=================================================================== RCS file: /cvs/mandoc/man.c,v retrieving revision 1.61 retrieving revision 1.65 diff -u -p -r1.61 -r1.65 --- mandoc/man.c 2010/05/07 15:49:36 1.61 +++ mandoc/man.c 2010/05/12 16:46:28 1.65 @@ -1,4 +1,4 @@ -/* $Id: man.c,v 1.61 2010/05/07 15:49:36 kristaps Exp $ */ +/* $Id: man.c,v 1.65 2010/05/12 16:46:28 kristaps Exp $ */ /* * Copyright (c) 2008, 2009 Kristaps Dzonsons * @@ -395,19 +395,54 @@ man_ptext(struct man *m, int line, char *buf) goto descope; } - /* Warn if the last un-escaped character is whitespace. */ + /* + * 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 (1 == i || ('\\' != buf[i - 2])) + if (' ' == buf[i - 1] || '\t' == buf[i - 1]) { + if (i > 1 && '\\' != buf[i - 2]) if ( ! man_pwarn(m, line, i - 1, WTSPACE)) 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'; + } + if ( ! man_word_alloc(m, line, 0, buf)) return(0); + /* + * 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); + + switch (buf[i - 1]) { + case ('.'): + if (i > 1 && '\\' == buf[i - 2]) + break; + /* FALLTHROUGH */ + case ('!'): + /* FALLTHROUGH */ + case ('?'): + m->last->flags |= MAN_EOS; + break; + default: + break; + + } + descope: /* * Co-ordinate what happens with having a next-line scope open: @@ -435,8 +470,7 @@ static int macrowarn(struct man *m, int ln, const char *buf) { if ( ! (MAN_IGN_MACRO & m->pflags)) - return(man_verr(m, ln, 0, - "unknown macro: %s%s", + return(man_verr(m, ln, 0, "unknown macro: %s%s", buf, strlen(buf) > 3 ? "..." : "")); return(man_vwarn(m, ln, 0, "unknown macro: %s%s", buf, strlen(buf) > 3 ? "..." : "")); @@ -462,6 +496,7 @@ man_pmacro(struct man *m, int ln, char *buf) * Skip whitespace between the control character and initial * text. "Whitespace" is both spaces and tabs. */ + if (' ' == buf[i] || '\t' == buf[i]) { i++; while (buf[i] && (' ' == buf[i] || '\t' == buf[i])) @@ -510,7 +545,10 @@ man_pmacro(struct man *m, int ln, char *buf) while (buf[i] && ' ' == buf[i]) i++; - /* Trailing whitespace? */ + /* + * Trailing whitespace. Note that tabs are allowed to be passed + * into the parser as "text", so we only warn about spaces here. + */ if ('\0' == buf[i] && ' ' == buf[i - 1]) if ( ! man_pwarn(m, ln, i - 1, WTSPACE))