vcs-svn: allow import of > 4GiB files
There is no reason in principle that an svn-format dump would not be able to represent a file whose length does not fit in a 32-bit integer. Use off_t consistently (instead of uint32_t) to represent file lengths so we can handle that. Most of our code is already ready to do that without this patch and already passes values of type off_t around. The type mismatch due to stragglers was noticed with gcc -Wtype-limits. Inspired-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Jonathan Nieder committed
Feb 2, 2012 at 04:19 UTC
2d54b9ea8b99df7b448f64a7847f7f3879964e12
3 files changed
+26
-14
vcs-svn/fast_export.c
+9
-6
@@ -227,15 +227,18 @@ static long apply_delta(off_t len, struct line_buffer *input,
227
return ret;
228
}
229
230
-void fast_export_data(uint32_t mode, uint32_t len, struct line_buffer *input)
230
+void fast_export_data(uint32_t mode, off_t len, struct line_buffer *input)
231
{
232
+ assert(len >= 0);
233
if (mode == REPO_MODE_LNK) {
234
/* svn symlink blobs start with "link " */
235
+ if (len < 5)
236
+ die("invalid dump: symlink too short for \"link\" prefix");
237
len -= 5;
238
if (buffer_skip_bytes(input, 5) != 5)
239
die_short_read(input);
240
}
238
- printf("data %"PRIu32"\n", len);
241
+ printf("data %"PRIuMAX"\n", (uintmax_t) len);
242
if (buffer_copy_bytes(input, len) != len)
243
die_short_read(input);
244
fputc('\n', stdout);
@@ -297,12 +300,12 @@ int fast_export_ls(const char *path, uint32_t *mode, struct strbuf *dataref)
300
301
void fast_export_blob_delta(uint32_t mode,
302
uint32_t old_mode, const char *old_data,
300
- uint32_t len, struct line_buffer *input)
303
+ off_t len, struct line_buffer *input)
304
{
305
long postimage_len;
303
- if (len > maximum_signed_value_of_type(off_t))
304
- die("enormous delta");
305
- postimage_len = apply_delta((off_t) len, input, old_data, old_mode);
306
+
307
+ assert(len >= 0);
308
+ postimage_len = apply_delta(len, input, old_data, old_mode);
309
if (mode == REPO_MODE_LNK) {
310
buffer_skip_bytes(&postimage, strlen("link "));
311
postimage_len -= strlen("link ");
vcs-svn/fast_export.h
+2
-2
@@ -14,10 +14,10 @@ void fast_export_begin_commit(uint32_t revision, const char *author,
14
const struct strbuf *log, const char *uuid,
15
const char *url, unsigned long timestamp);
16
void fast_export_end_commit(uint32_t revision);
17
-void fast_export_data(uint32_t mode, uint32_t len, struct line_buffer *input);
17
+void fast_export_data(uint32_t mode, off_t len, struct line_buffer *input);
18
void fast_export_blob_delta(uint32_t mode,
19
uint32_t old_mode, const char *old_data,
20
- uint32_t len, struct line_buffer *input);
20
+ off_t len, struct line_buffer *input);
21
22
/* If there is no such file at that rev, returns -1, errno == ENOENT. */
23
int fast_export_ls_rev(uint32_t rev, const char *path,
vcs-svn/svndump.c
+15
-6
@@ -40,7 +40,8 @@
40
static struct line_buffer input = LINE_BUFFER_INIT;
41
42
static struct {
43
- uint32_t action, propLength, textLength, srcRev, type;
43
+ uint32_t action, propLength, srcRev, type;
44
+ off_t text_length;
45
struct strbuf src, dst;
46
uint32_t text_delta, prop_delta;
47
} node_ctx;
@@ -61,7 +62,7 @@ static void reset_node_ctx(char *fname)
62
node_ctx.type = 0;
63
node_ctx.action = NODEACT_UNKNOWN;
64
node_ctx.propLength = LENGTH_UNKNOWN;
64
- node_ctx.textLength = LENGTH_UNKNOWN;
65
+ node_ctx.text_length = -1;
66
strbuf_reset(&node_ctx.src);
67
node_ctx.srcRev = 0;
68
strbuf_reset(&node_ctx.dst);
@@ -209,7 +210,7 @@ static void handle_node(void)
210
{
211
const uint32_t type = node_ctx.type;
212
const int have_props = node_ctx.propLength != LENGTH_UNKNOWN;
212
- const int have_text = node_ctx.textLength != LENGTH_UNKNOWN;
213
+ const int have_text = node_ctx.text_length != -1;
214
/*
215
* Old text for this node:
216
* NULL - directory or bug
@@ -291,12 +292,12 @@ static void handle_node(void)
292
}
293
if (!node_ctx.text_delta) {
294
fast_export_modify(node_ctx.dst.buf, node_ctx.type, "inline");
294
- fast_export_data(node_ctx.type, node_ctx.textLength, &input);
295
+ fast_export_data(node_ctx.type, node_ctx.text_length, &input);
296
return;
297
}
298
fast_export_modify(node_ctx.dst.buf, node_ctx.type, "inline");
299
fast_export_blob_delta(node_ctx.type, old_mode, old_data,
299
- node_ctx.textLength, &input);
300
+ node_ctx.text_length, &input);
301
}
302
303
static void begin_revision(void)
@@ -409,7 +410,15 @@ void svndump_read(const char *url)
410
break;
411
case sizeof("Text-content-length"):
412
if (!constcmp(t, "Text-content-length")) {
412
- node_ctx.textLength = atoi(val);
413
+ char *end;
414
+ uintmax_t textlen;
415
+
416
+ textlen = strtoumax(val, &end, 10);
417
+ if (!isdigit(*val) || *end)
418
+ die("invalid dump: non-numeric length %s", val);
419
+ if (textlen > maximum_signed_value_of_type(off_t))
420
+ die("unrepresentable length in dump: %s", val);
421
+ node_ctx.text_length = (off_t) textlen;
422
break;
423
}
424
if (constcmp(t, "Prop-content-length"))