Merge lp:~3v1n0/u1db-qt/objects-subfields-list-all into lp:u1db-qt
Status: | Merged |
---|---|
Approved by: | Cris Dywan |
Approved revision: | 132 |
Merged at revision: | 130 |
Proposed branch: | lp:~3v1n0/u1db-qt/objects-subfields-list-all |
Merge into: | lp:u1db-qt |
Prerequisite: | lp:~3v1n0/u1db-qt/uri-path-parsing |
Diff against target: |
157 lines (+120/-2) 3 files modified
examples/u1db-qt-example-7/u1db-qt-example-7.qml (+75/-0) src/query.cpp (+4/-2) tests/test-database.cpp (+41/-0) |
To merge this branch: | bzr merge lp:~3v1n0/u1db-qt/objects-subfields-list-all |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Cris Dywan | Approve | ||
PS Jenkins bot | continuous-integration | Approve | |
Review via email: mp+252059@code.launchpad.net |
Commit message
Query: allow adding more than one result for each doc, allowing indexing subfields of objects in a list
This is to be conformant to what the u1db specs (implemented in python) say.
Description of the change
Indexing subfields of ojects in a list doesn't work as expected in u1db-qt, we only list the first element and not everyone as defined in docs [1].
The reason was that we didn't add a result if we already had one for that doc... And this seems wrong to me.
In a next MP we should probably also fix the "Name a list" case (see [1] again).
As test-case you can also try [2].
[1] https:/
[2] http://
PASSED: Continuous integration, rev:132 jenkins. qa.ubuntu. com/job/ u1db-qt- ci/66/ jenkins. qa.ubuntu. com/job/ u1db-qt- vivid-amd64- ci/13 jenkins. qa.ubuntu. com/job/ u1db-qt- vivid-armhf- ci/13
http://
Executed test runs:
SUCCESS: http://
SUCCESS: http://
Click here to trigger a rebuild: s-jenkins. ubuntu- ci:8080/ job/u1db- qt-ci/66/ rebuild
http://