Support dynamic domains in mod_roster #3159
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov Report
@@ Coverage Diff @@
## master #3159 +/- ##
==========================================
+ Coverage 79.56% 80.30% +0.74%
==========================================
Files 396 396
Lines 32287 32260 -27
==========================================
+ Hits 25688 25908 +220
+ Misses 6599 6352 -247
Continue to review full report at Codecov.
|
chrzaszcz
force-pushed
the
mod_roster_with_dynamic_domains
branch
3 times, most recently
from
cba74c2
to
8a08909
Compare
Return "internal server error" for all IQ requests, not only for roster.
This is against RFC 6121 for roster set. For roster get it is unspecified, but very unexpected as well. Following commits will change the tests and the implementation.
According to RFC6121, section 2.3.3, it is forbidden to send roster set to another user's JID. In the current implementation there was only a check if the domain exists, which would leak information about other tenants served by the system. Handling of roster get to another user's JID is unspecified by the RFC, but it makes most sense to handle it the same way - it is better to return an error than to accept such stanzas, which are most likely sent by accident.
chrzaszcz
force-pushed
the
mod_roster_with_dynamic_domains
branch
2 times, most recently
from
d3ddd91
to
d85befc
Compare
For 'rosterusers' it was there but it was unused. Two more unused columns are removed: 'subscribe' and 'type'.
The 'raw_to_record' callback was a no-op for all backends.
Rework the backend callbacks: - Add the host type - Merge all 'get_roster_entry_*' variants into one function - Remove unused 'raw_to_record' - Define types for cleaner and shorter type specs Make 'set_roster_item_t' return an error if a non-existing item is removed. This allows to reduce the extra check before calling 'set_roster_item'.
chrzaszcz
force-pushed
the
mod_roster_with_dynamic_domains
branch
from
d85befc
to
765da85
Compare
- Update callbacks - Update prepared queries
Take it from the creds
The calls are preceded with 'does_user_exist', so we can assume the host type exists.
ets:lookup_element/2 was throwing 'badarg' for empty feature lists
Dynamic domains don't support BOSH and these test cases shouldn't depend on BOSH anyway.
chrzaszcz
force-pushed
the
mod_roster_with_dynamic_domains
branch
from
765da85
to
3e5cd0e
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Main changes:
mod_rosterto use host types.Side changes:
mod_roster. They started becoming tedious to maintain.mod_rosterto make thetest typesection consistent with the top-level ones by excluding the header from the template. This makes the templating itself a bit inconsistent, but IMO other modules can be converted as well. It makes sense to exclude constant parts from the template.Not changed:
roster_groupshook is only used bymod_pubsuband has no handlers. It can be removed, but we should figure out whether it needs a fix, as the resulting config field is always an empty list.Take a quick look at the commit messages for more information.