dee

Merge lp://qastaging/~kamstrup/dee/filter-models-ng into lp://qastaging/dee

filter-models-ng
Merge into trunk

Proposed by Mikkel Kamstrup Erlandsen on 2011-12-14

Status:

Merged

Approved by:

Michal Hruby on 2011-12-15

Approved revision:

324

Merged at revision:

320

Proposed branch:

lp://qastaging/~kamstrup/dee/filter-models-ng

Merge into:

lp://qastaging/dee

Diff against target:

1257 lines (+420/-277)

12 files modified

dee/Makefile.am (+2/-2)
dee/dee-filter-model.c (+23/-26)
dee/dee-filter-model.h (+4/-61)
dee/dee-filter.c (+198/-96)
dee/dee-filter.h (+103/-9)
dee/dee-model-reader.c (+2/-2)
dee/dee-model-reader.h (+5/-2)
dee/dee.h (+1/-1)
doc/reference/dee-1.0/dee-1.0-docs.sgml (+1/-1)
tests/test-filter-model.c (+52/-39)
vapi/Dee-1.0-custom.vala (+6/-14)
vapi/dee-1.0.vapi (+23/-24)

To merge this branch:

bzr merge lp://qastaging/~kamstrup/dee/filter-models-ng

High

Fix Released

Link a bug report

Reviewer	Review Type	Date Requested	Status
Michal Hruby (community)		2011-12-14	Approve on 2011-12-15
Review via email: mp+85706@code.qastaging.launchpad.net

This proposal supersedes a proposal from 2011-12-14.

Description of the change

Overhaul DeeFilter and DeeFilterModel to work with GI and be more like DeeModelReader and DeeIndex

This branch gives a slight API break in how DeeFilters are created and reverses the order of the arguments to dee_filter_model_new() to match the order in dee_{hash,tree}_index_new(). I tested manually that Python works with DeeFilterModel now.

(resubmitted because of quantum entanglement with the no-const branch)

Revision history for this message

Michal Hruby (mhr3) wrote on 2011-12-14: Posted in a previous version of this proposal

Seems to be mixed up with the no-const branch, can you please merge trunk/resubmit, or whatever is necessary?

Revision history for this message

Michal Hruby (mhr3) wrote on 2011-12-15:

I'm getting: `Warning: multiple "IDs" for constraint linkend: DeeFilter.` when generating the documentation, not sure what's the root cause of that, but it wasn't there in trunk.

What's missing is regenerated vapi (we really need to hook that up into the build system properly) + no longer forcing Dee.Filter to be compact class.

One of the things I was never sure about is whether GI-langs are able to use functions that have in the C signature two function pointers and a single userdata + destroy (ie current dee_filter_new), but it doesn't seem likely.

review: Needs Information

Revision history for this message

Mikkel Kamstrup Erlandsen (kamstrup) wrote on 2011-12-15:

> I'm getting: `Warning: multiple "IDs" for constraint linkend: DeeFilter.` when
> generating the documentation, not sure what's the root cause of that, but it
> wasn't there in trunk.

I'll take a quick look; but it doesn't worry me too much right now as the docs themselves look fine. So I think we can land even if I don't find a fix right away.

> What's missing is regenerated vapi (we really need to hook that up into the
> build system properly) + no longer forcing Dee.Filter to be compact class.

Done

> One of the things I was never sure about is whether GI-langs are able to use
> functions that have in the C signature two function pointers and a single
> userdata + destroy (ie current dee_filter_new), but it doesn't seem likely.

This did cross my mind, and I thin you're right that it might now work. GDBus and other APIsdoes this as well though - but since we can easily expand the set of DeeFilters we distribute I am not too concerned about this.

lp://qastaging/~kamstrup/dee/filter-models-ng updated on 2011-12-15

323. By Mikkel Kamstrup Erlandsen on 2011-12-15

Update VAPI for DeeFilter and DeeFilterModel

324. By Mikkel Kamstrup Erlandsen on 2011-12-15