Offer blame only where it works

The blame page answers 404 for a symlink, since git blames a regular
file alone, and it refuses a binary file and one past max-blob-size,
yet a tree listing offered blame for all of them, and so did a
symlink's own page. Each row already knows its size, and git calls a
file binary by its first 8000 bytes alone, which a stream reads
without inflating the rest. With blame on, a listing of 2,000
delta-packed files went from about 11 to 15 ms.

1 file changed, 37 insertions, 4 deletions

diff --git a/source/ui-tree.c b/source/ui-tree.cindex 8e17cc4..63b8984 100644--- a/source/ui-tree.c+++ b/source/ui-tree.c@@ -17,6 +17,8 @@ #include "ui-shared.h" #include "ui-tree.h" +#include <odb/streaming.h>+ #define HEXDUMP_ROW_BYTES 32 #define HEXDUMP_ROW_HALF (HEXDUMP_ROW_BYTES / 2) #define HEXDUMP_GAP_WIDTH 4@@ -25,6 +27,10 @@ // max-blob-size would make a page several times its own size. #define HEXDUMP_MAX_BYTES (64 * 1024) +// How far into a buffer buffer_is_binary looks, the window git judges a file+// binary by.+#define BINARY_SNIFF_BYTES 8000+ // How many single-child directories a listing row follows into one path. #define MAX_DIR_CHAIN_LEVELS 15 @@ -147,7 +153,7 @@ static void print_binary_buffer(char *buf, size_t size, const char *path, const  * false means a complete error page went out in place of the blob, so nothing  * may be added to it.  */-static bool print_object(const struct object_id *oid, const char *path,+static bool print_object(const struct object_id *oid, unsigned mode, const char *path, 	const char *filename, const char *rev) { 	enum object_type type;@@ -182,7 +188,7 @@ static bool print_object(const struct object_id *oid, const char *path, 	cgit_print_layout_start(); 	htmlf("<div class='blob-header'>blob: %s (", oid_to_hex(oid)); 	cgit_plain_link("plain", NULL, NULL, ctx.qry.head, rev, path);-	if (ctx.repo->enable_blame && !is_binary) {+	if (ctx.repo->enable_blame && S_ISREG(mode) && !is_binary) { 		html(") ("); 		cgit_blame_link("blame", NULL, NULL, ctx.qry.head, rev, path); 	}@@ -259,6 +265,26 @@ static void print_dir_chain(const struct object_id *oid, const char *name, 	strbuf_setlen(fullpath, initial_length); } +/*+ * Whether git counts the blob as binary, read through a stream so that only+ * the bytes buffer_is_binary looks at are inflated, however large the blob.+ */+static bool blob_starts_binary(const struct object_id *oid)+{+	char buf[BINARY_SNIFF_BYTES];+	struct odb_stream *stream;+	size_t got = 0;+	ssize_t n;++	stream = odb_stream_from_object(the_repository->objects, oid, NULL);+	if (!stream)+		return false;+	while (got < sizeof(buf) && (n = odb_stream_read(stream, buf + got, sizeof(buf) - got)) > 0)+		got += n;+	odb_stream_close(stream);+	return buffer_is_binary(buf, got);+}+ static void print_ls_row(const struct object_id *oid, const char *name, 	unsigned mode, struct walk_tree_context *walk) {@@ -345,7 +371,14 @@ static void print_ls_row(const struct object_id *oid, const char *name, 		cgit_stats_link("stats", NULL, "button", ctx.qry.head, fullpath.buf); 	if (!S_ISGITLINK(mode)) 		cgit_plain_link("plain", NULL, "button", ctx.qry.head, walk->rev, fullpath.buf);-	if (!S_ISDIR(mode) && !S_ISGITLINK(mode) && ctx.repo->enable_blame)+	// The blame page refuses a binary file and one past max-blob-size, so+	// neither is offered it.+	if (+		S_ISREG(mode)              &&+		ctx.repo->enable_blame     &&+		!cgit_blob_too_large(size) &&+		!blob_starts_binary(oid)+	) 		cgit_blame_link("blame", NULL, "button", ctx.qry.head, walk->rev, fullpath.buf); 	html("</td></tr>\n"); @@ -438,7 +471,7 @@ static int walk_tree(const struct object_id *oid, struct strbuf *base, 			ls_head(); 			return READ_TREE_RECURSIVE; 		} else {-			bool shown = print_object(oid, fullpath.buf, pathname, walk->rev);+			bool shown = print_object(oid, mode, fullpath.buf, pathname, walk->rev);  			walk->state = shown ? WALK_BLOB_SHOWN : WALK_ERROR_SHOWN; 			strbuf_release(&fullpath);
diff --git a/source/ui-tree.c b/source/ui-tree.c
index 8e17cc4..63b8984 100644
--- a/source/ui-tree.c
+++ b/source/ui-tree.c
@@ -17,6 +17,8 @@
 #include "ui-shared.h"
 #include "ui-tree.h"
 
+#include <odb/streaming.h>
+
 #define HEXDUMP_ROW_BYTES 32
 #define HEXDUMP_ROW_HALF (HEXDUMP_ROW_BYTES / 2)
 #define HEXDUMP_GAP_WIDTH 4
@@ -25,6 +27,10 @@
 // max-blob-size would make a page several times its own size.
 #define HEXDUMP_MAX_BYTES (64 * 1024)
 
+// How far into a buffer buffer_is_binary looks, the window git judges a file
+// binary by.
+#define BINARY_SNIFF_BYTES 8000
+
 // How many single-child directories a listing row follows into one path.
 #define MAX_DIR_CHAIN_LEVELS 15
 
@@ -147,7 +153,7 @@ static void print_binary_buffer(char *buf, size_t size, const char *path, const
  * false means a complete error page went out in place of the blob, so nothing
  * may be added to it.
  */
-static bool print_object(const struct object_id *oid, const char *path,
+static bool print_object(const struct object_id *oid, unsigned mode, const char *path,
 	const char *filename, const char *rev)
 {
 	enum object_type type;
@@ -182,7 +188,7 @@ static bool print_object(const struct object_id *oid, const char *path,
 	cgit_print_layout_start();
 	htmlf("<div class='blob-header'>blob: %s (", oid_to_hex(oid));
 	cgit_plain_link("plain", NULL, NULL, ctx.qry.head, rev, path);
-	if (ctx.repo->enable_blame && !is_binary) {
+	if (ctx.repo->enable_blame && S_ISREG(mode) && !is_binary) {
 		html(") (");
 		cgit_blame_link("blame", NULL, NULL, ctx.qry.head, rev, path);
 	}
@@ -259,6 +265,26 @@ static void print_dir_chain(const struct object_id *oid, const char *name,
 	strbuf_setlen(fullpath, initial_length);
 }
 
+/*
+ * Whether git counts the blob as binary, read through a stream so that only
+ * the bytes buffer_is_binary looks at are inflated, however large the blob.
+ */
+static bool blob_starts_binary(const struct object_id *oid)
+{
+	char buf[BINARY_SNIFF_BYTES];
+	struct odb_stream *stream;
+	size_t got = 0;
+	ssize_t n;
+
+	stream = odb_stream_from_object(the_repository->objects, oid, NULL);
+	if (!stream)
+		return false;
+	while (got < sizeof(buf) && (n = odb_stream_read(stream, buf + got, sizeof(buf) - got)) > 0)
+		got += n;
+	odb_stream_close(stream);
+	return buffer_is_binary(buf, got);
+}
+
 static void print_ls_row(const struct object_id *oid, const char *name,
 	unsigned mode, struct walk_tree_context *walk)
 {
@@ -345,7 +371,14 @@ static void print_ls_row(const struct object_id *oid, const char *name,
 		cgit_stats_link("stats", NULL, "button", ctx.qry.head, fullpath.buf);
 	if (!S_ISGITLINK(mode))
 		cgit_plain_link("plain", NULL, "button", ctx.qry.head, walk->rev, fullpath.buf);
-	if (!S_ISDIR(mode) && !S_ISGITLINK(mode) && ctx.repo->enable_blame)
+	// The blame page refuses a binary file and one past max-blob-size, so
+	// neither is offered it.
+	if (
+		S_ISREG(mode)              &&
+		ctx.repo->enable_blame     &&
+		!cgit_blob_too_large(size) &&
+		!blob_starts_binary(oid)
+	)
 		cgit_blame_link("blame", NULL, "button", ctx.qry.head, walk->rev, fullpath.buf);
 	html("</td></tr>\n");
 
@@ -438,7 +471,7 @@ static int walk_tree(const struct object_id *oid, struct strbuf *base,
 			ls_head();
 			return READ_TREE_RECURSIVE;
 		} else {
-			bool shown = print_object(oid, fullpath.buf, pathname, walk->rev);
+			bool shown = print_object(oid, mode, fullpath.buf, pathname, walk->rev);
 
 			walk->state = shown ? WALK_BLOB_SHOWN : WALK_ERROR_SHOWN;
 			strbuf_release(&fullpath);