[Commented On] D9789: clone: make sure we warm the cache after a clone

[indygreg (Gregory Szorc)]

indygreg added a comment.


  I'm `-0` on this change. Caches are caches and IMO should only be populated on demand.
  
  My specific complaint is I worry about performance implications here. There are likely automated processes (e.g. in CI) that clone a repo and don't perform any operation that touches the cache. Populating the cache could introduce non-trivial, avoidable overhead.
  
  Was any performance testing performed on this patch?

INLINE COMMENTS

> test-tags.t:735-744
>  Cache should contain the head only, even though other nodes have tags data
>  
>    $ f --size --hexdump tagsclient/.hg/cache/hgtagsfnodes1
>    tagsclient/.hg/cache/hgtagsfnodes1: size=96
> -  0000: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
> -  0010: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
> -  0020: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
> -  0030: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
> -  0040: ff ff ff ff ff ff ff ff 40 f0 35 8c 19 e0 a7 d3 |........ at .5.....|
> +  0000: 96 ee 1d 73 00 00 00 00 00 00 00 00 00 00 00 00 |...s............|
> +  0010: 00 00 00 00 00 00 00 00 c4 da b0 c2 94 65 e1 c6 |.............e..|
> +  0020: 0d f7 f0 dd 32 04 ea 57 78 c8 97 97 79 fc d5 95 |....2..Wx...y...|

The test description is now inconsistent with reality. Please update the description and consider deleting redundant tests.

REPOSITORY
  rHG Mercurial

CHANGES SINCE LAST ACTION
  https://phab.mercurial-scm.org/D9789/new/

REVISION DETAIL
  https://phab.mercurial-scm.org/D9789

To: marmoute, #hg-reviewers, joerg.sonnenberger, mharbison72
Cc: indygreg, joerg.sonnenberger, martinvonz, mercurial-patches
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mercurial-scm.org/pipermail/mercurial-patches/attachments/20210124/f8a6ec1a/attachment-0002.html>