[QGIS-Developer] What to do about WFS test failures?

classic Classic list List threaded Threaded
20 messages Options
Reply | Threaded
Open this post in threaded view
|

[QGIS-Developer] What to do about WFS test failures?

Nyall Dawson
Hi devs,

Just raising the question of what we should do about the constant WFS
test failures we get on Travis. I'd estimate 1 in 3 builds fails
because of WFS related tests hanging.

I'm very reluctant to disable all these tests, because:

1. WFS is super important.
2. I think there may be a real issue here - I've got at least one
customer who is having WFS issues with 3.2 and master. BUT: on the
other hand Travis has always been flaky with any test which uses
threads, regardless of which area of code it's from. So it could just
be Travis playing up again, in which case we'd need to disable these
tests like we do most of the other thread-related tests...

The current situation is basically unworkable. Sooooo.... ideas?

Nyall
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Alessandro Pasotti-2


On Sat, Sep 1, 2018 at 11:56 AM Nyall Dawson <[hidden email]> wrote:
Hi devs,

Just raising the question of what we should do about the constant WFS
test failures we get on Travis. I'd estimate 1 in 3 builds fails
because of WFS related tests hanging.

I'm very reluctant to disable all these tests, because:

1. WFS is super important.
2. I think there may be a real issue here - I've got at least one
customer who is having WFS issues with 3.2 and master. BUT: on the
other hand Travis has always been flaky with any test which uses
threads, regardless of which area of code it's from. So it could just
be Travis playing up again, in which case we'd need to disable these
tests like we do most of the other thread-related tests...

The current situation is basically unworkable. Sooooo.... ideas?

Nyall


Sorry I do not have any solution,  the only consideration that comes up to my mind is that if sum up all the time wasted by developers by unrelated/random Travis failures we would probably be very badly surprised  (as I usually say Travis is basically a very slow binary entropy generator).

Talking about the tests, WFS tests, and (even for different reasons) http-based integration tests (some auth tests, qgsfiledownloader, many server OWS tests) are apparently more fragile than others, mainly because they depend on (sometimes external) http services, but they have proven in the past to be very effective in spotting out regressions on very important feature likes WMS, WFS etc., some of them are also important because they ensure that QGIS client can talk to its server component.

The very bad things about disabling tests are:

- their development costed a lot of time and efforts and disabling them is not an incentive for the developers to write more tests [1]
- the tests are obviously important to prevent regressions
- before disabling a test we should make 100% sure that we are not overlooking a real bug (or what are the tests written for in the first place?)


So, my recommendation - not very useful in the short term I'm afraid - would be to start exploring other more reliable options for our CI, even if they are more expensive.

For the time being, as a temporary solution, there are no other options than disabling though.

Maybe a good idea would be to allocate some money specifically for the tests (a dedicated grant, a dedicated hackfest?) just an idea...

[1] of course sometimes it would just mean to write "better" and more robust tests, but I suspect that this is not generally the case.

--
Alessandro Pasotti
w3:   www.itopen.it

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Richard Duivenvoorde
In reply to this post by Nyall Dawson
On 09/01/2018 11:56 AM, Nyall Dawson wrote:

> Hi devs,
>
> Just raising the question of what we should do about the constant WFS
> test failures we get on Travis. I'd estimate 1 in 3 builds fails
> because of WFS related tests hanging.
>
> I'm very reluctant to disable all these tests, because:
>
> 1. WFS is super important.
> 2. I think there may be a real issue here - I've got at least one
> customer who is having WFS issues with 3.2 and master. BUT: on the
> other hand Travis has always been flaky with any test which uses
> threads, regardless of which area of code it's from. So it could just
> be Travis playing up again, in which case we'd need to disable these
> tests like we do most of the other thread-related tests...
>
> The current situation is basically unworkable. Sooooo.... ideas?

To add: QGIS recently has some issues and list messages concerning wfs
too. I searched some:

https://issues.qgis.org/issues/19702

https://lists.osgeo.org/pipermail/qgis-user/2018-August/043237.html

https://lists.osgeo.org/pipermail/qgis-user/2018-August/043241.html

And myself I also have the feeling that WFS is less stable/usable then
in 2.18 (I release a list of national WFS/WMS/WCS's in one of my plugins).

So I think it is not only CI that has problems, it looks like changes in
WFS made it more tricky to use?

Regards,

Richard Duivenvoorde
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Alessandro Pasotti-2


On Sat, Sep 1, 2018 at 12:37 PM Richard Duivenvoorde <[hidden email]> wrote:
On 09/01/2018 11:56 AM, Nyall Dawson wrote:
> Hi devs,
>
> Just raising the question of what we should do about the constant WFS
> test failures we get on Travis. I'd estimate 1 in 3 builds fails
> because of WFS related tests hanging.
>
> I'm very reluctant to disable all these tests, because:
>
> 1. WFS is super important.
> 2. I think there may be a real issue here - I've got at least one
> customer who is having WFS issues with 3.2 and master. BUT: on the
> other hand Travis has always been flaky with any test which uses
> threads, regardless of which area of code it's from. So it could just
> be Travis playing up again, in which case we'd need to disable these
> tests like we do most of the other thread-related tests...
>
> The current situation is basically unworkable. Sooooo.... ideas?

To add: QGIS recently has some issues and list messages concerning wfs
too. I searched some:

https://issues.qgis.org/issues/19702

https://lists.osgeo.org/pipermail/qgis-user/2018-August/043237.html

https://lists.osgeo.org/pipermail/qgis-user/2018-August/043241.html

And myself I also have the feeling that WFS is less stable/usable then
in 2.18 (I release a list of national WFS/WMS/WCS's in one of my plugins).

So I think it is not only CI that has problems, it looks like changes in
WFS made it more tricky to use?


Richard, please let's focus on Travis in this thread.

WFS UX issues on 3.x are a complete different topic.


--
Alessandro Pasotti
w3:   www.itopen.it

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Richard Duivenvoorde
On 09/01/2018 12:43 PM, Alessandro Pasotti wrote:

> Richard, please let's focus on Travis in this thread.
>
> WFS UX issues on 3.x are a complete different topic.

Ok, sorry. What I just wanted to add: it looks like that 'default'-wfs
behaviour has changed, raising wfs problems .

I just wanted to raise awareness of this, thinking the CI failures could
have the same cause.

Regards,

Richard Duivenvoorde

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Nyall Dawson
In reply to this post by Richard Duivenvoorde
On Sat, 1 Sep 2018 at 20:37, Richard Duivenvoorde <[hidden email]> wrote:

> And myself I also have the feeling that WFS is less stable

This is my experience too, which is why I'm not going to 100%
attribute these errors to Travis!

Nyall

> /usable then
> in 2.18 (I release a list of national WFS/WMS/WCS's in one of my plugins).
>
> So I think it is not only CI that has problems, it looks like changes in
> WFS made it more tricky to use?
>
> Regards,
>
> Richard Duivenvoorde
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Mathieu Pellerin
Can these failures be reproduced locally if one repeatedly tests those cases?

On Mon, Sep 3, 2018, 07:02 Nyall Dawson <[hidden email]> wrote:
On Sat, 1 Sep 2018 at 20:37, Richard Duivenvoorde <[hidden email]> wrote:

> And myself I also have the feeling that WFS is less stable

This is my experience too, which is why I'm not going to 100%
attribute these errors to Travis!

Nyall

> /usable then
> in 2.18 (I release a list of national WFS/WMS/WCS's in one of my plugins).
>
> So I think it is not only CI that has problems, it looks like changes in
> WFS made it more tricky to use?
>
> Regards,
>
> Richard Duivenvoorde
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Tom Chadwin
I can't offer any helpful suggestions, but just to let you know I finally had
to disable all my plugin WFS tests. I used to cope, by rerunning failed
Travis runs, but by about three months ago, it seemed no longer usable -
failure after failure.

I was using a third-party WFS, and perhaps I could have got round this by
adding a WFS provider to the test docker image, but in this plugin's case, I
didn't think it worth the significant effort to do so. The WM(T)S tests also
use third-party sources and seem stable, so perhaps this wasn't the
underlying issue anyway.

If an improvement or solution could be found, it would be great to reinstate
these tests.

Thanks

Tom



-----
Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Régis Haubourg
Hi all,
I very much think that the WFS client is an really bad state, and is not really reliable, especially in WFS-T context.
The good news is that we just have been funded to refactor it !
The work should start in september and land in 3.6. I will let our dev's come here with more technical details about the goals. I hope we will also be able to take benefit of this to this the OGC compliancy of the client here.
Best regards,
Régis

Le lun. 3 sept. 2018 à 11:36, Tom Chadwin <[hidden email]> a écrit :
I can't offer any helpful suggestions, but just to let you know I finally had
to disable all my plugin WFS tests. I used to cope, by rerunning failed
Travis runs, but by about three months ago, it seemed no longer usable -
failure after failure.

I was using a third-party WFS, and perhaps I could have got round this by
adding a WFS provider to the test docker image, but in this plugin's case, I
didn't think it worth the significant effort to do so. The WM(T)S tests also
use third-party sources and seem stable, so perhaps this wasn't the
underlying issue anyway.

If an improvement or solution could be found, it would be great to reinstate
these tests.

Thanks

Tom



-----
Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Even Rouault-2
Régis,

Not that I'm against improvements, all the contrary, but just wanted to
underline that the provider was seriously refactored already in 2.16. Clearly
the lack of WFS-T support for 1.1 and 2.0 in the scope of those enhancements
can be a source of confusion currently for users. What do you have in mind as
refactoring exactly ?

Even

> Hi all,
> I very much think that the WFS client is an really bad state, and is not
> really reliable, especially in WFS-T context.
> The good news is that we just have been funded to refactor it !
> The work should start in september and land in 3.6. I will let our dev's
> come here with more technical details about the goals. I hope we will also
> be able to take benefit of this to this the OGC compliancy of the client
> here.
> Best regards,
> Régis
>
> Le lun. 3 sept. 2018 à 11:36, Tom Chadwin <[hidden email]> a
>
> écrit :
> > I can't offer any helpful suggestions, but just to let you know I finally
> > had
> > to disable all my plugin WFS tests. I used to cope, by rerunning failed
> > Travis runs, but by about three months ago, it seemed no longer usable -
> > failure after failure.
> >
> > I was using a third-party WFS, and perhaps I could have got round this by
> > adding a WFS provider to the test docker image, but in this plugin's case,
> > I
> > didn't think it worth the significant effort to do so. The WM(T)S tests
> > also
> > use third-party sources and seem stable, so perhaps this wasn't the
> > underlying issue anyway.
> >
> > If an improvement or solution could be found, it would be great to
> > reinstate
> > these tests.
> >
> > Thanks
> >
> > Tom
> >
> >
> >
> > -----
> > Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
> > --
> > Sent from:
> > http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
> > _______________________________________________
> > QGIS-Developer mailing list
> > [hidden email]
> > List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


--
Spatialys - Geospatial professional services
http://www.spatialys.com
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Régis Haubourg
Hi Even, thanks for pointing that, I missed that history.
I'll ask the dev's for the detailed improvements planned, I dont' have any detail currently (sorry for that)
Régis

Le lun. 3 sept. 2018 à 21:29, Even Rouault <[hidden email]> a écrit :
Régis,

Not that I'm against improvements, all the contrary, but just wanted to
underline that the provider was seriously refactored already in 2.16. Clearly
the lack of WFS-T support for 1.1 and 2.0 in the scope of those enhancements
can be a source of confusion currently for users. What do you have in mind as
refactoring exactly ?

Even

> Hi all,
> I very much think that the WFS client is an really bad state, and is not
> really reliable, especially in WFS-T context.
> The good news is that we just have been funded to refactor it !
> The work should start in september and land in 3.6. I will let our dev's
> come here with more technical details about the goals. I hope we will also
> be able to take benefit of this to this the OGC compliancy of the client
> here.
> Best regards,
> Régis
>
> Le lun. 3 sept. 2018 à 11:36, Tom Chadwin <[hidden email]> a
>
> écrit :
> > I can't offer any helpful suggestions, but just to let you know I finally
> > had
> > to disable all my plugin WFS tests. I used to cope, by rerunning failed
> > Travis runs, but by about three months ago, it seemed no longer usable -
> > failure after failure.
> >
> > I was using a third-party WFS, and perhaps I could have got round this by
> > adding a WFS provider to the test docker image, but in this plugin's case,
> > I
> > didn't think it worth the significant effort to do so. The WM(T)S tests
> > also
> > use third-party sources and seem stable, so perhaps this wasn't the
> > underlying issue anyway.
> >
> > If an improvement or solution could be found, it would be great to
> > reinstate
> > these tests.
> >
> > Thanks
> >
> > Tom
> >
> >
> >
> > -----
> > Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
> > --
> > Sent from:
> > http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
> > _______________________________________________
> > QGIS-Developer mailing list
> > [hidden email]
> > List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


--
Spatialys - Geospatial professional services
http://www.spatialys.com

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Jeremy Palmer-3
Can I also add that the refactoring that was funded in 2.16 added tests, paging support, filter query builder and dynamic caching. There is now a lot of complexity in the driver due to the complexity of WFS server implementations and the standard, plus the multi-threading code. If a refactor is proposed I would be against anything that doesn't deal current use cases and edge cases which have already been implemented. Maybe some analysis of the failed tests is the first place to start. WFS-T is a side issue to that.

Cheers,
Jeremy

On Tue, Sep 4, 2018 at 7:51 AM Régis Haubourg <[hidden email]> wrote:
Hi Even, thanks for pointing that, I missed that history.
I'll ask the dev's for the detailed improvements planned, I dont' have any detail currently (sorry for that)
Régis

Le lun. 3 sept. 2018 à 21:29, Even Rouault <[hidden email]> a écrit :
Régis,

Not that I'm against improvements, all the contrary, but just wanted to
underline that the provider was seriously refactored already in 2.16. Clearly
the lack of WFS-T support for 1.1 and 2.0 in the scope of those enhancements
can be a source of confusion currently for users. What do you have in mind as
refactoring exactly ?

Even

> Hi all,
> I very much think that the WFS client is an really bad state, and is not
> really reliable, especially in WFS-T context.
> The good news is that we just have been funded to refactor it !
> The work should start in september and land in 3.6. I will let our dev's
> come here with more technical details about the goals. I hope we will also
> be able to take benefit of this to this the OGC compliancy of the client
> here.
> Best regards,
> Régis
>
> Le lun. 3 sept. 2018 à 11:36, Tom Chadwin <[hidden email]> a
>
> écrit :
> > I can't offer any helpful suggestions, but just to let you know I finally
> > had
> > to disable all my plugin WFS tests. I used to cope, by rerunning failed
> > Travis runs, but by about three months ago, it seemed no longer usable -
> > failure after failure.
> >
> > I was using a third-party WFS, and perhaps I could have got round this by
> > adding a WFS provider to the test docker image, but in this plugin's case,
> > I
> > didn't think it worth the significant effort to do so. The WM(T)S tests
> > also
> > use third-party sources and seem stable, so perhaps this wasn't the
> > underlying issue anyway.
> >
> > If an improvement or solution could be found, it would be great to
> > reinstate
> > these tests.
> >
> > Thanks
> >
> > Tom
> >
> >
> >
> > -----
> > Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
> > --
> > Sent from:
> > http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
> > _______________________________________________
> > QGIS-Developer mailing list
> > [hidden email]
> > List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


--
Spatialys - Geospatial professional services
http://www.spatialys.com
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

carlob
What about starting with/focusing on WFS 3? The new version is really cleaner and seems much more efficient.
The current WFS implementation in QGIS is much better than previous versions, even if sometimes making a virtual OGR file il the only way to use some services. There are really bad server implementations out there, additionally an automated solution struggles against misleading and lazy XML informations. This is a broken idea or at least one that asks for close cooperation between server and client.
Just my remaining cent.
c


On Mon, Sep 3, 2018 at 10:01 PM, Jeremy Palmer <[hidden email]> wrote:
Can I also add that the refactoring that was funded in 2.16 added tests, paging support, filter query builder and dynamic caching. There is now a lot of complexity in the driver due to the complexity of WFS server implementations and the standard, plus the multi-threading code. If a refactor is proposed I would be against anything that doesn't deal current use cases and edge cases which have already been implemented. Maybe some analysis of the failed tests is the first place to start. WFS-T is a side issue to that.

Cheers,
Jeremy


On Tue, Sep 4, 2018 at 7:51 AM Régis Haubourg <[hidden email]> wrote:
Hi Even, thanks for pointing that, I missed that history.
I'll ask the dev's for the detailed improvements planned, I dont' have any detail currently (sorry for that)
Régis

Le lun. 3 sept. 2018 à 21:29, Even Rouault <[hidden email]> a écrit :
Régis,

Not that I'm against improvements, all the contrary, but just wanted to
underline that the provider was seriously refactored already in 2.16. Clearly
the lack of WFS-T support for 1.1 and 2.0 in the scope of those enhancements
can be a source of confusion currently for users. What do you have in mind as
refactoring exactly ?

Even

> Hi all,
> I very much think that the WFS client is an really bad state, and is not
> really reliable, especially in WFS-T context.
> The good news is that we just have been funded to refactor it !
> The work should start in september and land in 3.6. I will let our dev's
> come here with more technical details about the goals. I hope we will also
> be able to take benefit of this to this the OGC compliancy of the client
> here.
> Best regards,
> Régis
>
> Le lun. 3 sept. 2018 à 11:36, Tom Chadwin <[hidden email]> a
>
> écrit :
> > I can't offer any helpful suggestions, but just to let you know I finally
> > had
> > to disable all my plugin WFS tests. I used to cope, by rerunning failed
> > Travis runs, but by about three months ago, it seemed no longer usable -
> > failure after failure.
> >
> > I was using a third-party WFS, and perhaps I could have got round this by
> > adding a WFS provider to the test docker image, but in this plugin's case,
> > I
> > didn't think it worth the significant effort to do so. The WM(T)S tests
> > also
> > use third-party sources and seem stable, so perhaps this wasn't the
> > underlying issue anyway.
> >
> > If an improvement or solution could be found, it would be great to
> > reinstate
> > these tests.
> >
> > Thanks
> >
> > Tom
> >
> >
> >
> > -----
> > Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
> > --
> > Sent from:
> > http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
> > _______________________________________________
> > QGIS-Developer mailing list
> > [hidden email]
> > List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


--
Spatialys - Geospatial professional services
http://www.spatialys.com
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer



--
--------------------------------------------------------------------------
Carlo A. Bertelli
   Charta servizi e sistemi per il territorio e la storia ambientale srl
          Dipendenze del palazzo Doria,
          vc. alla Chiesa della Maddalena 9/2 16124      Genova (Italy)
          tel./fax +39(0)10 2475439  +39 0108566195  mobile:+39 393 1590711
   e-mail: [hidden email]      http://www.chartasrl.eu
--------------------------------------------------------------------------




_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Jeremy Palmer-3
I agree WFS 3.0 is a much better implementation and it would be great if a implementation is started soon to track the current standards development. However, we still have (and will have for a long time) a user need to support WFS 1.0 and 2.0 - so this still needs to be deal with. 

On Tue, Sep 4, 2018 at 8:32 AM Carlo A. Bertelli (Charta s.r.l.) <[hidden email]> wrote:
What about starting with/focusing on WFS 3? The new version is really cleaner and seems much more efficient.
The current WFS implementation in QGIS is much better than previous versions, even if sometimes making a virtual OGR file il the only way to use some services. There are really bad server implementations out there, additionally an automated solution struggles against misleading and lazy XML informations. This is a broken idea or at least one that asks for close cooperation between server and client.
Just my remaining cent.
c


On Mon, Sep 3, 2018 at 10:01 PM, Jeremy Palmer <[hidden email]> wrote:
Can I also add that the refactoring that was funded in 2.16 added tests, paging support, filter query builder and dynamic caching. There is now a lot of complexity in the driver due to the complexity of WFS server implementations and the standard, plus the multi-threading code. If a refactor is proposed I would be against anything that doesn't deal current use cases and edge cases which have already been implemented. Maybe some analysis of the failed tests is the first place to start. WFS-T is a side issue to that.

Cheers,
Jeremy


On Tue, Sep 4, 2018 at 7:51 AM Régis Haubourg <[hidden email]> wrote:
Hi Even, thanks for pointing that, I missed that history.
I'll ask the dev's for the detailed improvements planned, I dont' have any detail currently (sorry for that)
Régis

Le lun. 3 sept. 2018 à 21:29, Even Rouault <[hidden email]> a écrit :
Régis,

Not that I'm against improvements, all the contrary, but just wanted to
underline that the provider was seriously refactored already in 2.16. Clearly
the lack of WFS-T support for 1.1 and 2.0 in the scope of those enhancements
can be a source of confusion currently for users. What do you have in mind as
refactoring exactly ?

Even

> Hi all,
> I very much think that the WFS client is an really bad state, and is not
> really reliable, especially in WFS-T context.
> The good news is that we just have been funded to refactor it !
> The work should start in september and land in 3.6. I will let our dev's
> come here with more technical details about the goals. I hope we will also
> be able to take benefit of this to this the OGC compliancy of the client
> here.
> Best regards,
> Régis
>
> Le lun. 3 sept. 2018 à 11:36, Tom Chadwin <[hidden email]> a
>
> écrit :
> > I can't offer any helpful suggestions, but just to let you know I finally
> > had
> > to disable all my plugin WFS tests. I used to cope, by rerunning failed
> > Travis runs, but by about three months ago, it seemed no longer usable -
> > failure after failure.
> >
> > I was using a third-party WFS, and perhaps I could have got round this by
> > adding a WFS provider to the test docker image, but in this plugin's case,
> > I
> > didn't think it worth the significant effort to do so. The WM(T)S tests
> > also
> > use third-party sources and seem stable, so perhaps this wasn't the
> > underlying issue anyway.
> >
> > If an improvement or solution could be found, it would be great to
> > reinstate
> > these tests.
> >
> > Thanks
> >
> > Tom
> >
> >
> >
> > -----
> > Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
> > --
> > Sent from:
> > http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
> > _______________________________________________
> > QGIS-Developer mailing list
> > [hidden email]
> > List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


--
Spatialys - Geospatial professional services
http://www.spatialys.com
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer



--
--------------------------------------------------------------------------
Carlo A. Bertelli
   Charta servizi e sistemi per il territorio e la storia ambientale srl
          Dipendenze del palazzo Doria,
          vc. alla Chiesa della Maddalena 9/2 16124      Genova (Italy)
          tel./fax +39(0)10 2475439  +39 0108566195  mobile:+39 393 1590711
   e-mail: [hidden email]      http://www.chartasrl.eu
--------------------------------------------------------------------------




_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Jonathan Moules-4
In reply to this post by Régis Haubourg

I'd suggest based on Tom's post that the work should include some sort of reliable way of testing the WFS client (and of course full test coverage) - I don't know if GeoServer is Docker-happy these days, but the default install in a VM should be sufficient for the task (it comes with test layers) - although GeoServer doesn't support WFS-T for writing.

Cheers,
Jonathan

On 03/09/2018 12:04, Régis Haubourg wrote:
Hi all,
I very much think that the WFS client is an really bad state, and is not really reliable, especially in WFS-T context.
The good news is that we just have been funded to refactor it !
The work should start in september and land in 3.6. I will let our dev's come here with more technical details about the goals. I hope we will also be able to take benefit of this to this the OGC compliancy of the client here.
Best regards,
Régis

Le lun. 3 sept. 2018 à 11:36, Tom Chadwin <[hidden email]> a écrit :
I can't offer any helpful suggestions, but just to let you know I finally had
to disable all my plugin WFS tests. I used to cope, by rerunning failed
Travis runs, but by about three months ago, it seemed no longer usable -
failure after failure.

I was using a third-party WFS, and perhaps I could have got round this by
adding a WFS provider to the test docker image, but in this plugin's case, I
didn't think it worth the significant effort to do so. The WM(T)S tests also
use third-party sources and seem stable, so perhaps this wasn't the
underlying issue anyway.

If an improvement or solution could be found, it would be great to reinstate
these tests.

Thanks

Tom



-----
Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Nyall Dawson
In reply to this post by Even Rouault-2
On Tue, 4 Sep 2018 at 05:50, Even Rouault <[hidden email]> wrote:
>
> Régis,
>
> Not that I'm against improvements, all the contrary, but just wanted to
> underline that the provider was seriously refactored already in 2.16. Clearly
> the lack of WFS-T support for 1.1 and 2.0 in the scope of those enhancements
> can be a source of confusion currently for users. What do you have in mind as
> refactoring exactly ?

I'm with Even here... please, proceed with extreme caution.

We may be seeing issues with the WFS provider since 3.2, but I do not
think a full rewrite is needed/wanted here, and potentially will cause
many regressions given how finicky WFS servers are and how many
server-specific fixes have been required in the current
implementation.

Nyall


>
> Even
>
> > Hi all,
> > I very much think that the WFS client is an really bad state, and is not
> > really reliable, especially in WFS-T context.
> > The good news is that we just have been funded to refactor it !
> > The work should start in september and land in 3.6. I will let our dev's
> > come here with more technical details about the goals. I hope we will also
> > be able to take benefit of this to this the OGC compliancy of the client
> > here.
> > Best regards,
> > Régis
> >
> > Le lun. 3 sept. 2018 à 11:36, Tom Chadwin <[hidden email]> a
> >
> > écrit :
> > > I can't offer any helpful suggestions, but just to let you know I finally
> > > had
> > > to disable all my plugin WFS tests. I used to cope, by rerunning failed
> > > Travis runs, but by about three months ago, it seemed no longer usable -
> > > failure after failure.
> > >
> > > I was using a third-party WFS, and perhaps I could have got round this by
> > > adding a WFS provider to the test docker image, but in this plugin's case,
> > > I
> > > didn't think it worth the significant effort to do so. The WM(T)S tests
> > > also
> > > use third-party sources and seem stable, so perhaps this wasn't the
> > > underlying issue anyway.
> > >
> > > If an improvement or solution could be found, it would be great to
> > > reinstate
> > > these tests.
> > >
> > > Thanks
> > >
> > > Tom
> > >
> > >
> > >
> > > -----
> > > Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
> > > --
> > > Sent from:
> > > http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
> > > _______________________________________________
> > > QGIS-Developer mailing list
> > > [hidden email]
> > > List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> > > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>
>
> --
> Spatialys - Geospatial professional services
> http://www.spatialys.com
> _______________________________________________
> QGIS-Developer mailing list
> [hidden email]
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

geowolf
In reply to this post by Jonathan Moules-4
On Tue, Sep 4, 2018 at 8:51 AM Jonathan Moules <[hidden email]> wrote:

I'd suggest based on Tom's post that the work should include some sort of reliable way of testing the WFS client (and of course full test coverage) - I don't know if GeoServer is Docker-happy these days, but the default install in a VM should be sufficient for the task (it comes with test layers) - although GeoServer doesn't support WFS-T for writing.

"GeoServer doesn't support WFS-T for writing" ... what??? :-)

The default data directory disables writes for any user but the administrator, you can change it, or just authenticate
as the admin to run WFS-T

Cheers
Andrea

==

GeoServer Professional Services from the experts! Visit http://goo.gl/it488V for more information. == Ing. Andrea Aime @geowolf Technical Lead GeoSolutions S.A.S. Via di Montramito 3/A 55054 Massarosa (LU) phone: +39 0584 962313 fax: +39 0584 1660272 mob: +39 339 8844549 http://www.geo-solutions.it http://twitter.com/geosolutions_it ------------------------------------------------------- Con riferimento alla normativa sul trattamento dei dati personali (Reg. UE 2016/679 - Regolamento generale sulla protezione dei dati “GDPR”), si precisa che ogni circostanza inerente alla presente email (il suo contenuto, gli eventuali allegati, etc.) è un dato la cui conoscenza è riservata al/i solo/i destinatario/i indicati dallo scrivente. Se il messaggio Le è giunto per errore, è tenuta/o a cancellarlo, ogni altra operazione è illecita. Le sarei comunque grato se potesse darmene notizia. This email is intended only for the person or entity to which it is addressed and may contain information that is privileged, confidential or otherwise protected from disclosure. We remind that - as provided by European Regulation 2016/679 “GDPR” - copying, dissemination or use of this e-mail or the information herein by anyone other than the intended recipient is prohibited. If you have received this email by mistake, please notify us immediately by telephone or e-mail.


_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Régis Haubourg
In reply to this post by Jonathan Moules-4
Hi,
I just checked the goals of the work on the WFS provider. It is not a refactor in fat, sorry for having used that word. The work consists in having a closer look to the WFS and snapping cache interactions because they cause lots of problems when the distant database has triggers that would require WFS cache reloading and then snapping cache reloading.
Maybe Vincent can tell more about that, but it's not a major change then. I should'nt have raised the hand here, sorry for the noise.
Regards,
Régis

Le mar. 4 sept. 2018 à 08:51, Jonathan Moules <[hidden email]> a écrit :

I'd suggest based on Tom's post that the work should include some sort of reliable way of testing the WFS client (and of course full test coverage) - I don't know if GeoServer is Docker-happy these days, but the default install in a VM should be sufficient for the task (it comes with test layers) - although GeoServer doesn't support WFS-T for writing.

Cheers,
Jonathan

On 03/09/2018 12:04, Régis Haubourg wrote:
Hi all,
I very much think that the WFS client is an really bad state, and is not really reliable, especially in WFS-T context.
The good news is that we just have been funded to refactor it !
The work should start in september and land in 3.6. I will let our dev's come here with more technical details about the goals. I hope we will also be able to take benefit of this to this the OGC compliancy of the client here.
Best regards,
Régis

Le lun. 3 sept. 2018 à 11:36, Tom Chadwin <[hidden email]> a écrit :
I can't offer any helpful suggestions, but just to let you know I finally had
to disable all my plugin WFS tests. I used to cope, by rerunning failed
Travis runs, but by about three months ago, it seemed no longer usable -
failure after failure.

I was using a third-party WFS, and perhaps I could have got round this by
adding a WFS provider to the test docker image, but in this plugin's case, I
didn't think it worth the significant effort to do so. The WM(T)S tests also
use third-party sources and seem stable, so perhaps this wasn't the
underlying issue anyway.

If an improvement or solution could be found, it would be great to reinstate
these tests.

Thanks

Tom



-----
Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Tim Sutton-6
In reply to this post by Jonathan Moules-4
Hi 



On 03 Sep 2018, at 22:54, Jonathan Moules <[hidden email]> wrote:

I'd suggest based on Tom's post that the work should include some sort of reliable way of testing the WFS client (and of course full test coverage) - I don't know if GeoServer is Docker-happy these days


Kartoza publishes geoserver images here:



Regards

Tim


, but the default install in a VM should be sufficient for the task (it comes with test layers) - although GeoServer doesn't support WFS-T for writing.

Cheers,
Jonathan

On 03/09/2018 12:04, Régis Haubourg wrote:
Hi all,
I very much think that the WFS client is an really bad state, and is not really reliable, especially in WFS-T context.
The good news is that we just have been funded to refactor it !
The work should start in september and land in 3.6. I will let our dev's come here with more technical details about the goals. I hope we will also be able to take benefit of this to this the OGC compliancy of the client here.
Best regards,
Régis

Le lun. 3 sept. 2018 à 11:36, Tom Chadwin <[hidden email]> a écrit :
I can't offer any helpful suggestions, but just to let you know I finally had
to disable all my plugin WFS tests. I used to cope, by rerunning failed
Travis runs, but by about three months ago, it seemed no longer usable -
failure after failure.

I was using a third-party WFS, and perhaps I could have got round this by
adding a WFS provider to the test docker image, but in this plugin's case, I
didn't think it worth the significant effort to do so. The WM(T)S tests also
use third-party sources and seem stable, so perhaps this wasn't the
underlying issue anyway.

If an improvement or solution could be found, it would be great to reinstate
these tests.

Thanks

Tom



-----
Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer








Tim Sutton

Co-founder: Kartoza
Ex Project chair: QGIS.org

Visit http://kartoza.com to find out about open source:

Desktop GIS programming services
Geospatial web development
GIS Training
Consulting Services

Skype: timlinux 
IRC: timlinux on #qgis at freenode.net


_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

signature.asc (499 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: What to do about WFS test failures?

Nyall Dawson
In reply to this post by Nyall Dawson
On Sat, 1 Sep 2018 at 19:56, Nyall Dawson <[hidden email]> wrote:

> 2. I think there may be a real issue here - I've got at least one
> customer who is having WFS issues with 3.2 and master. BUT: on the
> other hand Travis has always been flaky with any test which uses
> threads, regardless of which area of code it's from. So it could just
> be Travis playing up again, in which case we'd need to disable these
> tests like we do most of the other thread-related tests...

For those following at home -- I think the underlying issue (a bug)
has been found and fixed... fingers crossed!

Nyall
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer