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

Re: RFC: Flat directory for notes, or fan-out? Both!

From
Johannes Schindelin <johannes.schindelin@gmx.de>
Date
Feb 11, 2009, 20:57 UTC
Message-ID
<alpine.DEB.1.00.0902112152270.10279@pacific.mpi-cbg.de>
In-Reply-To
<20090211200227.GA27961@coredump.intra.peff.net>
Hi,
On Wed, 11 Feb 2009, Jeff King wrote:
Show 13 quoted lines
> On Tue, Feb 10, 2009 at 10:10:43PM +0100, Johannes Schindelin wrote:
> 
> > > I wonder if we can solve this by introducing a local cache that is a flat
> > > file that looks like:
> > [...]
> > Or we could use an on-disk hashmap.  Oh, wait...
> 
> That was my first thought, as well. Maybe your original implementation
> wasn't so bad, after all. :)
> 
> I searched through the archive to find a list of criticisms, but I
> didn't see any. So I guess the problem was just a concern that it might
> end up too complex.
Nope, the issue was that it would take too long to recreate IIRC.

BTW I am no longer a fan of the on-disk cache; I think it is an ugly solution to a problem that should be solved without ugliness using a flexible directory layout in the note ref' tree.

I mean, we really can allow different directory layouts as Sam described, with a few benefits, and only slight downsides. If we support multiple levels anyway, the code to allow arbitrary splits is not complicated (see Sam's patch).

Even the merging should not pose any problem at all; we need a custom merge driver anyway, and there is no reason whatsoever why we should not just teach the merge driver to remove the slashes before comparing the filie names.

At edit time, we can afford to perform a check that is a little more expensive than it would have been otherwise.

Ciao, Dscho

Previous: Jeff KingNext: Junio C Hamano
Message 26 of 33 in “RFC: Flat directory for notes, or fan-out? Both!”
  1. Johannes SchindelinFeb 9, 2009
  2. Boyd Stephen Smith Jr.Feb 10, 2009
  3. Jeff KingFeb 10, 2009
  4. Boyd Stephen Smith Jr.Feb 11, 2009
  5. Linus TorvaldsFeb 11, 2009
  6. Sam VilainFeb 11, 2009
  7. Linus TorvaldsFeb 11, 2009
  8. Sam VilainFeb 11, 2009
  9. Johannes SchindelinFeb 11, 2009
  10. Jeff KingFeb 10, 2009
  11. Johannes SchindelinFeb 10, 2009
  12. Jeff KingFeb 10, 2009
  13. Johannes SchindelinFeb 10, 2009
  14. Junio C HamanoFeb 10, 2009
  15. Shawn O. PearceFeb 10, 2009
  16. Johannes SchindelinFeb 10, 2009
  17. Shawn O. PearceFeb 10, 2009
  18. Johannes SchindelinFeb 10, 2009
  19. Junio C HamanoFeb 10, 2009
  20. Shawn O. PearceFeb 10, 2009
  21. Johannes SchindelinFeb 10, 2009
  22. Thomas RastFeb 10, 2009
  23. Thomas RastFeb 10, 2009
  24. Junio C HamanoFeb 10, 2009
  25. Jeff KingFeb 11, 2009
  26. Johannes SchindelinFeb 11, 2009
  27. Junio C HamanoFeb 11, 2009
  28. Johannes SchindelinFeb 11, 2009
  29. Shawn O. PearceFeb 10, 2009
  30. Johannes SchindelinFeb 10, 2009
  31. Shawn O. PearceFeb 10, 2009
  32. Sam VilainFeb 11, 2009
  33. Sam VilainFeb 11, 2009

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.