Handle binary files in diffs

This teaches all diff-related operations (i.e. ui-log, ui-diff and ui-patch)
how to handle binary files.

Signed-off-by: Lars Hjemli <hjemli@gmail.com>

5 files changed, 52 insertions, 7 deletions

diff --git a/cgit.h b/cgit.h
index 4fe94c6..8b0defa 100644
--- a/cgit.h
+++ b/cgit.h
@@ -1,243 +1,245 @@
 #ifndef CGIT_H
 #define CGIT_H
 
 
 #include <git-compat-util.h>
 #include <cache.h>
 #include <grep.h>
 #include <object.h>
 #include <tree.h>
 #include <commit.h>
 #include <tag.h>
 #include <diff.h>
 #include <diffcore.h>
 #include <refs.h>
 #include <revision.h>
 #include <log-tree.h>
 #include <archive.h>
+#include <xdiff-interface.h>
 #include <xdiff/xdiff.h>
 #include <utf8.h>
 
 
 /*
  * Dateformats used on misc. pages
  */
 #define FMT_LONGDATE "%Y-%m-%d %H:%M:%S (%Z)"
 #define FMT_SHORTDATE "%Y-%m-%d"
 #define FMT_ATOMDATE "%Y-%m-%dT%H:%M:%SZ"
 
 
 /*
  * Limits used for relative dates
  */
 #define TM_MIN    60
 #define TM_HOUR  (TM_MIN * 60)
 #define TM_DAY   (TM_HOUR * 24)
 #define TM_WEEK  (TM_DAY * 7)
 #define TM_YEAR  (TM_DAY * 365)
 #define TM_MONTH (TM_YEAR / 12.0)
 
 
 /*
  * Default encoding
  */
 #define PAGE_ENCODING "UTF-8"
 
 typedef void (*configfn)(const char *name, const char *value);
 typedef void (*filepair_fn)(struct diff_filepair *pair);
 typedef void (*linediff_fn)(char *line, int len);
 
 struct cgit_repo {
 	char *url;
 	char *name;
 	char *path;
 	char *desc;
 	char *owner;
 	char *defbranch;
 	char *group;
 	char *module_link;
 	char *readme;
 	char *clone_url;
 	int snapshots;
 	int enable_log_filecount;
 	int enable_log_linecount;
 	int max_stats;
 	time_t mtime;
 };
 
 struct cgit_repolist {
 	int length;
 	int count;
 	struct cgit_repo *repos;
 };
 
 struct commitinfo {
 	struct commit *commit;
 	char *author;
 	char *author_email;
 	unsigned long author_date;
 	char *committer;
 	char *committer_email;
 	unsigned long committer_date;
 	char *subject;
 	char *msg;
 	char *msg_encoding;
 };
 
 struct taginfo {
 	char *tagger;
 	char *tagger_email;
 	unsigned long tagger_date;
 	char *msg;
 };
 
 struct refinfo {
 	const char *refname;
 	struct object *object;
 	union {
 		struct taginfo *tag;
 		struct commitinfo *commit;
 	};
 };
 
 struct reflist {
 	struct refinfo **refs;
 	int alloc;
 	int count;
 };
 
 struct cgit_query {
 	int has_symref;
 	int has_sha1;
 	char *raw;
 	char *repo;
 	char *page;
 	char *search;
 	char *grep;
 	char *head;
 	char *sha1;
 	char *sha2;
 	char *path;
 	char *name;
 	char *mimetype;
 	char *url;
 	char *period;
 	int   ofs;
 	int nohead;
 	char *sort;
 	int showmsg;
 };
 
 struct cgit_config {
 	char *agefile;
 	char *cache_root;
 	char *clone_prefix;
 	char *css;
 	char *favicon;
 	char *footer;
 	char *index_header;
 	char *index_info;
 	char *logo;
 	char *logo_link;
 	char *module_link;
 	char *repo_group;
 	char *robots;
 	char *root_title;
 	char *root_desc;
 	char *root_readme;
 	char *script_name;
 	char *virtual_root;
 	int cache_size;
 	int cache_dynamic_ttl;
 	int cache_max_create_time;
 	int cache_repo_ttl;
 	int cache_root_ttl;
 	int cache_static_ttl;
 	int enable_index_links;
 	int enable_log_filecount;
 	int enable_log_linecount;
 	int local_time;
 	int max_repo_count;
 	int max_commit_count;
 	int max_lock_attempts;
 	int max_msg_len;
 	int max_repodesc_len;
 	int max_stats;
 	int nocache;
 	int renamelimit;
 	int snapshots;
 	int summary_branches;
 	int summary_log;
 	int summary_tags;
 };
 
 struct cgit_page {
 	time_t modified;
 	time_t expires;
 	size_t size;
 	char *mimetype;
 	char *charset;
 	char *filename;
 	char *title;
 };
 
 struct cgit_context {
 	struct cgit_query qry;
 	struct cgit_config cfg;
 	struct cgit_repo *repo;
 	struct cgit_page page;
 };
 
 struct cgit_snapshot_format {
 	const char *suffix;
 	const char *mimetype;
 	write_archive_fn_t write_func;
 	int bit;
 };
 
 extern const char *cgit_version;
 
 extern struct cgit_repolist cgit_repolist;
 extern struct cgit_context ctx;
 extern const struct cgit_snapshot_format cgit_snapshot_formats[];
 
 extern struct cgit_repo *cgit_add_repo(const char *url);
 extern struct cgit_repo *cgit_get_repoinfo(const char *url);
 extern void cgit_repo_config_cb(const char *name, const char *value);
 
 extern int chk_zero(int result, char *msg);
 extern int chk_positive(int result, char *msg);
 extern int chk_non_negative(int result, char *msg);
 
 extern char *trim_end(const char *str, char c);
 extern char *strlpart(char *txt, int maxlen);
 extern char *strrpart(char *txt, int maxlen);
 
 extern void cgit_add_ref(struct reflist *list, struct refinfo *ref);
 extern int cgit_refs_cb(const char *refname, const unsigned char *sha1,
 			int flags, void *cb_data);
 
 extern void *cgit_free_commitinfo(struct commitinfo *info);
 
 extern int cgit_diff_files(const unsigned char *old_sha1,
 			   const unsigned char *new_sha1,
-			   linediff_fn fn);
+			   unsigned long *old_size, unsigned long *new_size,
+			   int *binary, linediff_fn fn);
 
 extern void cgit_diff_tree(const unsigned char *old_sha1,
 			   const unsigned char *new_sha1,
 			   filepair_fn fn, const char *prefix);
 
 extern void cgit_diff_commit(struct commit *commit, filepair_fn fn);
 
 extern char *fmt(const char *format,...);
 
 extern struct commitinfo *cgit_parse_commit(struct commit *commit);
 extern struct taginfo *cgit_parse_tag(struct tag *tag);
 extern void cgit_parse_url(const char *url);
 
 extern const char *cgit_repobasename(const char *reponame);
 
 extern int cgit_parse_snapshots_mask(const char *str);
 
 
 #endif /* CGIT_H */
diff --git a/shared.c b/shared.c
index 578a544..dbb84d8 100644
--- a/shared.c
+++ b/shared.c
@@ -68,281 +68,290 @@ struct cgit_repo *cgit_add_repo(const char *url)
 struct cgit_repo *cgit_get_repoinfo(const char *url)
 {
 	int i;
 	struct cgit_repo *repo;
 
 	for (i=0; i<cgit_repolist.count; i++) {
 		repo = &cgit_repolist.repos[i];
 		if (!strcmp(repo->url, url))
 			return repo;
 	}
 	return NULL;
 }
 
 void *cgit_free_commitinfo(struct commitinfo *info)
 {
 	free(info->author);
 	free(info->author_email);
 	free(info->committer);
 	free(info->committer_email);
 	free(info->subject);
 	free(info->msg);
 	free(info->msg_encoding);
 	free(info);
 	return NULL;
 }
 
 char *trim_end(const char *str, char c)
 {
 	int len;
 	char *s, *t;
 
 	if (str == NULL)
 		return NULL;
 	t = (char *)str;
 	len = strlen(t);
 	while(len > 0 && t[len - 1] == c)
 		len--;
 
 	if (len == 0)
 		return NULL;
 
 	c = t[len];
 	t[len] = '\0';
 	s = xstrdup(t);
 	t[len] = c;
 	return s;
 }
 
 char *strlpart(char *txt, int maxlen)
 {
 	char *result;
 
 	if (!txt)
 		return txt;
 
 	if (strlen(txt) <= maxlen)
 		return txt;
 	result = xmalloc(maxlen + 1);
 	memcpy(result, txt, maxlen - 3);
 	result[maxlen-1] = result[maxlen-2] = result[maxlen-3] = '.';
 	result[maxlen] = '\0';
 	return result;
 }
 
 char *strrpart(char *txt, int maxlen)
 {
 	char *result;
 
 	if (!txt)
 		return txt;
 
 	if (strlen(txt) <= maxlen)
 		return txt;
 	result = xmalloc(maxlen + 1);
 	memcpy(result + 3, txt + strlen(txt) - maxlen + 4, maxlen - 3);
 	result[0] = result[1] = result[2] = '.';
 	return result;
 }
 
 void cgit_add_ref(struct reflist *list, struct refinfo *ref)
 {
 	size_t size;
 
 	if (list->count >= list->alloc) {
 		list->alloc += (list->alloc ? list->alloc : 4);
 		size = list->alloc * sizeof(struct refinfo *);
 		list->refs = xrealloc(list->refs, size);
 	}
 	list->refs[list->count++] = ref;
 }
 
 struct refinfo *cgit_mk_refinfo(const char *refname, const unsigned char *sha1)
 {
 	struct refinfo *ref;
 
 	ref = xmalloc(sizeof (struct refinfo));
 	ref->refname = xstrdup(refname);
 	ref->object = parse_object(sha1);
 	switch (ref->object->type) {
 	case OBJ_TAG:
 		ref->tag = cgit_parse_tag((struct tag *)ref->object);
 		break;
 	case OBJ_COMMIT:
 		ref->commit = cgit_parse_commit((struct commit *)ref->object);
 		break;
 	}
 	return ref;
 }
 
 int cgit_refs_cb(const char *refname, const unsigned char *sha1, int flags,
 		  void *cb_data)
 {
 	struct reflist *list = (struct reflist *)cb_data;
 	struct refinfo *info = cgit_mk_refinfo(refname, sha1);
 
 	if (info)
 		cgit_add_ref(list, info);
 	return 0;
 }
 
 void cgit_diff_tree_cb(struct diff_queue_struct *q,
 		       struct diff_options *options, void *data)
 {
 	int i;
 
 	for (i = 0; i < q->nr; i++) {
 		if (q->queue[i]->status == 'U')
 			continue;
 		((filepair_fn)data)(q->queue[i]);
 	}
 }
 
 static int load_mmfile(mmfile_t *file, const unsigned char *sha1)
 {
 	enum object_type type;
 
 	if (is_null_sha1(sha1)) {
 		file->ptr = (char *)"";
 		file->size = 0;
 	} else {
 		file->ptr = read_sha1_file(sha1, &type, 
 		                           (unsigned long *)&file->size);
 	}
 	return 1;
 }
 
 /*
  * Receive diff-buffers from xdiff and concatenate them as
  * needed across multiple callbacks.
  *
  * This is basically a copy of xdiff-interface.c/xdiff_outf(),
  * ripped from git and modified to use globals instead of
  * a special callback-struct.
  */
 char *diffbuf = NULL;
 int buflen = 0;
 
 int filediff_cb(void *priv, mmbuffer_t *mb, int nbuf)
 {
 	int i;
 
 	for (i = 0; i < nbuf; i++) {
 		if (mb[i].ptr[mb[i].size-1] != '\n') {
 			/* Incomplete line */
 			diffbuf = xrealloc(diffbuf, buflen + mb[i].size);
 			memcpy(diffbuf + buflen, mb[i].ptr, mb[i].size);
 			buflen += mb[i].size;
 			continue;
 		}
 
 		/* we have a complete line */
 		if (!diffbuf) {
 			((linediff_fn)priv)(mb[i].ptr, mb[i].size);
 			continue;
 		}
 		diffbuf = xrealloc(diffbuf, buflen + mb[i].size);
 		memcpy(diffbuf + buflen, mb[i].ptr, mb[i].size);
 		((linediff_fn)priv)(diffbuf, buflen + mb[i].size);
 		free(diffbuf);
 		diffbuf = NULL;
 		buflen = 0;
 	}
 	if (diffbuf) {
 		((linediff_fn)priv)(diffbuf, buflen);
 		free(diffbuf);
 		diffbuf = NULL;
 		buflen = 0;
 	}
 	return 0;
 }
 
 int cgit_diff_files(const unsigned char *old_sha1,
-		     const unsigned char *new_sha1,
-		     linediff_fn fn)
+		    const unsigned char *new_sha1, unsigned long *old_size,
+		    unsigned long *new_size, int *binary, linediff_fn fn)
 {
 	mmfile_t file1, file2;
 	xpparam_t diff_params;
 	xdemitconf_t emit_params;
 	xdemitcb_t emit_cb;
 
 	if (!load_mmfile(&file1, old_sha1) || !load_mmfile(&file2, new_sha1))
 		return 1;
 
+	*old_size = file1.size;
+	*new_size = file2.size;
+
+	if (buffer_is_binary(file1.ptr, file1.size) ||
+	    buffer_is_binary(file2.ptr, file2.size)) {
+		*binary = 1;
+		return 0;
+	}
+
 	memset(&diff_params, 0, sizeof(diff_params));
 	memset(&emit_params, 0, sizeof(emit_params));
 	memset(&emit_cb, 0, sizeof(emit_cb));
 	diff_params.flags = XDF_NEED_MINIMAL;
 	emit_params.ctxlen = 3;
 	emit_params.flags = XDL_EMIT_FUNCNAMES;
 	emit_cb.outf = filediff_cb;
 	emit_cb.priv = fn;
 	xdl_diff(&file1, &file2, &diff_params, &emit_params, &emit_cb);
 	return 0;
 }
 
 void cgit_diff_tree(const unsigned char *old_sha1,
 		    const unsigned char *new_sha1,
 		    filepair_fn fn, const char *prefix)
 {
 	struct diff_options opt;
 	int ret;
 	int prefixlen;
 
 	diff_setup(&opt);
 	opt.output_format = DIFF_FORMAT_CALLBACK;
 	opt.detect_rename = 1;
 	opt.rename_limit = ctx.cfg.renamelimit;
 	DIFF_OPT_SET(&opt, RECURSIVE);
 	opt.format_callback = cgit_diff_tree_cb;
 	opt.format_callback_data = fn;
 	if (prefix) {
 		opt.nr_paths = 1;
 		opt.paths = &prefix;
 		prefixlen = strlen(prefix);
 		opt.pathlens = &prefixlen;
 	}
 	diff_setup_done(&opt);
 
 	if (old_sha1 && !is_null_sha1(old_sha1))
 		ret = diff_tree_sha1(old_sha1, new_sha1, "", &opt);
 	else
 		ret = diff_root_tree_sha1(new_sha1, "", &opt);
 	diffcore_std(&opt);
 	diff_flush(&opt);
 }
 
 void cgit_diff_commit(struct commit *commit, filepair_fn fn)
 {
 	unsigned char *old_sha1 = NULL;
 
 	if (commit->parents)
 		old_sha1 = commit->parents->item->object.sha1;
 	cgit_diff_tree(old_sha1, commit->object.sha1, fn, NULL);
 }
 
 int cgit_parse_snapshots_mask(const char *str)
 {
 	const struct cgit_snapshot_format *f;
 	static const char *delim = " \t,:/|;";
 	int tl, sl, rv = 0;
 
 	/* favor legacy setting */
 	if(atoi(str))
 		return 1;
 	for(;;) {
 		str += strspn(str,delim);
 		tl = strcspn(str,delim);
 		if (!tl)
 			break;
 		for (f = cgit_snapshot_formats; f->suffix; f++) {
 			sl = strlen(f->suffix);
 			if((tl == sl && !strncmp(f->suffix, str, tl)) ||
 			   (tl == sl-1 && !strncmp(f->suffix+1, str, tl-1))) {
 				rv |= f->bit;
 				break;
 			}
 		}
 		str += tl;
 	}
 	return rv;
 }
diff --git a/ui-diff.c b/ui-diff.c
index 878496d..2196745 100644
--- a/ui-diff.c
+++ b/ui-diff.c
@@ -1,291 +1,313 @@
 /* ui-diff.c: show diff between two blobs
  *
  * Copyright (C) 2006 Lars Hjemli
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
  */
 
 #include "cgit.h"
 #include "html.h"
 #include "ui-shared.h"
 
 unsigned char old_rev_sha1[20];
 unsigned char new_rev_sha1[20];
 
 static int files, slots;
 static int total_adds, total_rems, max_changes;
 static int lines_added, lines_removed;
 
 static struct fileinfo {
 	char status;
 	unsigned char old_sha1[20];
 	unsigned char new_sha1[20];
 	unsigned short old_mode;
 	unsigned short new_mode;
 	char *old_path;
 	char *new_path;
 	unsigned int added;
 	unsigned int removed;
+	unsigned long old_size;
+	unsigned long new_size;
+	int binary:1;
 } *items;
 
 
 static void print_fileinfo(struct fileinfo *info)
 {
 	char *class;
 
 	switch (info->status) {
 	case DIFF_STATUS_ADDED:
 		class = "add";
 		break;
 	case DIFF_STATUS_COPIED:
 		class = "cpy";
 		break;
 	case DIFF_STATUS_DELETED:
 		class = "del";
 		break;
 	case DIFF_STATUS_MODIFIED:
 		class = "upd";
 		break;
 	case DIFF_STATUS_RENAMED:
 		class = "mov";
 		break;
 	case DIFF_STATUS_TYPE_CHANGED:
 		class = "typ";
 		break;
 	case DIFF_STATUS_UNKNOWN:
 		class = "unk";
 		break;
 	case DIFF_STATUS_UNMERGED:
 		class = "stg";
 		break;
 	default:
 		die("bug: unhandled diff status %c", info->status);
 	}
 
 	html("<tr>");
 	htmlf("<td class='mode'>");
 	if (is_null_sha1(info->new_sha1)) {
 		cgit_print_filemode(info->old_mode);
 	} else {
 		cgit_print_filemode(info->new_mode);
 	}
 
 	if (info->old_mode != info->new_mode &&
 	    !is_null_sha1(info->old_sha1) &&
 	    !is_null_sha1(info->new_sha1)) {
 		html("<span class='modechange'>[");
 		cgit_print_filemode(info->old_mode);
 		html("]</span>");
 	}
 	htmlf("</td><td class='%s'>", class);
 	cgit_diff_link(info->new_path, NULL, NULL, ctx.qry.head, ctx.qry.sha1,
 		       ctx.qry.sha2, info->new_path);
 	if (info->status == DIFF_STATUS_COPIED || info->status == DIFF_STATUS_RENAMED)
 		htmlf(" (%s from %s)",
 		      info->status == DIFF_STATUS_COPIED ? "copied" : "renamed",
 		      info->old_path);
 	html("</td><td class='right'>");
+	if (info->binary) {
+		htmlf("bin</td><td class='graph'>%d -> %d bytes",
+		      info->old_size, info->new_size);
+		return;
+	}
 	htmlf("%d", info->added + info->removed);
 	html("</td><td class='graph'>");
 	htmlf("<table summary='file diffstat' width='%d%%'><tr>", (max_changes > 100 ? 100 : max_changes));
 	htmlf("<td class='add' style='width: %.1f%%;'/>",
 	      info->added * 100.0 / max_changes);
 	htmlf("<td class='rem' style='width: %.1f%%;'/>",
 	      info->removed * 100.0 / max_changes);
 	htmlf("<td class='none' style='width: %.1f%%;'/>",
 	      (max_changes - info->removed - info->added) * 100.0 / max_changes);
 	html("</tr></table></td></tr>\n");
 }
 
 static void count_diff_lines(char *line, int len)
 {
 	if (line && (len > 0)) {
 		if (line[0] == '+')
 			lines_added++;
 		else if (line[0] == '-')
 			lines_removed++;
 	}
 }
 
 static void inspect_filepair(struct diff_filepair *pair)
 {
+	int binary = 0;
+	unsigned long old_size = 0;
+	unsigned long new_size = 0;
 	files++;
 	lines_added = 0;
 	lines_removed = 0;
-	cgit_diff_files(pair->one->sha1, pair->two->sha1, count_diff_lines);
+	cgit_diff_files(pair->one->sha1, pair->two->sha1, &old_size, &new_size,
+			&binary, count_diff_lines);
 	if (files >= slots) {
 		if (slots == 0)
 			slots = 4;
 		else
 			slots = slots * 2;
 		items = xrealloc(items, slots * sizeof(struct fileinfo));
 	}
 	items[files-1].status = pair->status;
 	hashcpy(items[files-1].old_sha1, pair->one->sha1);
 	hashcpy(items[files-1].new_sha1, pair->two->sha1);
 	items[files-1].old_mode = pair->one->mode;
 	items[files-1].new_mode = pair->two->mode;
 	items[files-1].old_path = xstrdup(pair->one->path);
 	items[files-1].new_path = xstrdup(pair->two->path);
 	items[files-1].added = lines_added;
 	items[files-1].removed = lines_removed;
+	items[files-1].old_size = old_size;
+	items[files-1].new_size = new_size;
+	items[files-1].binary = binary;
 	if (lines_added + lines_removed > max_changes)
 		max_changes = lines_added + lines_removed;
 	total_adds += lines_added;
 	total_rems += lines_removed;
 }
 
 void cgit_print_diffstat(const unsigned char *old_sha1,
 			 const unsigned char *new_sha1)
 {
 	int i;
 
 	html("<div class='diffstat-header'>");
 	cgit_diff_link("Diffstat", NULL, NULL, ctx.qry.head, ctx.qry.sha1,
 		       ctx.qry.sha2, NULL);
 	html("</div>");
 	html("<table summary='diffstat' class='diffstat'>");
 	max_changes = 0;
 	cgit_diff_tree(old_sha1, new_sha1, inspect_filepair, NULL);
 	for(i = 0; i<files; i++)
 		print_fileinfo(&items[i]);
 	html("</table>");
 	html("<div class='diffstat-summary'>");
 	htmlf("%d files changed, %d insertions, %d deletions",
 	      files, total_adds, total_rems);
 	html("</div>");
 }
 
 
 /*
  * print a single line returned from xdiff
  */
 static void print_line(char *line, int len)
 {
 	char *class = "ctx";
 	char c = line[len-1];
 
 	if (line[0] == '+')
 		class = "add";
 	else if (line[0] == '-')
 		class = "del";
 	else if (line[0] == '@')
 		class = "hunk";
 
 	htmlf("<div class='%s'>", class);
 	line[len-1] = '\0';
 	html_txt(line);
 	html("</div>");
 	line[len-1] = c;
 }
 
 static void header(unsigned char *sha1, char *path1, int mode1,
 		   unsigned char *sha2, char *path2, int mode2)
 {
 	char *abbrev1, *abbrev2;
 	int subproject;
 
 	subproject = (S_ISGITLINK(mode1) || S_ISGITLINK(mode2));
 	html("<div class='head'>");
 	html("diff --git a/");
 	html_txt(path1);
 	html(" b/");
 	html_txt(path2);
 
 	if (is_null_sha1(sha1))
 		path1 = "dev/null";
 	if (is_null_sha1(sha2))
 		path2 = "dev/null";
 
 	if (mode1 == 0)
 		htmlf("<br/>new file mode %.6o", mode2);
 
 	if (mode2 == 0)
 		htmlf("<br/>deleted file mode %.6o", mode1);
 
 	if (!subproject) {
 		abbrev1 = xstrdup(find_unique_abbrev(sha1, DEFAULT_ABBREV));
 		abbrev2 = xstrdup(find_unique_abbrev(sha2, DEFAULT_ABBREV));
 		htmlf("<br/>index %s..%s", abbrev1, abbrev2);
 		free(abbrev1);
 		free(abbrev2);
 		if (mode1 != 0 && mode2 != 0) {
 			htmlf(" %.6o", mode1);
 			if (mode2 != mode1)
 				htmlf("..%.6o", mode2);
 		}
 		html("<br/>--- a/");
 		if (mode1 != 0)
 			cgit_tree_link(path1, NULL, NULL, ctx.qry.head,
 				       sha1_to_hex(old_rev_sha1), path1);
 		else
 			html_txt(path1);
 		html("<br/>+++ b/");
 		if (mode2 != 0)
 			cgit_tree_link(path2, NULL, NULL, ctx.qry.head,
 				       sha1_to_hex(new_rev_sha1), path2);
 		else
 			html_txt(path2);
 	}
 	html("</div>");
 }
 
 static void filepair_cb(struct diff_filepair *pair)
 {
+	unsigned long old_size = 0;
+	unsigned long new_size = 0;
+	int binary = 0;
+
 	header(pair->one->sha1, pair->one->path, pair->one->mode,
 	       pair->two->sha1, pair->two->path, pair->two->mode);
 	if (S_ISGITLINK(pair->one->mode) || S_ISGITLINK(pair->two->mode)) {
 		if (S_ISGITLINK(pair->one->mode))
 			print_line(fmt("-Subproject %s", sha1_to_hex(pair->one->sha1)), 52);
 		if (S_ISGITLINK(pair->two->mode))
 			print_line(fmt("+Subproject %s", sha1_to_hex(pair->two->sha1)), 52);
 		return;
 	}
-	if (cgit_diff_files(pair->one->sha1, pair->two->sha1, print_line))
+	if (cgit_diff_files(pair->one->sha1, pair->two->sha1, &old_size, 
+			    &new_size, &binary, print_line))
 		cgit_print_error("Error running diff");
+	if (binary)
+		html("Binary files differ");
 }
 
 void cgit_print_diff(const char *new_rev, const char *old_rev, const char *prefix)
 {
 	enum object_type type;
 	unsigned long size;
 	struct commit *commit, *commit2;
 
 	if (!new_rev)
 		new_rev = ctx.qry.head;
 	get_sha1(new_rev, new_rev_sha1);
 	type = sha1_object_info(new_rev_sha1, &size);
 	if (type == OBJ_BAD) {
 		cgit_print_error(fmt("Bad object name: %s", new_rev));
 		return;
 	}
 	commit = lookup_commit_reference(new_rev_sha1);
 	if (!commit || parse_commit(commit))
 		cgit_print_error(fmt("Bad commit: %s", sha1_to_hex(new_rev_sha1)));
 
 	if (old_rev)
 		get_sha1(old_rev, old_rev_sha1);
 	else if (commit->parents && commit->parents->item)
 		hashcpy(old_rev_sha1, commit->parents->item->object.sha1);
 	else
 		hashclr(old_rev_sha1);
 
 	if (!is_null_sha1(old_rev_sha1)) {
 		type = sha1_object_info(old_rev_sha1, &size);
 		if (type == OBJ_BAD) {
 			cgit_print_error(fmt("Bad object name: %s", sha1_to_hex(old_rev_sha1)));
 			return;
 		}
 		commit2 = lookup_commit_reference(old_rev_sha1);
 		if (!commit2 || parse_commit(commit2))
 			cgit_print_error(fmt("Bad commit: %s", sha1_to_hex(old_rev_sha1)));
 	}
 	cgit_print_diffstat(old_rev_sha1, new_rev_sha1);
 
 	html("<table summary='diff' class='diff'>");
 	html("<tr><td>");
 	cgit_diff_tree(old_rev_sha1, new_rev_sha1, filepair_cb, prefix);
 	html("</td></tr>");
 	html("</table>");
 }
diff --git a/ui-log.c b/ui-log.c
index 3202848..ba2ab03 100644
--- a/ui-log.c
+++ b/ui-log.c
@@ -1,223 +1,228 @@
 /* ui-log.c: functions for log output
  *
  * Copyright (C) 2006 Lars Hjemli
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
  */
 
 #include "cgit.h"
 #include "html.h"
 #include "ui-shared.h"
 
 int files, add_lines, rem_lines;
 
 void count_lines(char *line, int size)
 {
 	if (size <= 0)
 		return;
 
 	if (line[0] == '+')
 		add_lines++;
 
 	else if (line[0] == '-')
 		rem_lines++;
 }
 
 void inspect_files(struct diff_filepair *pair)
 {
+	unsigned long old_size = 0;
+	unsigned long new_size = 0;
+	int binary = 0;
+
 	files++;
 	if (ctx.repo->enable_log_linecount)
-		cgit_diff_files(pair->one->sha1, pair->two->sha1, count_lines);
+		cgit_diff_files(pair->one->sha1, pair->two->sha1, &old_size,
+				&new_size, &binary, count_lines);
 }
 
 void show_commit_decorations(struct commit *commit)
 {
 	struct name_decoration *deco;
 	static char buf[1024];
 
 	buf[sizeof(buf) - 1] = 0;
 	deco = lookup_decoration(&name_decoration, &commit->object);
 	while (deco) {
 		if (!prefixcmp(deco->name, "refs/heads/")) {
 			strncpy(buf, deco->name + 11, sizeof(buf) - 1);
 			cgit_log_link(buf, NULL, "branch-deco", buf, NULL, NULL,
 				0, NULL, NULL, ctx.qry.showmsg);
 		}
 		else if (!prefixcmp(deco->name, "tag: refs/tags/")) {
 			strncpy(buf, deco->name + 15, sizeof(buf) - 1);
 			cgit_tag_link(buf, NULL, "tag-deco", ctx.qry.head, buf);
 		}
 		else if (!prefixcmp(deco->name, "refs/remotes/")) {
 			strncpy(buf, deco->name + 13, sizeof(buf) - 1);
 			cgit_log_link(buf, NULL, "remote-deco", NULL,
 				sha1_to_hex(commit->object.sha1), NULL,
 				0, NULL, NULL, ctx.qry.showmsg);
 		}
 		else {
 			strncpy(buf, deco->name, sizeof(buf) - 1);
 			cgit_commit_link(buf, NULL, "deco", ctx.qry.head,
 				sha1_to_hex(commit->object.sha1));
 		}
 		deco = deco->next;
 	}
 }
 
 void print_commit(struct commit *commit)
 {
 	struct commitinfo *info;
 	char *tmp;
 	int cols = 2;
 
 	info = cgit_parse_commit(commit);
 	htmlf("<tr%s><td>",
 		ctx.qry.showmsg ? " class='logheader'" : "");
 	tmp = fmt("id=%s", sha1_to_hex(commit->object.sha1));
 	tmp = cgit_pageurl(ctx.repo->url, "commit", tmp);
 	html_link_open(tmp, NULL, NULL);
 	cgit_print_age(commit->date, TM_WEEK * 2, FMT_SHORTDATE);
 	html_link_close();
 	htmlf("</td><td%s>",
 		ctx.qry.showmsg ? " class='logsubject'" : "");
 	cgit_commit_link(info->subject, NULL, NULL, ctx.qry.head,
 			 sha1_to_hex(commit->object.sha1));
 	show_commit_decorations(commit);
 	html("</td><td>");
 	html_txt(info->author);
 	if (ctx.repo->enable_log_filecount) {
 		files = 0;
 		add_lines = 0;
 		rem_lines = 0;
 		cgit_diff_commit(commit, inspect_files);
 		html("</td><td>");
 		htmlf("%d", files);
 		if (ctx.repo->enable_log_linecount) {
 			html("</td><td>");
 			htmlf("-%d/+%d", rem_lines, add_lines);
 		}
 	}
 	html("</td></tr>\n");
 	if (ctx.qry.showmsg) {
 		if (ctx.repo->enable_log_filecount) {
 			cols++;
 			if (ctx.repo->enable_log_linecount)
 				cols++;
 		}
 		htmlf("<tr class='nohover'><td/><td colspan='%d' class='logmsg'>",
 			cols);
 		html_txt(info->msg);
 		html("</td></tr>\n");
 	}
 	cgit_free_commitinfo(info);
 }
 
 static const char *disambiguate_ref(const char *ref)
 {
 	unsigned char sha1[20];
 	const char *longref;
 
 	longref = fmt("refs/heads/%s", ref);
 	if (get_sha1(longref, sha1) == 0)
 		return longref;
 
 	return ref;
 }
 
 void cgit_print_log(const char *tip, int ofs, int cnt, char *grep, char *pattern,
 		    char *path, int pager)
 {
 	struct rev_info rev;
 	struct commit *commit;
 	const char *argv[] = {NULL, NULL, NULL, NULL, NULL};
 	int argc = 2;
 	int i, columns = 3;
 
 	if (!tip)
 		tip = ctx.qry.head;
 
 	argv[1] = disambiguate_ref(tip);
 
 	if (grep && pattern && (!strcmp(grep, "grep") ||
 				!strcmp(grep, "author") ||
 				!strcmp(grep, "committer")))
 		argv[argc++] = fmt("--%s=%s", grep, pattern);
 
 	if (path) {
 		argv[argc++] = "--";
 		argv[argc++] = path;
 	}
 	init_revisions(&rev, NULL);
 	rev.abbrev = DEFAULT_ABBREV;
 	rev.commit_format = CMIT_FMT_DEFAULT;
 	rev.verbose_header = 1;
 	rev.show_root_diff = 0;
 	setup_revisions(argc, argv, &rev, NULL);
 	load_ref_decorations();
 	rev.show_decorations = 1;
 	rev.grep_filter.regflags |= REG_ICASE;
 	compile_grep_patterns(&rev.grep_filter);
 	prepare_revision_walk(&rev);
 
 	if (pager)
 		html("<table class='list nowrap'>");
 
 	html("<tr class='nohover'><th class='left'>Age</th>"
 	      "<th class='left'>Commit message");
 	if (pager) {
 		html(" (");
 		cgit_log_link(ctx.qry.showmsg ? "Collapse" : "Expand", NULL,
 			      NULL, ctx.qry.head, ctx.qry.sha1,
 			      ctx.qry.path, ctx.qry.ofs, ctx.qry.grep,
 			      ctx.qry.search, ctx.qry.showmsg ? 0 : 1);
 		html(")");
 	}
 	html("</th><th class='left'>Author</th>");
 	if (ctx.repo->enable_log_filecount) {
 		html("<th class='left'>Files</th>");
 		columns++;
 		if (ctx.repo->enable_log_linecount) {
 			html("<th class='left'>Lines</th>");
 			columns++;
 		}
 	}
 	html("</tr>\n");
 
 	if (ofs<0)
 		ofs = 0;
 
 	for (i = 0; i < ofs && (commit = get_revision(&rev)) != NULL; i++) {
 		free(commit->buffer);
 		commit->buffer = NULL;
 		free_commit_list(commit->parents);
 		commit->parents = NULL;
 	}
 
 	for (i = 0; i < cnt && (commit = get_revision(&rev)) != NULL; i++) {
 		print_commit(commit);
 		free(commit->buffer);
 		commit->buffer = NULL;
 		free_commit_list(commit->parents);
 		commit->parents = NULL;
 	}
 	if (pager) {
 		htmlf("</table><div class='pager'>",
 		     columns);
 		if (ofs > 0) {
 			cgit_log_link("[prev]", NULL, NULL, ctx.qry.head,
 				      ctx.qry.sha1, ctx.qry.path,
 				      ofs - cnt, ctx.qry.grep,
 				      ctx.qry.search, ctx.qry.showmsg);
 			html("&nbsp;");
 		}
 		if ((commit = get_revision(&rev)) != NULL) {
 			cgit_log_link("[next]", NULL, NULL, ctx.qry.head,
 				      ctx.qry.sha1, ctx.qry.path,
 				      ofs + cnt, ctx.qry.grep,
 				      ctx.qry.search, ctx.qry.showmsg);
 		}
 		html("</div>");
 	} else if ((commit = get_revision(&rev)) != NULL) {
 		html("<tr class='nohover'><td colspan='3'>");
 		cgit_log_link("[...]", NULL, NULL, ctx.qry.head, NULL, NULL, 0,
 			      NULL, NULL, ctx.qry.showmsg);
 		html("</td></tr>\n");
diff --git a/ui-patch.c b/ui-patch.c
index 1d77336..5d665d3 100644
--- a/ui-patch.c
+++ b/ui-patch.c
@@ -1,118 +1,125 @@
 /* ui-patch.c: generate patch view
  *
  * Copyright (C) 2007 Lars Hjemli
  *
  * Licensed under GNU General Public License v2
  *   (see COPYING for full license text)
  */
 
 #include "cgit.h"
 #include "html.h"
 #include "ui-shared.h"
 
 static void print_line(char *line, int len)
 {
 	char c = line[len-1];
 
 	line[len-1] = '\0';
 	htmlf("%s\n", line);
 	line[len-1] = c;
 }
 
 static void header(unsigned char *sha1, char *path1, int mode1,
 		   unsigned char *sha2, char *path2, int mode2)
 {
 	char *abbrev1, *abbrev2;
 	int subproject;
 
 	subproject = (S_ISGITLINK(mode1) || S_ISGITLINK(mode2));
 	htmlf("diff --git a/%s b/%s\n", path1, path2);
 
 	if (is_null_sha1(sha1))
 		path1 = "dev/null";
 	if (is_null_sha1(sha2))
 		path2 = "dev/null";
 
 	if (mode1 == 0)
 		htmlf("new file mode %.6o\n", mode2);
 
 	if (mode2 == 0)
 		htmlf("deleted file mode %.6o\n", mode1);
 
 	if (!subproject) {
 		abbrev1 = xstrdup(find_unique_abbrev(sha1, DEFAULT_ABBREV));
 		abbrev2 = xstrdup(find_unique_abbrev(sha2, DEFAULT_ABBREV));
 		htmlf("index %s..%s", abbrev1, abbrev2);
 		free(abbrev1);
 		free(abbrev2);
 		if (mode1 != 0 && mode2 != 0) {
 			htmlf(" %.6o", mode1);
 			if (mode2 != mode1)
 				htmlf("..%.6o", mode2);
 		}
 		htmlf("\n--- a/%s\n", path1);
 		htmlf("+++ b/%s\n", path2);
 	}
 }
 
 static void filepair_cb(struct diff_filepair *pair)
 {
+	unsigned long old_size = 0;
+	unsigned long new_size = 0;
+	int binary = 0;
+
 	header(pair->one->sha1, pair->one->path, pair->one->mode,
 	       pair->two->sha1, pair->two->path, pair->two->mode);
 	if (S_ISGITLINK(pair->one->mode) || S_ISGITLINK(pair->two->mode)) {
 		if (S_ISGITLINK(pair->one->mode))
 			print_line(fmt("-Subproject %s", sha1_to_hex(pair->one->sha1)), 52);
 		if (S_ISGITLINK(pair->two->mode))
 			print_line(fmt("+Subproject %s", sha1_to_hex(pair->two->sha1)), 52);
 		return;
 	}
-	if (cgit_diff_files(pair->one->sha1, pair->two->sha1, print_line))
+	if (cgit_diff_files(pair->one->sha1, pair->two->sha1, &old_size,
+			    &new_size, &binary, print_line))
 		html("Error running diff");
+	if (binary)
+		html("Binary files differ\n");
 }
 
 void cgit_print_patch(char *hex)
 {
 	struct commit *commit;
 	struct commitinfo *info;
 	unsigned char sha1[20], old_sha1[20];
 	char *patchname;
 
 	if (!hex)
 		hex = ctx.qry.head;
 
 	if (get_sha1(hex, sha1)) {
 		cgit_print_error(fmt("Bad object id: %s", hex));
 		return;
 	}
 	commit = lookup_commit_reference(sha1);
 	if (!commit) {
 		cgit_print_error(fmt("Bad commit reference: %s", hex));
 		return;
 	}
 	info = cgit_parse_commit(commit);
 
 	if (commit->parents && commit->parents->item)
 		hashcpy(old_sha1, commit->parents->item->object.sha1);
 	else
 		hashclr(old_sha1);
 
 	patchname = fmt("%s.patch", sha1_to_hex(sha1));
 	ctx.page.mimetype = "text/plain";
 	ctx.page.filename = patchname;
 	cgit_print_http_headers(&ctx);
 	htmlf("From %s Mon Sep 17 00:00:00 2001\n", sha1_to_hex(sha1));
 	htmlf("From: %s %s\n", info->author, info->author_email);
 	html("Date: ");
 	cgit_print_date(info->author_date, "%a, %d %b %Y %H:%M:%S %z%n", ctx.cfg.local_time);
 	htmlf("Subject: %s\n\n", info->subject);
 	if (info->msg && *info->msg) {
 		htmlf("%s", info->msg);
 		if (info->msg[strlen(info->msg) - 1] != '\n')
 			html("\n");
 	}
 	html("---\n");
 	cgit_diff_tree(old_sha1, sha1, filepair_cb, NULL);
 	html("--\n");
 	htmlf("cgit %s\n", CGIT_VERSION);
 	cgit_free_commitinfo(info);
 }