[Qgis-developer] Missing the support for scale parameter in WMS Legend

[Andrea Peri]

Hi Nyal,
Unfortunatelly no.
No news on this.
I have no the skill to understand what is the code interested.
I know it should on one the the committed link due a report send us at
that time.

Now I try to install the qgis 2.2 or qgis 2.4, and try to test it again.
Maybe I could do a video.

>(* insert the usual disclaimer about "no one should ever sponsor a
> feature without an explicit requirement for unit tests in the
> contract")

But for some enhancement like this, "the wms parameter scale" s quite
difficult to do a unit test.
Please pay attention that the enhancement funded from us was only in
the provider wms (the client).
The qgis-server almost at that time don't support this feature (the
abiity to have distinct scale rendering for the same layer).
This was a feature enhancement forward the other wms-server like
geoserver or mapserver.
So I guess will be quite difficult to do a UNIT-TEST.
Even if we want to do a unit-test, it need a wms service always
available and under the control of the qgis team, and obviously it
should be a qgis-server. So there was not more chance to be able to
test this enhancement that is specifically for the wms client
interface.

When is developed an enhancement for a WMS in a GIS client the test
should be done on more than the own wms service (qgis-server).
So need a mapserver, a geoserver and a qgis-server always availables.
I know this is quite impossible so also is quite impossible to do a
well formed unit-test in the wms section.

A.

2017-03-13 2:31 GMT+01:00 Nyall Dawson <nyall.dawson at gmail.com>:
> On 23 February 2017 at 17:34, Nyall Dawson <nyall.dawson at gmail.com> wrote:
>> On 23 February 2017 at 17:06, Andrea Peri <aperi2007 at gmail.com> wrote:
>>> Hi,
>>> we at Tuscany Region in 2013 fundedthe adding of the support of scale
>>> legend in the wms provider.
>>> That (if I remenber was added to the 2.2 or 2.4 version of qgis).
>>>
>>> These are the commit for that adding:
>>>
>>> https://github.com/qgis/QGIS/commit/8fb20f038d9d6c67bafc932ccea1d688ef2d5363
>>>
>>> https://github.com/qgis/QGIS/commit/f6a6a0c017db6306691dca9c9008ef0100574d98
>>>
>>> https://github.com/qgis/QGIS/commit/32da13b20c6a408a975d0e93415a73d20a3c4b59
>>>
>>> https://github.com/qgis/QGIS/commit/48002dbf18392c1d227983b97e8dc65d7eb1f6b6
>>>
>>> https://github.com/qgis/QGIS/commit/d7708437632972b293fe7a511008a910e61c001d
>>>
>>> From the 2.6 version as I remenber it was removed
>>
>> Which parts are missing or broken? I just quickly flicked through
>> those commits and the code at least all seems to still be there in
>> 2.18/master. (I didn't test the functionality, and noticed that
>> unfortunately there's no unit tests covering any of the original
>> work.)
>
> Andrea,
>
> What was the outcome here? Is there missing functionality?
>
> I'd like to know either way - it's a very bad situation for the
> project's reputation if funded features are broken like this*.
>
> Nyall
>
>
> (* insert the usual disclaimer about "no one should ever sponsor a
> feature without an explicit requirement for unit tests in the
> contract")



-- 
-----------------
Andrea Peri
. . . . . . . . .
qwerty àèìòù
-----------------