=================================================================== RCS file: /cvs/mandoc/mdoc.c,v retrieving revision 1.128 retrieving revision 1.138 diff -u -p -r1.128 -r1.138 --- mandoc/mdoc.c 2010/05/12 16:01:01 1.128 +++ mandoc/mdoc.c 2010/05/25 12:37:20 1.138 @@ -1,4 +1,4 @@ -/* $Id: mdoc.c,v 1.128 2010/05/12 16:01:01 kristaps Exp $ */ +/* $Id: mdoc.c,v 1.138 2010/05/25 12:37:20 kristaps Exp $ */ /* * Copyright (c) 2008, 2009 Kristaps Dzonsons * @@ -28,65 +28,10 @@ #include #include +#include "mandoc.h" #include "libmdoc.h" #include "libmandoc.h" -const char *const __mdoc_merrnames[MERRMAX] = { - "trailing whitespace", /* ETAILWS */ - "unexpected quoted parameter", /* EQUOTPARM */ - "unterminated quoted parameter", /* EQUOTTERM */ - "argument parameter suggested", /* EARGVAL */ - "macro disallowed in prologue", /* EBODYPROL */ - "macro disallowed in body", /* EPROLBODY */ - "text disallowed in prologue", /* ETEXTPROL */ - "blank line disallowed", /* ENOBLANK */ - "text parameter too long", /* ETOOLONG */ - "invalid escape sequence", /* EESCAPE */ - "invalid character", /* EPRINT */ - "document has no body", /* ENODAT */ - "document has no prologue", /* ENOPROLOGUE */ - "expected line arguments", /* ELINE */ - "invalid AT&T argument", /* EATT */ - "default name not yet set", /* ENAME */ - "missing list type", /* ELISTTYPE */ - "missing display type", /* EDISPTYPE */ - "too many display types", /* EMULTIDISP */ - "too many list types", /* EMULTILIST */ - "NAME section must be first", /* ESECNAME */ - "badly-formed NAME section", /* ENAMESECINC */ - "argument repeated", /* EARGREP */ - "expected boolean parameter", /* EBOOL */ - "inconsistent column syntax", /* ECOLMIS */ - "nested display invalid", /* ENESTDISP */ - "width argument missing", /* EMISSWIDTH */ - "invalid section for this manual section", /* EWRONGMSEC */ - "section out of conventional order", /* ESECOOO */ - "section repeated", /* ESECREP */ - "invalid standard argument", /* EBADSTAND */ - "multi-line arguments discouraged", /* ENOMULTILINE */ - "multi-line arguments suggested", /* EMULTILINE */ - "line arguments discouraged", /* ENOLINE */ - "prologue macro out of conventional order", /* EPROLOOO */ - "prologue macro repeated", /* EPROLREP */ - "invalid manual section", /* EBADMSEC */ - "invalid section", /* EBADSEC */ - "invalid font mode", /* EFONT */ - "invalid date syntax", /* EBADDATE */ - "invalid number format", /* ENUMFMT */ - "superfluous width argument", /* ENOWIDTH */ - "system: utsname error", /* EUTSNAME */ - "obsolete macro", /* EOBS */ - "end-of-line scope violation", /* EIMPBRK */ - "empty macro ignored", /* EIGNE */ - "unclosed explicit scope", /* EOPEN */ - "unterminated quoted phrase", /* EQUOTPHR */ - "closure macro without prior context", /* ENOCTX */ - "no description found for library", /* ELIB */ - "bad child for parent context", /* EBADCHILD */ - "list arguments preceding type", /* ENOTYPE */ - "deprecated comment style", /* EBADCOMMENT */ -}; - const char *const __mdoc_macronames[MDOC_MAX] = { "Ap", "Dd", "Dt", "Os", "Sh", "Ss", "Pp", "D1", @@ -151,9 +96,10 @@ static struct mdoc_node *node_alloc(struct mdoc *, int enum mdoct, enum mdoc_type); static int node_append(struct mdoc *, struct mdoc_node *); -static int mdoc_ptext(struct mdoc *, int, char *); -static int mdoc_pmacro(struct mdoc *, int, char *); -static int macrowarn(struct mdoc *, int, const char *); +static int mdoc_ptext(struct mdoc *, int, char *, int); +static int mdoc_pmacro(struct mdoc *, int, char *, int); +static int macrowarn(struct mdoc *, int, + const char *, int); const struct mdoc_node * @@ -191,6 +137,8 @@ mdoc_free1(struct mdoc *mdoc) free(mdoc->meta.arch); if (mdoc->meta.vol) free(mdoc->meta.vol); + if (mdoc->meta.msec) + free(mdoc->meta.msec); } @@ -243,15 +191,13 @@ mdoc_free(struct mdoc *mdoc) * Allocate volatile and non-volatile parse resources. */ struct mdoc * -mdoc_alloc(void *data, int pflags, const struct mdoc_cb *cb) +mdoc_alloc(void *data, int pflags, mandocmsg msg) { struct mdoc *p; p = mandoc_calloc(1, sizeof(struct mdoc)); - if (cb) - memcpy(&p->cb, cb, sizeof(struct mdoc_cb)); - + p->msg = msg; p->data = data; p->pflags = pflags; @@ -283,68 +229,35 @@ mdoc_endparse(struct mdoc *m) * the macro (mdoc_pmacro()) or text parser (mdoc_ptext()). */ int -mdoc_parseln(struct mdoc *m, int ln, char *buf) +mdoc_parseln(struct mdoc *m, int ln, char *buf, int offs) { if (MDOC_HALT & m->flags) return(0); - return('.' == *buf ? mdoc_pmacro(m, ln, buf) : - mdoc_ptext(m, ln, buf)); + m->flags |= MDOC_NEWLINE; + return(('.' == buf[offs] || '\'' == buf[offs]) ? + mdoc_pmacro(m, ln, buf, offs) : + mdoc_ptext(m, ln, buf, offs)); } int -mdoc_verr(struct mdoc *mdoc, int ln, int pos, - const char *fmt, ...) +mdoc_vmsg(struct mdoc *mdoc, enum mandocerr t, + int ln, int pos, const char *fmt, ...) { char buf[256]; va_list ap; - if (NULL == mdoc->cb.mdoc_err) - return(0); - va_start(ap, fmt); - (void)vsnprintf(buf, sizeof(buf) - 1, fmt, ap); + vsnprintf(buf, sizeof(buf) - 1, fmt, ap); va_end(ap); - return((*mdoc->cb.mdoc_err)(mdoc->data, ln, pos, buf)); + return((*mdoc->msg)(t, mdoc->data, ln, pos, buf)); } int -mdoc_vwarn(struct mdoc *mdoc, int ln, int pos, const char *fmt, ...) -{ - char buf[256]; - va_list ap; - - if (NULL == mdoc->cb.mdoc_warn) - return(0); - - va_start(ap, fmt); - (void)vsnprintf(buf, sizeof(buf) - 1, fmt, ap); - va_end(ap); - - return((*mdoc->cb.mdoc_warn)(mdoc->data, ln, pos, buf)); -} - - -int -mdoc_err(struct mdoc *m, int line, int pos, int iserr, enum merr type) -{ - const char *p; - - p = __mdoc_merrnames[(int)type]; - assert(p); - - if (iserr) - return(mdoc_verr(m, line, pos, p)); - - return(mdoc_vwarn(m, line, pos, p)); -} - - -int mdoc_macro(struct mdoc *m, enum mdoct tok, int ln, int pp, int *pos, char *buf) { @@ -354,13 +267,13 @@ mdoc_macro(struct mdoc *m, enum mdoct tok, if (MDOC_PROLOGUE & mdoc_macros[tok].flags && MDOC_PBODY & m->flags) - return(mdoc_perr(m, ln, pp, EPROLBODY)); + return(mdoc_pmsg(m, ln, pp, MANDOCERR_BADBODY)); /* If we're in the prologue, deny "body" macros. */ if ( ! (MDOC_PROLOGUE & mdoc_macros[tok].flags) && ! (MDOC_PBODY & m->flags)) { - if ( ! mdoc_pwarn(m, ln, pp, EBODYPROL)) + if ( ! mdoc_pmsg(m, ln, pp, MANDOCERR_BADPROLOG)) return(0); if (NULL == m->meta.title) m->meta.title = mandoc_strdup("unknown"); @@ -453,7 +366,9 @@ node_alloc(struct mdoc *m, int line, int pos, p->pos = pos; p->tok = tok; p->type = type; - + if (MDOC_NEWLINE & m->flags) + p->flags |= MDOC_LINE; + m->flags &= ~MDOC_NEWLINE; return(p); } @@ -625,32 +540,71 @@ mdoc_node_delete(struct mdoc *m, struct mdoc_node *p) * control character. */ static int -mdoc_ptext(struct mdoc *m, int line, char *buf) +mdoc_ptext(struct mdoc *m, int line, char *buf, int offs) { - int i; + char *c, *ws, *end; /* Ignore bogus comments. */ - if ('\\' == buf[0] && '.' == buf[1] && '\"' == buf[2]) - return(mdoc_pwarn(m, line, 0, EBADCOMMENT)); + if ('\\' == buf[offs] && + '.' == buf[offs + 1] && + '"' == buf[offs + 2]) + return(mdoc_pmsg(m, line, offs, MANDOCERR_BADCOMMENT)); /* No text before an initial macro. */ if (SEC_NONE == m->lastnamed) - return(mdoc_perr(m, line, 0, ETEXTPROL)); + return(mdoc_pmsg(m, line, offs, MANDOCERR_NOTEXT)); - /* Literal just gets pulled in as-is. */ - - if (MDOC_LITERAL & m->flags) - return(mdoc_word_alloc(m, line, 0, buf)); + /* + * Search for the beginning of unescaped trailing whitespace (ws) + * and for the first character not to be output (end). + */ + ws = NULL; + for (c = end = buf + offs; *c; c++) { + switch (*c) { + case '-': + if (mandoc_hyph(buf + offs, c)) + *c = ASCII_HYPH; + break; + case ' ': + if (NULL == ws) + ws = c; + continue; + case '\t': + /* + * Always warn about trailing tabs, + * even outside literal context, + * where they should be put on the next line. + */ + if (NULL == ws) + ws = c; + /* + * Strip trailing tabs in literal context only; + * outside, they affect the next line. + */ + if (MDOC_LITERAL & m->flags) + continue; + break; + case '\\': + /* Skip the escaped character, too, if any. */ + if (c[1]) + c++; + /* FALLTHROUGH */ + default: + ws = NULL; + break; + } + end = c + 1; + } + *end = '\0'; - /* Check for a blank line, which may also consist of spaces. */ + if (ws) + if ( ! mdoc_pmsg(m, line, (int)(ws-buf), MANDOCERR_EOLNSPACE)) + return(0); - for (i = 0; ' ' == buf[i]; i++) - /* Skip to first non-space. */ ; - - if ('\0' == buf[i]) { - if ( ! mdoc_pwarn(m, line, 0, ENOBLANK)) + if ('\0' == buf[offs] && ! (MDOC_LITERAL & m->flags)) { + if ( ! mdoc_pmsg(m, line, (int)(c-buf), MANDOCERR_NOBLANKLN)) return(0); /* @@ -658,75 +612,45 @@ mdoc_ptext(struct mdoc *m, int line, char *buf) * 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)) + if ( ! mdoc_elem_alloc(m, line, offs, MDOC_Pp, NULL)) return(0); m->next = MDOC_NEXT_SIBLING; return(1); } - /* - * Warn if the last un-escaped character is whitespace. Then - * strip away the remaining spaces (tabs stay!). - */ - - i = (int)strlen(buf); - assert(i); - - if (' ' == buf[i - 1] || '\t' == buf[i - 1]) { - if (i > 1 && '\\' != buf[i - 2]) - if ( ! mdoc_pwarn(m, line, i - 1, ETAILWS)) - return(0); - - for (--i; i && ' ' == buf[i]; i--) - /* Spin back to non-space. */ ; - - /* Jump ahead of escaped whitespace. */ - i += '\\' == buf[i] ? 2 : 1; - - buf[i] = '\0'; - } - - /* Allocate the whole word. */ - - if ( ! mdoc_word_alloc(m, line, 0, buf)) + if ( ! mdoc_word_alloc(m, line, offs, buf+offs)) return(0); + if (MDOC_LITERAL & m->flags) + return(1); + /* * End-of-sentence check. If the last character is an unescaped * EOS character, then flag the node as being the end of a * sentence. The front-end will know how to interpret this. */ - assert(i); + assert(buf < end); - switch (buf[i - 1]) { - case ('.'): - if (i > 1 && '\\' == buf[i - 2]) - break; - /* FALLTHROUGH */ - case ('!'): - /* FALLTHROUGH */ - case ('?'): + if (mandoc_eos(buf+offs, (size_t)(end-buf-offs))) m->last->flags |= MDOC_EOS; - break; - default: - break; - } - return(1); } static int -macrowarn(struct mdoc *m, int ln, const char *buf) +macrowarn(struct mdoc *m, int ln, const char *buf, int offs) { - if ( ! (MDOC_IGN_MACRO & m->pflags)) - return(mdoc_verr(m, ln, 0, "unknown macro: %s%s", - buf, strlen(buf) > 3 ? "..." : "")); - return(mdoc_vwarn(m, ln, 0, "unknown macro: %s%s", - buf, strlen(buf) > 3 ? "..." : "")); + int rc; + + rc = mdoc_vmsg(m, MANDOCERR_MACRO, ln, offs, + "unknown macro: %s%s", + buf, strlen(buf) > 3 ? "..." : ""); + + /* FIXME: logic should be in driver. */ + return(MDOC_IGN_MACRO & m->pflags ? rc : 0); } @@ -735,18 +659,20 @@ macrowarn(struct mdoc *m, int ln, const char *buf) * character. */ int -mdoc_pmacro(struct mdoc *m, int ln, char *buf) +mdoc_pmacro(struct mdoc *m, int ln, char *buf, int offs) { enum mdoct tok; - int i, j; + int i, j, sv; char mac[5]; /* Empty lines are ignored. */ - if ('\0' == buf[1]) + offs++; + + if ('\0' == buf[offs]) return(1); - i = 1; + i = offs; /* Accept whitespace after the initial control char. */ @@ -758,6 +684,8 @@ mdoc_pmacro(struct mdoc *m, int ln, char *buf) return(1); } + sv = i; + /* Copy the first word into a nil-terminated buffer. */ for (j = 0; j < 4; j++, i++) { @@ -770,19 +698,21 @@ mdoc_pmacro(struct mdoc *m, int ln, char *buf) if (isgraph((u_char)buf[i])) continue; - return(mdoc_perr(m, ln, i, EPRINT)); + if ( ! mdoc_pmsg(m, ln, i, MANDOCERR_BADCHAR)) + return(0); + i--; } - mac[j] = 0; + mac[j] = '\0'; if (j == 4 || j < 2) { - if ( ! macrowarn(m, ln, mac)) + if ( ! macrowarn(m, ln, mac, sv)) goto err; return(1); } if (MDOC_MAX == (tok = mdoc_hash_find(mac))) { - if ( ! macrowarn(m, ln, mac)) + if ( ! macrowarn(m, ln, mac, sv)) goto err; return(1); } @@ -798,14 +728,14 @@ mdoc_pmacro(struct mdoc *m, int ln, char *buf) */ if ('\0' == buf[i] && ' ' == buf[i - 1]) - if ( ! mdoc_pwarn(m, ln, i - 1, ETAILWS)) + if ( ! mdoc_pmsg(m, ln, i - 1, MANDOCERR_EOLNSPACE)) goto err; /* * Begin recursive parse sequence. Since we're at the start of * the line, we don't need to do callable/parseable checks. */ - if ( ! mdoc_macro(m, tok, ln, 1, &i, buf)) + if ( ! mdoc_macro(m, tok, ln, sv, &i, buf)) goto err; return(1);