Support `T{' and `T}' data blocks. When a standalone `T{' is

encountered as a line's last data cell, move into TBL_PART_CDATA mode
whilst leaving the cell's designation as TBL_DATA_NONE.  When new data
arrives that's not a standalone `T}', append it to the cell contends.
Close out and warn appropriately.

8 files changed, 86 insertions, 23 deletions

diff --git a/libroff.h b/libroff.h
index 70c21398..b4e043a7 100644
--- a/libroff.h
+++ b/libroff.h
@@ -1,4 +1,4 @@
-/*	$Id: libroff.h,v 1.15 2011/01/02 10:10:57 kristaps Exp $ */
+/*	$Id: libroff.h,v 1.16 2011/01/04 15:02:00 kristaps Exp $ */
 /*
  * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
  *
@@ -22,7 +22,8 @@ __BEGIN_DECLS
 enum	tbl_part {
 	TBL_PART_OPTS, /* in options (first line) */
 	TBL_PART_LAYOUT, /* describing layout */
-	TBL_PART_DATA  /* creating data rows */
+	TBL_PART_DATA, /* creating data rows */
+	TBL_PART_CDATA /* continue previous row */
 };
 
 struct	tbl_node {
@@ -52,6 +53,7 @@ enum rofferr 	 tbl_read(struct tbl_node *, int, const char *, int);
 int		 tbl_option(struct tbl_node *, int, const char *);
 int		 tbl_layout(struct tbl_node *, int, const char *);
 int		 tbl_data(struct tbl_node *, int, const char *);
+int		 tbl_cdata(struct tbl_node *, int, const char *);
 const struct tbl_span *tbl_span(const struct tbl_node *);
 void		 tbl_end(struct tbl_node *);
 
diff --git a/main.c b/main.c
index 63b778f4..2be68a93 100644
--- a/main.c
+++ b/main.c
@@ -1,4 +1,4 @@
-/*	$Id: main.c,v 1.134 2011/01/04 12:06:21 kristaps Exp $ */
+/*	$Id: main.c,v 1.135 2011/01/04 15:02:00 kristaps Exp $ */
 /*
  * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2010 Ingo Schwarze <schwarze@openbsd.org>
@@ -190,6 +190,7 @@ static	const char * const	mandocerrs[MANDOCERR_MAX] = {
 	"no table layout cells specified",
 	"no table data cells specified",
 	"ignore data in cell",
+	"data block still open",
 
 	"input stack limit exceeded, infinite loop?",
 	"skipping bad character",
diff --git a/mandoc.h b/mandoc.h
index 75d8f1d3..7da2e541 100644
--- a/mandoc.h
+++ b/mandoc.h
@@ -1,4 +1,4 @@
-/*	$Id: mandoc.h,v 1.47 2011/01/04 12:06:21 kristaps Exp $ */
+/*	$Id: mandoc.h,v 1.48 2011/01/04 15:02:00 kristaps Exp $ */
 /*
  * Copyright (c) 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
  *
@@ -112,6 +112,7 @@ enum	mandocerr {
 	MANDOCERR_TBLNOLAYOUT, /* no table layout cells specified */
 	MANDOCERR_TBLNODATA, /* no table data cells specified */
 	MANDOCERR_TBLIGNDATA, /* ignore data in cell */
+	MANDOCERR_TBLBLOCK, /* data block still open */
 
 	MANDOCERR_ROFFLOOP, /* input stack limit exceeded, infinite loop? */
 	MANDOCERR_BADCHAR, /* skipping bad character */
@@ -224,6 +225,7 @@ struct	tbl_row {
 };
 
 enum	tbl_datt {
+	TBL_DATA_NONE,
 	TBL_DATA_DATA,
 	TBL_DATA_HORIZ,
 	TBL_DATA_DHORIZ,
diff --git a/roff.7 b/roff.7
index cdf20a27..1967a1d7 100644
--- a/roff.7
+++ b/roff.7
@@ -1,4 +1,4 @@
-.\"	$Id: roff.7,v 1.21 2011/01/04 12:06:21 kristaps Exp $
+.\"	$Id: roff.7,v 1.22 2011/01/04 15:02:00 kristaps Exp $
 .\"
 .\" Copyright (c) 2010 Kristaps Dzonsons <kristaps@bsd.lv>
 .\" Copyright (c) 2010 Ingo Schwarze <schwarze@openbsd.org>
@@ -736,6 +736,11 @@ or
 .Cm \e=
 is specified, a line is drawn within the data field (i.e., terminating
 within the cell and not draw to the border).
+If the last cell of a line is
+.Cm T{ ,
+all subsequent lines are included as part of the cell until
+.Cm T}
+is specified on its own line.
 .Sh COMPATIBILITY
 This section documents compatibility between mandoc and other other
 .Nm
diff --git a/tbl.c b/tbl.c
index 1ea37f5a..ba591ceb 100644
--- a/tbl.c
+++ b/tbl.c
@@ -1,4 +1,4 @@
-/*	$Id: tbl.c,v 1.20 2011/01/03 13:59:21 kristaps Exp $ */
+/*	$Id: tbl.c,v 1.21 2011/01/04 15:02:00 kristaps Exp $ */
 /*
  * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
  *
@@ -52,7 +52,9 @@ tbl_read(struct tbl_node *tbl, int ln, const char *p, int offs)
 		return(tbl_option(tbl, ln, p) ? ROFF_IGN : ROFF_ERR);
 	case (TBL_PART_LAYOUT):
 		return(tbl_layout(tbl, ln, p) ? ROFF_IGN : ROFF_ERR);
-	case (TBL_PART_DATA):
+	case (TBL_PART_CDATA):
+		return(tbl_cdata(tbl, ln, p) ? ROFF_TBL : ROFF_IGN);
+	default:
 		break;
 	}
 
@@ -122,6 +124,8 @@ tbl_free(struct tbl_node *p)
 void
 tbl_restart(int line, int pos, struct tbl_node *tbl)
 {
+	if (TBL_PART_CDATA == tbl->part)
+		TBL_MSG(tbl, MANDOCERR_TBLBLOCK, tbl->line, tbl->pos);
 
 	tbl->part = TBL_PART_LAYOUT;
 	tbl->line = line;
@@ -148,5 +152,8 @@ tbl_end(struct tbl_node *tbl)
 
 	if (tbl->last_span)
 		tbl->last_span->flags |= TBL_SPAN_LAST;
+
+	if (TBL_PART_CDATA == tbl->part)
+		TBL_MSG(tbl, MANDOCERR_TBLBLOCK, tbl->line, tbl->pos);
 }
 
diff --git a/tbl_data.c b/tbl_data.c
index 40264b9c..030f8c12 100644
--- a/tbl_data.c
+++ b/tbl_data.c
@@ -1,4 +1,4 @@
-/*	$Id: tbl_data.c,v 1.10 2011/01/04 12:06:21 kristaps Exp $ */
+/*	$Id: tbl_data.c,v 1.11 2011/01/04 15:02:00 kristaps Exp $ */
 /*
  * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
  *
@@ -14,6 +14,10 @@
  * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
  */
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
 #include <assert.h>
 #include <ctype.h>
 #include <stdlib.h>
@@ -24,10 +28,10 @@
 #include "libmandoc.h"
 #include "libroff.h"
 
-static	void	data(struct tbl_node *, struct tbl_span *, 
+static	int	data(struct tbl_node *, struct tbl_span *, 
 			int, const char *, int *);
 
-void
+static int
 data(struct tbl_node *tbl, struct tbl_span *dp, 
 		int ln, const char *p, int *pos)
 {
@@ -47,10 +51,9 @@ data(struct tbl_node *tbl, struct tbl_span *dp,
 				TBL_CELL_DVERT == cp->pos))
 		cp = cp->next;
 
-	/* FIXME: warn about losing data contents if cell is HORIZ. */
-
 	dat = mandoc_calloc(1, sizeof(struct tbl_dat));
 	dat->layout = cp;
+	dat->pos = TBL_DATA_NONE;
 
 	if (NULL == dat->layout)
 		TBL_MSG(tbl, MANDOCERR_TBLEXTRADAT, ln, *pos);
@@ -65,6 +68,17 @@ data(struct tbl_node *tbl, struct tbl_span *dp,
 	while (p[*pos] && p[*pos] != tbl->opts.tab)
 		(*pos)++;
 
+	/*
+	 * Check for a continued-data scope opening.  This consists of a
+	 * trailing `T{' at the end of the line.  Subsequent lines,
+	 * until a standalone `T}', are included in our cell.
+	 */
+
+	if (*pos - sv == 2 && 'T' == p[sv] && '{' == p[sv + 1]) {
+		tbl->part = TBL_PART_CDATA;
+		return(0);
+	}
+
 	dat->string = mandoc_malloc(*pos - sv + 1);
 	memcpy(dat->string, &p[sv], *pos - sv);
 	dat->string[*pos - sv] = '\0';
@@ -83,10 +97,40 @@ data(struct tbl_node *tbl, struct tbl_span *dp,
 	else
 		dat->pos = TBL_DATA_DATA;
 
+	if (NULL == dat->layout)
+		return(1);
+
 	if (TBL_CELL_HORIZ == dat->layout->pos ||
 			TBL_CELL_DHORIZ == dat->layout->pos)
 		if (TBL_DATA_DATA == dat->pos && '\0' != *dat->string)
 			TBL_MSG(tbl, MANDOCERR_TBLIGNDATA, ln, sv);
+
+	return(1);
+}
+
+int
+tbl_cdata(struct tbl_node *tbl, int ln, const char *p)
+{
+	struct tbl_dat	*dat;
+	size_t	 	 sz;
+
+	if (0 == strcmp(p, "T}")) {
+		tbl->part = TBL_PART_DATA;
+		return(1);
+	}
+
+	dat = tbl->last_span->last;
+	dat->pos = TBL_DATA_DATA;
+
+	if (dat->string) {
+		sz = strlen(p) + strlen(dat->string) + 2;
+		dat->string = mandoc_realloc(dat->string, sz);
+		strlcat(dat->string, " ", sz);
+		strlcat(dat->string, p, sz);
+	} else
+		dat->string = mandoc_strdup(p);
+
+	return(0);
 }
 
 int
@@ -141,8 +185,11 @@ tbl_data(struct tbl_node *tbl, int ln, const char *p)
 
 	dp->pos = TBL_SPAN_DATA;
 
+	/* This returns 0 when TBL_PART_CDATA is entered. */
+
 	while ('\0' != p[pos])
-		data(tbl, dp, ln, p, &pos);
+		if ( ! data(tbl, dp, ln, p, &pos))
+			return(0);
 
 	return(1);
 }
diff --git a/tbl_html.c b/tbl_html.c
index e3c710dc..2abbf4ca 100644
--- a/tbl_html.c
+++ b/tbl_html.c
@@ -1,4 +1,4 @@
-/*	$Id: tbl_html.c,v 1.1 2011/01/04 10:29:41 kristaps Exp $ */
+/*	$Id: tbl_html.c,v 1.2 2011/01/04 15:02:00 kristaps Exp $ */
 /*
  * Copyright (c) 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -63,7 +63,8 @@ print_tbl(struct html *h, const struct tbl_span *sp)
 		}
 		tt = print_otag(h, TAG_TD, 0, NULL);
 		if (dp) {
-			print_text(h, dp->string);
+			if (dp->string)
+				print_text(h, dp->string);
 			dp = dp->next;
 		}
 		print_tagq(h, tt);
diff --git a/tbl_term.c b/tbl_term.c
index a6c78e1d..267c3d14 100644
--- a/tbl_term.c
+++ b/tbl_term.c
@@ -1,4 +1,4 @@
-/*	$Id: tbl_term.c,v 1.9 2011/01/04 13:21:45 kristaps Exp $ */
+/*	$Id: tbl_term.c,v 1.10 2011/01/04 15:02:00 kristaps Exp $ */
 /*
  * Copyright (c) 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -253,6 +253,9 @@ tbl_data(struct termp *tp, const struct tbl *tbl,
 	}
 
 	switch (dp->pos) {
+	case (TBL_DATA_NONE):
+		tbl_char(tp, ASCII_NBRSP, tbp->width);
+		return;
 	case (TBL_DATA_HORIZ):
 		/* FALLTHROUGH */
 	case (TBL_DATA_NHORIZ):
@@ -420,14 +423,9 @@ tbl_calc(struct termp *tp, const struct tbl_span *sp)
 	hp = sp->head;
 
 	for ( ; sp; sp = sp->next) {
-		switch (sp->pos) {
-		case (TBL_DATA_HORIZ):
-			/* FALLTHROUGH */
-		case (TBL_DATA_DHORIZ):
+		if (TBL_SPAN_DATA != sp->pos)
 			continue;
-		default:
-			break;
-		}
+
 		for (dp = sp->first; dp; dp = dp->next) {
 			if (NULL == dp->layout)
 				continue;