From 94cd81895ea53a20400d13b0e9c2bcf756ed382b Mon Sep 17 00:00:00 2001
From: Ingo Schwarze <schwarze@openbsd.org>
Date: Thu, 26 Jan 2017 18:28:18 +0000
Subject: Fix -man -Thtml formatting after .nf (which has nothing to do with
 "literal", by the way, it means "no fill"):

* Use <pre> such that whitespace is preserved.
* Preserve lines breaks.
* For font alternating macros, avoid node recursion which required
scary juggling with the fill state.  Instead, simply print the text
children directly.

Missing feature first noticed by kristaps@ in 2011,
the again reported by afresh1@ in 2016,
and finally reported here: https://github.com/Debian/debiman/issues/21 ,
which i only found because of Shane Kerr's comment here:
https://plus.google.com/110314300533310775053/posts/H1eaw9Yskoc
---
 TODO       |  11 +------
 html.c     |   5 ++-
 html.h     |   3 +-
 man_html.c | 105 +++++++++++++++++++++++++++----------------------------------
 4 files changed, 52 insertions(+), 72 deletions(-)

diff --git a/TODO b/TODO
index 30313e90..f4b9a74e 100644
--- a/TODO
+++ b/TODO
@@ -1,6 +1,6 @@
 ************************************************************************
 * Official mandoc TODO.
-* $Id: TODO,v 1.224 2017/01/25 02:14:43 schwarze Exp $
+* $Id: TODO,v 1.225 2017/01/26 18:28:18 schwarze Exp $
 ************************************************************************
 
 Many issues are annotated for difficulty as follows:
@@ -193,10 +193,6 @@ are mere guesses, and some may be wrong.
   uqs@  Thu, 2 Jun 2011 11:33:35 +0200
   loc *  exist **  algo ***  size *  imp **
 
---- missing man features -----------------------------------------------
-
-- -T[x]html doesn't stipulate non-collapsing spaces in literal mode
-
 --- missing tbl features -----------------------------------------------
 
 - horizontal lines in the layout still consume data cells
@@ -408,11 +404,6 @@ are mere guesses, and some may be wrong.
 
 --- HTML issues --------------------------------------------------------
 
-- In -man -Thtml, .nf does not preserve indentation.
-  It should either convert blanks to &nbsp;
-  or use <pre> rather than <div> (like .Bd -literal does).
-  Reported by afresh1@  12 Apr 2016 14:35:45 -0700
-
 - .Bf at the beginning of a paragraph inserts a bogus 1ex horizontal
   space, see for example random(3).  Introduced in
   http://mdocml.bsd.lv/cgi-bin/cvsweb/mdoc_html.c.diff?r1=1.91&r2=1.92
diff --git a/html.c b/html.c
index cee98aea..7af682e7 100644
--- a/html.c
+++ b/html.c
@@ -1,4 +1,4 @@
-/*	$Id: html.c,v 1.201 2017/01/25 02:14:43 schwarze Exp $ */
+/*	$Id: html.c,v 1.202 2017/01/26 18:28:18 schwarze Exp $ */
 /*
  * Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011-2015, 2017 Ingo Schwarze <schwarze@openbsd.org>
@@ -114,7 +114,6 @@ static	const char	*const roffscales[SCALE_MAX] = {
 
 static	void	 a2width(const char *, struct roffsu *);
 static	void	 print_byte(struct html *, char);
-static	void	 print_endline(struct html *);
 static	void	 print_endword(struct html *);
 static	void	 print_indent(struct html *);
 static	void	 print_word(struct html *, const char *);
@@ -812,7 +811,7 @@ print_byte(struct html *h, char c)
  * If something was printed on the current output line, end it.
  * Not to be called right after print_indent().
  */
-static void
+void
 print_endline(struct html *h)
 {
 	if (h->col == 0)
diff --git a/html.h b/html.h
index 19532c49..a393ff62 100644
--- a/html.h
+++ b/html.h
@@ -1,4 +1,4 @@
-/*	$Id: html.h,v 1.78 2017/01/19 16:59:30 schwarze Exp $ */
+/*	$Id: html.h,v 1.79 2017/01/26 18:28:18 schwarze Exp $ */
 /*
  * Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2017 Ingo Schwarze <schwarze@openbsd.org>
@@ -127,5 +127,6 @@ void		  print_tblclose(struct html *);
 void		  print_tbl(struct html *, const struct tbl_span *);
 void		  print_eqn(struct html *, const struct eqn *);
 void		  print_paragraph(struct html *);
+void		  print_endline(struct html *);
 
 int		  html_strlen(const char *);
diff --git a/man_html.c b/man_html.c
index 641e0e33..804859f8 100644
--- a/man_html.c
+++ b/man_html.c
@@ -1,4 +1,4 @@
-/*	$Id: man_html.c,v 1.129 2017/01/21 01:20:32 schwarze Exp $ */
+/*	$Id: man_html.c,v 1.130 2017/01/26 18:28:18 schwarze Exp $ */
 /*
  * Copyright (c) 2008-2012, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2013, 2014, 2015, 2017 Ingo Schwarze <schwarze@openbsd.org>
@@ -32,7 +32,6 @@
 #include "html.h"
 #include "main.h"
 
-/* TODO: preserve ident widths. */
 /* FIXME: have PD set the default vspace width. */
 
 #define	INDENT		  5
@@ -43,8 +42,7 @@
 			  struct html *h
 
 struct	mhtml {
-	int		  fl;
-#define	MANH_LITERAL	 (1 << 0) /* literal context */
+	struct tag	 *nofill;
 };
 
 struct	htmlman {
@@ -72,9 +70,9 @@ static	int		  man_SS_pre(MAN_ARGS);
 static	int		  man_UR_pre(MAN_ARGS);
 static	int		  man_alt_pre(MAN_ARGS);
 static	int		  man_br_pre(MAN_ARGS);
+static	int		  man_fill_pre(MAN_ARGS);
 static	int		  man_ign_pre(MAN_ARGS);
 static	int		  man_in_pre(MAN_ARGS);
-static	int		  man_literal_pre(MAN_ARGS);
 static	void		  man_root_post(MAN_ARGS);
 static	void		  man_root_pre(MAN_ARGS);
 
@@ -101,8 +99,8 @@ static	const struct htmlman mans[MAN_MAX] = {
 	{ man_alt_pre, NULL }, /* IR */
 	{ man_alt_pre, NULL }, /* RI */
 	{ man_br_pre, NULL }, /* sp */
-	{ man_literal_pre, NULL }, /* nf */
-	{ man_literal_pre, NULL }, /* fi */
+	{ man_fill_pre, NULL }, /* nf */
+	{ man_fill_pre, NULL }, /* fi */
 	{ NULL, NULL }, /* RE */
 	{ man_RS_pre, NULL }, /* RS */
 	{ man_ign_pre, NULL }, /* DT */
@@ -112,8 +110,8 @@ static	const struct htmlman mans[MAN_MAX] = {
 	{ man_in_pre, NULL }, /* in */
 	{ man_ign_pre, NULL }, /* ft */
 	{ man_OP_pre, NULL }, /* OP */
-	{ man_literal_pre, NULL }, /* EX */
-	{ man_literal_pre, NULL }, /* EE */
+	{ man_fill_pre, NULL }, /* EX */
+	{ man_fill_pre, NULL }, /* EE */
 	{ man_UR_pre, NULL }, /* UR */
 	{ NULL, NULL }, /* UE */
 	{ man_ign_pre, NULL }, /* ll */
@@ -150,7 +148,7 @@ html_man(void *arg, const struct roff_man *man)
 	struct html	*h;
 	struct tag	*t;
 
-	memset(&mh, 0, sizeof(mh));
+	mh.nofill = NULL;
 	h = (struct html *)arg;
 
 	if ((h->oflags & HTML_FRAGMENT) == 0) {
@@ -200,6 +198,8 @@ print_man_node(MAN_ARGS)
 
 	child = 1;
 	t = h->tags.head;
+	if (t == mh->nofill)
+		t = t->next;
 
 	switch (n->type) {
 	case ROFFT_TEXT:
@@ -207,12 +207,11 @@ print_man_node(MAN_ARGS)
 			print_paragraph(h);
 			return;
 		}
-		if (n->flags & NODE_LINE && (*n->string == ' ' ||
-		    (n->prev != NULL && mh->fl & MANH_LITERAL &&
-		     ! (h->flags & HTML_NONEWLINE))))
+		if (mh->nofill == NULL &&
+		    n->flags & NODE_LINE && *n->string == ' ')
 			print_otag(h, TAG_BR, "");
 		print_text(h, n->string);
-		return;
+		break;
 	case ROFFT_EQN:
 		print_eqn(h, n->eqn);
 		break;
@@ -252,16 +251,15 @@ print_man_node(MAN_ARGS)
 		print_man_nodelist(man, n->child, mh, h);
 
 	/* This will automatically close out any font scope. */
-	print_stagq(h, t);
+	print_stagq(h, mh->nofill == NULL ? t : mh->nofill);
 
-	switch (n->type) {
-	case ROFFT_EQN:
-		break;
-	default:
-		if (mans[n->tok].post)
-			(*mans[n->tok].post)(man, n, mh, h);
-		break;
-	}
+	if (n->type != ROFFT_TEXT && n->type != ROFFT_EQN &&
+	    mans[n->tok].post != NULL)
+		(*mans[n->tok].post)(man, n, mh, h);
+
+	if (mh->nofill != NULL &&
+	    (n->next == NULL || n->next->flags & NODE_LINE))
+		print_endline(h);
 }
 
 static int
@@ -349,13 +347,11 @@ man_br_pre(MAN_ARGS)
 static int
 man_SH_pre(MAN_ARGS)
 {
-	if (n->type == ROFFT_BLOCK) {
-		mh->fl &= ~MANH_LITERAL;
-		return 1;
-	} else if (n->type == ROFFT_BODY)
-		return 1;
-
-	print_otag(h, TAG_H1, "c", "Sh");
+	if (n->type == ROFFT_BLOCK && mh->nofill != NULL) {
+		print_tagq(h, mh->nofill);
+		mh->nofill = NULL;
+	} else if (n->type == ROFFT_HEAD)
+		print_otag(h, TAG_H1, "c", "Sh");
 	return 1;
 }
 
@@ -363,17 +359,11 @@ static int
 man_alt_pre(MAN_ARGS)
 {
 	const struct roff_node	*nn;
-	int		 i, savelit;
+	int		 i;
 	enum htmltag	 fp;
 	struct tag	*t;
 
-	if ((savelit = mh->fl & MANH_LITERAL))
-		print_otag(h, TAG_BR, "");
-
-	mh->fl &= ~MANH_LITERAL;
-
 	for (i = 0, nn = n->child; nn; nn = nn->next, i++) {
-		t = NULL;
 		switch (n->tok) {
 		case MAN_BI:
 			fp = i % 2 ? TAG_I : TAG_B;
@@ -400,18 +390,14 @@ man_alt_pre(MAN_ARGS)
 		if (i)
 			h->flags |= HTML_NOSPACE;
 
-		if (TAG_MAX != fp)
+		if (fp != TAG_MAX)
 			t = print_otag(h, fp, "");
 
-		print_man_node(man, nn, mh, h);
+		print_text(h, nn->string);
 
-		if (t)
+		if (fp != TAG_MAX)
 			print_tagq(h, t);
 	}
-
-	if (savelit)
-		mh->fl |= MANH_LITERAL;
-
 	return 0;
 }
 
@@ -427,13 +413,11 @@ man_SM_pre(MAN_ARGS)
 static int
 man_SS_pre(MAN_ARGS)
 {
-	if (n->type == ROFFT_BLOCK) {
-		mh->fl &= ~MANH_LITERAL;
-		return 1;
-	} else if (n->type == ROFFT_BODY)
-		return 1;
-
-	print_otag(h, TAG_H2, "c", "Ss");
+	if (n->type == ROFFT_BLOCK && mh->nofill != NULL) {
+		print_tagq(h, mh->nofill);
+		mh->nofill = NULL;
+	} else if (n->type == ROFFT_HEAD)
+		print_otag(h, TAG_H2, "c", "Ss");
 	return 1;
 }
 
@@ -552,15 +536,20 @@ man_I_pre(MAN_ARGS)
 }
 
 static int
-man_literal_pre(MAN_ARGS)
+man_fill_pre(MAN_ARGS)
 {
-
 	if (MAN_fi == n->tok || MAN_EE == n->tok) {
-		print_otag(h, TAG_BR, "");
-		mh->fl &= ~MANH_LITERAL;
-	} else
-		mh->fl |= MANH_LITERAL;
-
+		if (mh->nofill != NULL) {
+			print_tagq(h, mh->nofill);
+			mh->nofill = NULL;
+		} else
+			print_otag(h, TAG_BR, "");
+	} else {
+		if (mh->nofill == NULL)
+			mh->nofill = print_otag(h, TAG_PRE, "");
+		else
+			print_otag(h, TAG_BR, "");
+	}
 	return 0;
 }
 
-- 
cgit v1.2.3-56-ge451