git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH v4 4/5] archive-tar: write extended headers for far-future mtime

From
Jeff King <peff@peff.net>
Date
Jun 30, 2016, 09:09 UTC
Message-ID
<20160630090920.GD17463@sigill.intra.peff.net>
In-Reply-To
<20160630090614.GA16725@sigill.intra.peff.net>

The ustar format represents timestamps as seconds since the epoch, but only has room to store 11 octal digits. To express anything larger, we need to use an extended header. This is exactly the same case we fixed for the size field in the previous commit, and the solution here follows the same pattern.

This is even mentioned as an issue in f2f0267 (archive-tar: use xsnprintf for trivial formatting, 2015-09-24), but since it only affected things far in the future, it wasn't deemed worth dealing with. But note that my calculations claiming thousands of years were off there; because our xsnprintf produces a NUL byte, we only have until the year 2242 to fix this.

Given that this is just around the corner (geologically speaking, anyway), and because it's easy to fix, let's just make it work. Unlike the previous fix for "size", where we had to write an individual extended header for each file, we can write one global header (since we have only one mtime for the whole archive).

There's a slight bit of trickiness there. We may already be writing a global header with a "comment" field for the commit sha1. So we need to write our new field into the same header. To do this, we push the decision of whether to write such a header down into write_global_extended_header(), which will now assemble the header as it sees fit, and will return early if we have nothing to write (in practice, we'll only have a large mtime if it comes from a commit, but this makes it also work if you set your system clock ahead such that time() returns a huge value).

Note that we don't (and never did) handle negative timestamps (i.e., before 1970). This would probably not be too hard to support in the same way, but since git does not support negative timestamps at all, I didn't bother here.

After writing the extended header, we munge the timestamp in the ustar headers to the maximum-allowable size. This is wrong, but it's the least-wrong thing we can provide to a tar implementation that doesn't understand pax headers (it's also what GNU tar does).

Helped-by: René Scharfe <l.s.r@web.de>
Signed-off-by: Jeff King <peff@peff.net>
---
 archive-tar.c       | 19 ++++++++++++++++---
 t/t5000-tar-tree.sh |  4 ++--
 2 files changed, 18 insertions(+), 5 deletions(-)
diff --git a/archive-tar.c b/archive-tar.c
index 57a1540..d671bc3 100644
--- a/archive-tar.c
+++ b/archive-tar.c
@@ -22,8 +22,11 @@ static int write_tar_filter_archive(const struct archiver *ar,
  * This is the max value that a ustar size header can specify, as it is fixed
  * at 11 octal digits. POSIX specifies that we switch to extended headers at
  * this size.
+ *
+ * Likewise for the mtime (which happens to use a buffer of the same size).
  */
 #define USTAR_MAX_SIZE 077777777777UL
+#define USTAR_MAX_MTIME 077777777777UL
 
 /* writes out the whole block, but only if it is full */
 static void write_if_needed(void)
@@ -324,7 +327,18 @@ static int write_global_extended_header(struct archiver_args *args)
 	unsigned int mode;
 	int err = 0;
 
-	strbuf_append_ext_header(&ext_header, "comment", sha1_to_hex(sha1), 40);
+	if (sha1)
+		strbuf_append_ext_header(&ext_header, "comment",
+					 sha1_to_hex(sha1), 40);
+	if (args->time > USTAR_MAX_MTIME) {
+		strbuf_append_ext_header_uint(&ext_header, "mtime",
+					      args->time);
+		args->time = USTAR_MAX_MTIME;
+	}
+
+	if (!ext_header.len)
+		return 0;
+
 	memset(&header, 0, sizeof(header));
 	*header.typeflag = TYPEFLAG_GLOBAL_HEADER;
 	mode = 0100666;
@@ -409,8 +423,7 @@ static int write_tar_archive(const struct archiver *ar,
 {
 	int err = 0;
 
-	if (args->commit_sha1)
-		err = write_global_extended_header(args);
+	err = write_global_extended_header(args);
 	if (!err)
 		err = write_archive_entries(args, write_tar_entry);
 	if (!err)
diff --git a/t/t5000-tar-tree.sh b/t/t5000-tar-tree.sh
index 93c2d34..96d208d 100755
--- a/t/t5000-tar-tree.sh
+++ b/t/t5000-tar-tree.sh
@@ -383,11 +383,11 @@ test_expect_success 'set up repository with far-future commit' '
 		git commit -m "tempori parendum"
 '
 
-test_expect_failure 'generate tar with future mtime' '
+test_expect_success 'generate tar with future mtime' '
 	git archive HEAD >future.tar
 '
 
-test_expect_failure TAR_HUGE 'system tar can read our future mtime' '
+test_expect_success TAR_HUGE 'system tar can read our future mtime' '
 	echo 4147 >expect &&
 	tar_info future.tar | cut -d" " -f2 >actual &&
 	test_cmp expect actual
-- 
2.9.0.317.g65b4e7c
Previous: Torsten BögershausenNext: Jeff King
Message 8 of 42 in “friendlier handling of overflows in archive-tar”
  1. 0/5 friendlier handling of overflows in archive-tarJeff King, Jun 30, 2016
  2. 5/5 archive-tar: drop return valueJeff King, Jun 30, 2016
  3. 3/5 archive-tar: write extended headers for file sizes >= 8GBJeff King, Jun 30, 2016
  4. Johannes SixtJul 14, 2016
  5. Junio C HamanoJul 14, 2016
  6. Jeff KingJul 14, 2016
  7. Torsten BögershausenJul 15, 2016
  8. 4/5 archive-tar: write extended headers for far-future mtimeJeff King, Jun 30, 2016
  9. 1/5 t9300: factor out portable "head -c" replacementJeff King, Jun 30, 2016
  10. Eric SunshineJul 1, 2016
  11. Junio C HamanoJul 1, 2016
  12. Jeff KingJul 1, 2016
  13. 6/5 t5000: use test_match_signalJeff King, Jun 30, 2016
  14. 2/5 t5000: test tar files that overflow ustar headersJeff King, Jun 30, 2016
  15. Johannes SchindelinJul 14, 2016
  16. Johannes SixtJul 14, 2016
  17. Junio C HamanoJul 14, 2016
  18. Johannes SixtJul 14, 2016
  19. Jeff KingJul 14, 2016
  20. Junio C HamanoJul 14, 2016
  21. Jeff KingJul 14, 2016
  22. Torsten BögershausenJul 15, 2016
  23. Jeff KingJul 15, 2016
  24. Junio C HamanoJul 14, 2016
  25. Jeff KingJul 14, 2016
  26. Jeff KingJul 14, 2016
  27. Junio C HamanoJul 14, 2016
  28. Junio C HamanoJul 14, 2016
  29. Jeff KingJul 14, 2016
  30. Junio C HamanoJul 14, 2016
  31. Jeff KingJul 14, 2016
  32. Junio C HamanoJul 14, 2016
  33. Jeff KingJul 14, 2016
  34. Junio C HamanoJul 14, 2016
  35. 0/2 ulong may only be 32-bit wideJunio C Hamano, Jul 14, 2016
  36. 2/2 archive-tar: huge offset and future timestamps would not work on 32-bitJunio C Hamano, Jul 14, 2016
  37. Jeff KingJul 14, 2016
  38. Junio C HamanoJul 14, 2016
  39. Duy NguyenJul 16, 2016
  40. 1/2 t0006: skip "far in the future" test when unsigned long is not long enoughJunio C Hamano, Jul 14, 2016
  41. Johannes SchindelinJul 15, 2016
  42. Junio C HamanoJul 15, 2016

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.