[Updated] D9275: transaction: drop per-file extra data support

[indygreg (Gregory Szorc)]

This revision is now accepted and ready to land.
indygreg added a comment.
indygreg accepted this revision.


  I _think_ this is safe. But I would appreciate an extra set of eyes on the changed code in `revlog.py`. Essentially what's happening here is we were previously stuffing the number of revisions in the revlog in the in-memory data structure so we could recover the inline revlog if we had to roll back a `.i` -> `.d` revlog split that was incurred as part of the transaction.
  
  The key line not changed in this patch is `tr.replace(self.indexfile, trindex * self._io.size)`. And the new code gradually incrementing `trindex` as long as we're processing entries earlier than the `.i` offset recorded in the transaction seems correct to me.

REPOSITORY
  rHG Mercurial

BRANCH
  default

CHANGES SINCE LAST ACTION
  https://phab.mercurial-scm.org/D9275/new/

REVISION DETAIL
  https://phab.mercurial-scm.org/D9275

To: joerg.sonnenberger, indygreg, #hg-reviewers
Cc: mercurial-patches
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mercurial-scm.org/pipermail/mercurial-patches/attachments/20201107/39ae4c68/attachment-0002.html>