Review Request 121639: Split out KF5Contacts dependency

Mon Dec 29 15:08:57 UTC 2014

> On Dec. 29, 2014, 2:26 p.m., David Edmundson wrote:
> > src/metacontact.cpp, line 68
> > <https://git.reviewboard.kde.org/r/121639/diff/1/?file=335596#file335596line68>
> >
> >     deliberately public?

Eh well... it's the private implementation. they are usually public, no?

> On Dec. 29, 2014, 2:26 p.m., David Edmundson wrote:
> > src/persondata.cpp, line 156
> > <https://git.reviewboard.kde.org/r/121639/diff/1/?file=335599#file335599line156>
> >
> >     We're converting a list of variants to a string?
> >     
> >     I'm not entirely sure what the behaviour will be, but I doubt it's what we want.

Why do you say it's a list?

> On Dec. 29, 2014, 2:26 p.m., David Edmundson wrote:
> > autotests/fakecontactsource.cpp, line 57
> > <https://git.reviewboard.kde.org/r/121639/diff/1/?file=335573#file335573line57>
> >
> >     currently this is unused, which means we can't be  testing where we have multiple properties 
> >     
> >     also I don't really understand what this line would be doing, turning a variant into a string and then into a stringlist and then into a variant?

I guess I'll have to document this better.

The idea is that if you request "all-something" you are requesting all the alternatives of "something" and requesting "something" returns just the preferred value by any metric.

I guess this might not be a very intuitive API, maybe we'd prefer to have an extra argument in customProperty specifying the cardinality? something like enum Amount { Preferred, All } and then we know that All is a list of the type in Preferred.

> On Dec. 29, 2014, 2:26 p.m., David Edmundson wrote:
> > src/defaultcontactmonitor.cpp, line 38
> > <https://git.reviewboard.kde.org/r/121639/diff/1/?file=335589#file335589line38>
> >
> >     given we now control AbstractContact we can put the ID in that. 
> >     
> >     Will simplify a few APIs

So you propose it should be a property? or?

> On Dec. 29, 2014, 2:26 p.m., David Edmundson wrote:
> > src/metacontact.cpp, line 52
> > <https://git.reviewboard.kde.org/r/121639/diff/1/?file=335596#file335596line52>
> >
> >     merging behaviour isn't this simple.
> >     It change on the property, especially for cardinality.
> >     
> >     
> >     if I request a name I don't want to get a list, and it's especially important as we can use the redesign to not bother querying for data we're not going to show.
> >     
> >     
> >     It also makes the usage more complex where clients now have to start handling lists, and grouping all emails (now a list of lists) together.

See the first comment in here, note the "all-" part of the if().

- Aleix

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/121639/#review72493
-----------------------------------------------------------

On Dec. 24, 2014, 1:07 a.m., Aleix Pol Gonzalez wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/121639/
> -----------------------------------------------------------
> 
> (Updated Dec. 24, 2014, 1:07 a.m.)
> 
> 
> Review request for KDEPIM, Telepathy, Christian Mollekopf, David Edmundson, and Martin Klapetek.
> 
> 
> Repository: libkpeople
> 
> 
> Description
> -------
> 
> Last week we had a couple of discussions regarding KPeople relationship with KDE PIM and KContacts and we agreed that we want to keep KContacts as a framework that is good at dealing with vCards, which leaves KPeople without a good way to send around its data.
> 
> This patch proposes an internal AbstractContact class that the backends will have to re-implement. At the moment it only requires a "customProperty" that lets us fetch whatever is needed.
> Furthermore, it also includes a set of convenience methods in PersonData to have a type-safe API.
> 
> 
> Diffs
> -----
> 
>   CMakeLists.txt 3df5f0b 
>   autotests/CMakeLists.txt 676f170 
>   autotests/fakecontactsource.h 981fe03 
>   autotests/fakecontactsource.cpp 269f94f 
>   autotests/persondatatests.cpp 47a6213 
>   examples/CMakeLists.txt 74bd188 
>   examples/duplicates.cpp 72ff53b 
>   src/CMakeLists.txt 534066b 
>   src/backends/abstractcontact.h PRE-CREATION 
>   src/backends/abstractcontact.cpp PRE-CREATION 
>   src/backends/abstractpersonaction.h 703063e 
>   src/backends/abstractpersonaction.cpp be93d90 
>   src/backends/allcontactsmonitor.h 47e5dfd 
>   src/backends/allcontactsmonitor.cpp e517f17 
>   src/backends/basepersonsdatasource.h 3c25e28 
>   src/backends/basepersonsdatasource.cpp b794917 
>   src/backends/contactmonitor.h cf25953 
>   src/backends/contactmonitor.cpp ceb5d74 
>   src/declarative/personactionsmodel.cpp 4b34e15 
>   src/defaultcontactmonitor.cpp 36c31a7 
>   src/defaultcontactmonitor_p.h 8fc2812 
>   src/duplicatesfinder.cpp c6a1014 
>   src/global.h 9540623 
>   src/global.cpp 3ef834b 
>   src/match.cpp 3a93aba 
>   src/match_p.h 5727270 
>   src/metacontact.cpp bba6b65 
>   src/metacontact_p.h 0d738cc 
>   src/persondata.h 66b19a3 
>   src/persondata.cpp dab9f61 
>   src/personpluginmanager.cpp f249b5f 
>   src/personsmodel.h db907d4 
>   src/personsmodel.cpp ca6e972 
>   src/plugins/akonadi/akonadidatasource.cpp ae4ace1 
>   src/widgets/abstractfieldwidgetfactory.h 1d2d8aa 
>   src/widgets/persondetailsdialog.cpp 8ca0a99 
>   src/widgets/persondetailsview.cpp b7535bd 
>   src/widgets/plugins/emaildetailswidget.h 72a96ee 
>   src/widgets/plugins/emaildetailswidget.cpp 839c5f6 
>   src/widgets/plugins/emails.h ede1686 
>   src/widgets/plugins/emails.cpp d69101b 
> 
> Diff: https://git.reviewboard.kde.org/r/121639/diff/
> 
> 
> Testing
> -------
> 
> Builds, the test passes. ktp tests work, somewhat.
> Also there's few tests altogether.
> 
> 
> Thanks,
> 
> Aleix Pol Gonzalez
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-telepathy/attachments/20141229/e323d2ab/attachment.html>