const __mdoc_macronames[MDOC_MAX] = { /* LINTED */ "Dx", "%Q", "br", "sp", /* LINTED */ - "%U" + "%U", "Ta" }; const char *const __mdoc_argnames[MDOC_ARG_MAX] = { diff --git a/mdoc.h b/mdoc.h index 8b2c9a05..9fcc4d5a 100644 --- a/mdoc.h +++ b/mdoc.h @@ -149,6 +149,7 @@ enum mdoct { MDOC_br, MDOC_sp, MDOC__U, + MDOC_Ta, MDOC_MAX }; diff --git a/mdoc_action.c b/mdoc_action.c index 3d94c3fd..68e04210 100644 --- a/mdoc_action.c +++ b/mdoc_action.c @@ -194,6 +194,7 @@ static const struct actions mdoc_actions[MDOC_MAX] = { { NULL, NULL }, /* br */ { NULL, NULL }, /* sp */ { NULL, NULL }, /* %U */ + { NULL, NULL }, /* Ta */ }; #define RSORD_MAX 14 diff --git a/mdoc_argv.c b/mdoc_argv.c index a01791aa..ac344f44 100644 --- a/mdoc_argv.c +++ b/mdoc_argv.c @@ -210,6 +210,7 @@ static int mdoc_argflags[MDOC_MAX] = { 0, /* br */ 0, /* sp */ 0, /* %U */ + 0, /* Ta */ }; @@ -347,6 +348,13 @@ mdoc_args(struct mdoc *m, int line, int *pos, if (MDOC_It != tok) return(args(m, line, pos, buf, fl, v)); + /* + * We know that we're in an `It', so it's reasonable to expect + * us to be sitting in a `Bl'. Someday this may not be the case + * (if we allow random `It's sitting out there), so provide a + * safe fall-back into the default behaviour. + */ + for (n = m->last; n; n = n->parent) if (MDOC_Bl == n->tok) break; @@ -455,7 +463,7 @@ args(struct mdoc *m, int line, int *pos, break; if (pp > *v && ' ' != *(pp - 1)) continue; - if (' ' == *(pp + 2) || 0 == *(pp + 2)) + if (' ' == *(pp + 2) || '\0' == *(pp + 2)) break; } diff --git a/mdoc_html.c b/mdoc_html.c index c80d3c73..686c408a 100644 --- a/mdoc_html.c +++ b/mdoc_html.c @@ -255,6 +255,7 @@ static const struct htmlmdoc mdocs[MDOC_MAX] = { {mdoc_sp_pre, NULL}, /* br */ {mdoc_sp_pre, NULL}, /* sp */ {mdoc__x_pre, mdoc__x_post}, /* %U */ + {NULL, NULL}, /* Ta */ }; diff --git a/mdoc_macro.c b/mdoc_macro.c index 205512d6..20be5acb 100644 --- a/mdoc_macro.c +++ b/mdoc_macro.c @@ -44,13 +44,13 @@ static int in_line_eoln(MACRO_PROT_ARGS); static int in_line_argn(MACRO_PROT_ARGS); static int in_line(MACRO_PROT_ARGS); static int obsolete(MACRO_PROT_ARGS); +static int phrase_ta(MACRO_PROT_ARGS); static int append_delims(struct mdoc *, int, int *, char *); static enum mdoct lookup(enum mdoct, const char *); static enum mdoct lookup_raw(const char *); -static int phrase(struct mdoc *, int, int, - char *, enum margserr); +static int phrase(struct mdoc *, int, int, char *); static enum mdoct rew_alt(enum mdoct); static int rew_dobreak(enum mdoct, const struct mdoc_node *); @@ -186,6 +186,7 @@ const struct mdoc_macro __mdoc_macros[MDOC_MAX] = { { in_line_eoln, 0 }, /* br */ { in_line_eoln, 0 }, /* sp */ { in_line_eoln, 0 }, /* %U */ + { phrase_ta, MDOC_CALLABLE | MDOC_PARSED }, /* Ta */ }; const struct mdoc_macro * const mdoc_macros = __mdoc_macros; @@ -196,7 +197,7 @@ swarn(struct mdoc *mdoc, enum mdoc_type type, int line, int pos, const struct mdoc_node *p) { const char *n, *t, *tt; - int rc; + enum mandocerr ec; n = t = ""; tt = "block"; @@ -229,11 +230,12 @@ swarn(struct mdoc *mdoc, enum mdoc_type type, break; } - rc = mdoc_vmsg(mdoc, MANDOCERR_SCOPE, line, pos, - "%s scope breaks %s of %s", tt, t, n); + ec = (MDOC_IGN_SCOPE & mdoc->pflags) ? + MANDOCERR_SCOPE : MANDOCERR_SYNTSCOPE; - /* FIXME: logic should be in driver. */ - return(MDOC_IGN_SCOPE & mdoc->pflags ? rc : 0); + return(mdoc_vmsg(mdoc, ec, line, pos, + "%s scope breaks %s of %s", + tt, t, n)); } @@ -1006,14 +1008,30 @@ blk_full(MACRO_PROT_ARGS) for ( ; ; ) { la = *pos; - /* Initialise last-phrase-type with ARGS_PHRASE. */ - lac = ARGS_ERROR == ac ? ARGS_PHRASE : ac; + /* Initialise last-phrase-type with ARGS_PEND. */ + lac = ARGS_ERROR == ac ? ARGS_PEND : ac; ac = mdoc_args(m, line, pos, buf, tok, &p); if (ARGS_ERROR == ac) return(0); - if (ARGS_EOLN == ac) + + if (ARGS_EOLN == ac) { + if (ARGS_PPHRASE != lac && ARGS_PHRASE != lac) + break; + /* + * This is necessary: if the last token on a + * line is a `Ta' or tab, then we'll get + * ARGS_EOLN, so we must be smart enough to + * reopen our scope if the last parse was a + * phrase or partial phrase. + */ + if ( ! rew_sub(MDOC_BODY, m, tok, line, ppos)) + return(0); + if ( ! mdoc_body_alloc(m, line, ppos, tok)) + return(0); + body = m->last; break; + } /* * Emit leading punctuation (i.e., punctuation before @@ -1068,10 +1086,7 @@ blk_full(MACRO_PROT_ARGS) if (ARGS_PEND == ac && ARGS_PPHRASE == lac) m->flags |= MDOC_PPHRASE; - if (ARGS_PEND == ac) { - if ( ! phrase(m, line, la, buf, lac)) - return(0); - } else if ( ! phrase(m, line, la, buf, ac)) + if ( ! phrase(m, line, la, buf)) return(0); m->flags &= ~MDOC_PPHRASE; @@ -1624,26 +1639,24 @@ obsolete(MACRO_PROT_ARGS) * macro is encountered. */ static int -phrase(struct mdoc *m, int line, int ppos, char *buf, enum margserr ac) +phrase(struct mdoc *m, int line, int ppos, char *buf) { int la, pos; - enum margserr aac; + enum margserr ac; enum mdoct ntok; char *p; - assert(ARGS_PHRASE == ac || ARGS_PPHRASE == ac); - for (pos = ppos; ; ) { la = pos; - aac = mdoc_zargs(m, line, &pos, buf, 0, &p); + ac = mdoc_zargs(m, line, &pos, buf, 0, &p); - if (ARGS_ERROR == aac) + if (ARGS_ERROR == ac) return(0); - if (ARGS_EOLN == aac) + if (ARGS_EOLN == ac) break; - ntok = ARGS_QWORD == aac ? MDOC_MAX : lookup_raw(p); + ntok = ARGS_QWORD == ac ? MDOC_MAX : lookup_raw(p); if (MDOC_MAX == ntok) { if ( ! mdoc_word_alloc(m, line, la, p)) @@ -1660,3 +1673,64 @@ phrase(struct mdoc *m, int line, int ppos, char *buf, enum margserr ac) } +static int +phrase_ta(MACRO_PROT_ARGS) +{ + int la; + enum mdoct ntok; + enum margserr ac; + struct mdoc_node *n; + char *p; + + /* + * FIXME: this is overly restrictive: if the `Ta' is unexpected, + * it should simply error out with ARGSLOST. + */ + + n = m->last; + if ( ! rew_sub(MDOC_BODY, m, MDOC_It, line, ppos)) + return(0); + + /* + * FIXME: this is necessary in bogus constructions like + * .Bl -column foo bar + * .Sy foo Ta bar + * It is, however, an ugly way to do it. + * + * XXX; remove this when the above construct builds an implied + * `It' marker. + */ + if (NULL == m->last || + MDOC_BODY != m->last->type || + MDOC_It != m->last->tok) { + swarn(m, tok, line, ppos, n); + return(0); + } + + if ( ! mdoc_body_alloc(m, line, ppos, MDOC_It)) + return(0); + + for (;;) { + la = *pos; + ac = mdoc_zargs(m, line, pos, buf, 0, &p); + + if (ARGS_ERROR == ac) + return(0); + if (ARGS_EOLN == ac) + break; + + ntok = ARGS_QWORD == ac ? MDOC_MAX : lookup_raw(p); + + if (MDOC_MAX == ntok) { + if ( ! mdoc_word_alloc(m, line, la, p)) + return(0); + continue; + } + + if ( ! mdoc_macro(m, ntok, line, la, pos, buf)) + return(0); + return(append_delims(m, line, pos, buf)); + } + + return(1); +} diff --git a/mdoc_term.c b/mdoc_term.c index 9bd89edc..6344b22d 100644 --- a/mdoc_term.c +++ b/mdoc_term.c @@ -260,6 +260,7 @@ static const struct termact termacts[MDOC_MAX] = { { termp_sp_pre, NULL }, /* br */ { termp_sp_pre, NULL }, /* sp */ { termp_under_pre, termp____post }, /* %U */ + { NULL, NULL }, /* Ta */ }; diff --git a/mdoc_validate.c b/mdoc_validate.c index 8beae48a..4296cc3c 100644 --- a/mdoc_validate.c +++ b/mdoc_validate.c @@ -268,6 +268,7 @@ const struct valids mdoc_valids[MDOC_MAX] = { { NULL, posts_notext }, /* br */ { NULL, posts_sp }, /* sp */ { NULL, posts_text1 }, /* %U */ + { NULL, NULL }, /* Ta */ }; -- cgit