1129692 - Update comm-central for PLDHashTable changes in bug 1050035

Status: RESOLVED DUPLICATE of bug 1163331

(MailNews Core :: Backend, defect)

Description

[Nicholas Nethercote [inactive]]

Bug 1050035 changed PLDHashTable's API in the following ways.

- PLDHashTable now allocates its entry storage lazily. (nsTHashtable and
  friends do too, since they are just layers on top of PLDHashTable.) This is a
  nice space win because about 45% of all created PLDHashTables never get any
  elements inserted into them.

- As a result, PL_DHashTableInit() is now infallible. This is possible because
  the allocation of entry storage now only occurs on table insertion, in
  PL_DHashTableAdd().

- An infallible version of PL_DHashTableAdd() has been added. To use the
  fallible version you need a |fallible_t| argument.

- PLD_NewDHashTable() and PLD_HashTableDestroy() have been removed. You should
  now just use |new|+PL_DHashTableInit and PLD_HashTableDestroy()+|delete|,
  which are more obvious and only slightly more verbose. (And I have plans to
  add a destructor and an initializing constructor, so in the end you'll be
  able to just use |new| + |delete|).
  
I've prepared a patch to get c-c compiling again.

Comment 1

[Nicholas Nethercote [inactive]]

Attachment: Update comm-central for PLDHashTable changes in bug 1050035

This patch keeps all the PLD_HashTableAdd() calls fallible by adding a
|mozilla::fallible| argument. It might be possible to get rid of
TokenHash::mTableIsInitialized but I didn't do that.

The patch is enough to get c-c compiling. I tested that starting and
immediately shutting down Thunderbird works, but nothing beyond that. A TB dev
should test this and land it. Thanks.

Comment 2

[Kent James (:rkent)]

Comment on attachment 8559456 [details] [diff] [review]
Update comm-central for PLDHashTable changes in bug 1050035

Yes this looks fine. Thanks so much for doing this!

Comment 3

[Kent James (:rkent)]

Pushed https://hg.mozilla.org/comm-central/rev/1037eb322c0f

Comment 4

[Nicholas Nethercote [inactive]]

I had to back out the patches in bug 1050035 due to some intermittent crashes and assertion failures, so this patch will also need backing out. Sorry.

Comment 5

[Philip Chee]

Attachment: Backout patch

Comment 6

[Kent James (:rkent)]

Comment on attachment 8562989 [details] [diff] [review]
Backout patch

Review of attachment 8562989 [details] [diff] [review]:
-----------------------------------------------------------------

Looks good, thanks for doing this!

Comment 7

[Kent James (:rkent)]

Pushed https://hg.mozilla.org/comm-central/rev/0eb00ed42931

Comment 8

[Eyal Rozenberg]

It looks like the commit fur this bug on 2015-02-05 busted my Thunderbird extension, BiDi Mail UI; see bug 1187572. I see a backout patch on 2015-02-11, but it seems bustage lingered on since February. I only just noticed it a short while ago after TB v38.0 was out. Can you please advise regarding what I/you can/should do about this?

Comment 9

[Wayne Mery (:wsmwk)]

(In reply to Nicholas Nethercote [:njn] from comment #4)
> I had to back out the patches in bug 1050035 due to some intermittent
> crashes and assertion failures, so this patch will also need backing out.
> Sorry.

Can we polish this off?  
Bug 1050035, bug 1144649 and bug 1131901 are long resolved in version 38.

Comment 10

[Kent James (:rkent)]

This ended up fixed in a different bug.