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

[RFC/PATCH 0/7] user-configurable git-archive output formats

From
Jeff King <peff@github.com>
Date
Jun 15, 2011, 22:30 UTC
Message-ID
<20110615223030.GA16110@sigill.intra.peff.net>
In-Reply-To
<20110614204521.GA12776@sigill.intra.peff.net>
On Tue, Jun 14, 2011 at 04:45:21PM -0400, Jeff King wrote:
Show 8 quoted lines
> The gzip path is not configurable at all. Probably it should read the
> path and arguments from the config file. In fact, we could even allow
> arbitrary config like:
> 
>   [tarfilter "tgz"]
>     command = gzip -c
>     extension = tgz
>     extension = tar.gz

Here's a series implementing that. You can configure whatever you want, and it includes builtin gzip configuration by default. You can override to turn it off, or even switch it to run something like pigz instead.

My biggest reservation with the patches as-is is that they are very tar-centric and not orthogonal. Specifically, they won't handle:

  1. Other streamable archive formats you would want to pipe through
     compressors. Do any of these actually exist? I guess we could offer
     "pax" as a format eventually, and it might be like tar with
     different defaults? I dunno.
     Fixing this would not be too hard. Instead of these being
     "tarfilters", they would be "archive filters", and they would chain
     to some format, defaulting to "tar".  Since there is no other
     format right now, we could even punt on writing most of the code
     until somebody adds one. But we would want to get the naming of the
     config options right, since those are user-facing. Maybe
     "archivefilter" (unfortunately the more readable archive.filter is
     a little awkward with the way we parse config files)?
  2. In theory you might want to plug in external helpers that are not
     just stream filters, but actually their own container formats (like
     zip). I think people who want 7zip would want this.
     But how does git-archive interact with the helper? By definition
     the data it wants is the set of files, not a single stream. So
     either:
       a. We give the helper a temporary exported checkout, and it
          generates the stream from that.
       b. We use tar as the lingua franca of streaming file containers,
          and let the helper deal with converting to its preferred
          output format.
      Option (a) seems horribly inefficient on disk I/O. And if we did
      want to do that, I think it's largely unrelated to this patch
      series.
      You can actually do option (b) with this series. In its worst
      case, you can do the same as (a): just untar into a temporary
      directory and compress from there. But a well-written helper could
      convert tar into the output format on the fly.
The patches are:
  [1/7]: archive: reorder option parsing and config reading
  [2/7]: archive: add user-configurable tar-filter infrastructure
  [3/7]: archive: support user tar-filters via --format
  [4/7]: archive: advertise user tar-filters in --list
  [5/7]: archive: refactor format-guessing from filename
  [6/7]: archive: match extensions from user-configured formats
  [7/7]: archive: provide builtin .tar.gz filter
-Peff
Previous: Jeff KingNext: Jeff King
Message 8 of 56 in “archive: factor out write phase of tar format”
  1. 1/2 archive: factor out write phase of tar formatJeff King, Jun 14, 2011
  2. 2/2 archive: support gzipped tar filesJeff King, Jun 14, 2011
  3. J.H.Jun 14, 2011
  4. Jeff KingJun 14, 2011
  5. René ScharfeJun 14, 2011
  6. Jeff KingJun 14, 2011
  7. Jeff KingJun 14, 2011
  8. 0/7 user-configurable git-archive output formatsJeff King, Jun 15, 2011
  9. 1/7 archive: reorder option parsing and config readingJeff King, Jun 15, 2011
  10. 2/7 archive: add user-configurable tar-filter infrastructureJeff King, Jun 15, 2011
  11. Junio C HamanoJun 15, 2011
  12. Jeff KingJun 16, 2011
  13. 3/7 archive: support user tar-filters via --formatJeff King, Jun 15, 2011
  14. 4/7 archive: advertise user tar-filters in --listJeff King, Jun 15, 2011
  15. 5/7 archive: refactor format-guessing from filenameJeff King, Jun 15, 2011
  16. Junio C HamanoJun 15, 2011
  17. Jeff KingJun 16, 2011
  18. 6/7 archive: match extensions from user-configured formatsJeff King, Jun 15, 2011
  19. 7/7 archive: provide builtin .tar.gz filterJeff King, Jun 15, 2011
  20. Junio C HamanoJun 15, 2011
  21. Junio C HamanoJun 15, 2011
  22. Jeff KingJun 16, 2011
  23. Junio C HamanoJun 16, 2011
  24. Jeff KingJun 16, 2011
  25. Chris WebbJun 16, 2011
  26. Jeff KingJun 16, 2011
  27. Junio C HamanoJun 16, 2011
  28. Jeff KingJun 16, 2011
  29. John SzakmeisterJun 16, 2011
  30. Junio C HamanoJun 16, 2011
  31. Jeff KingJun 16, 2011
  32. René ScharfeJun 18, 2011
  33. Jakub NarebskiJun 18, 2011
  34. Junio C HamanoJun 20, 2011
  35. 0/9 configurable tar compressorsJeff King, Jun 22, 2011
  36. 1/9 archive: reorder option parsing and config readingJeff King, Jun 22, 2011
  37. 2/9 archive-tar: don't reload default config optionsJeff King, Jun 22, 2011
  38. 3/9 archive: refactor list of archive formatsJeff King, Jun 22, 2011
  39. Thiago FarinaJun 23, 2011
  40. Jeff KingJun 23, 2011
  41. 4/9 archive: pass archiver struct to write_archive callbackJeff King, Jun 22, 2011
  42. 5/9 archive: move file extension format-guessing lowerJeff King, Jun 22, 2011
  43. 6/9 archive: refactor file extension format-guessingJeff King, Jun 22, 2011
  44. 7/9 archive: implement configurable tar filtersJeff King, Jun 22, 2011
  45. Jeff KingJun 22, 2011
  46. René ScharfeJun 22, 2011
  47. Jeff KingJun 22, 2011
  48. 8/9 archive: provide builtin .tar.gz filterJeff King, Jun 22, 2011
  49. 9/9 upload-archive: allow user to turn off filtersJeff King, Jun 22, 2011
  50. Jeff KingJun 22, 2011
  51. Jeff KingJun 21, 2011
  52. René ScharfeJun 18, 2011
  53. Junio C HamanoJun 14, 2011
  54. Jeff KingJun 14, 2011
  55. Miles BaderJun 14, 2011
  56. Jeff KingJun 15, 2011

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.