Re: [Qgis-developer] What is the name for "pyspatialite" package in fedora 19?

[童言]

I see! Thanks a 
lot!-The 
linked section about Fedora lists build instructions for Qt5 which
is still experimental and probably not what you want.

But the rest of the document contains a lot of useful information.

-- 

Matthias Kuhn







___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS Layer Search Plugin

[Jeremy Palmer]

What it is:
1. It provides the user the ability to search all layers and all fields against 
a particular search string. It may be that the user already knows the layer he 
is interested in, but that is not always the case. It is possible that the user 
has a number of layers that contain similar data. I really do intend for it to 
search multiple layers, but LayerSearch has the ability to select a single 
layer and single column if desired.
2. It does not require any pre configuration other than the search feature and 
whether you want an exact match, a match at the beginning of a data field, or a 
match anywhere in the data.
3. The resulting matches display what layer and column the match was found in 
and by clicking on the match QGIS will zoom to that feature.
4. While searching on a large data set you can click an explore the found 
features even while it is still searching.




Not possible to consider pushing the query filter down the provider?

What it is not:
The fastest search engine in the world. It is not meant to be used to search a 
reference data set like Quick Finder does. I would not want it to do any pre 
indexing or pre configuration. That is not its purpose. I just tried a geonames 
data set with 2.2 million records,  LayerSearch completed the task in 45 
seconds on my laptop. That is acceptable to me. In the mean time I was able to 
look at the matches it was finding. Even though I say this, I am interested in 
optimizing it to be as fast as possible, but I don't what the user to have to 
index the data or do any pre configuration first.



BTW when you talk about moving to "Processing" what are you talking about? Are 
you talking about the Processing toolbox or something else? I am new to the 
development side of QGIS.


Cheers
Jeremy



This message contains information, which may be in confidence and may be 
subject to legal privilege. If you are not the intended recipient, you must not 
peruse, use, disseminate, distribute or copy this message. If you have received 
this message in error, please notify us immediately (Phone 0800 665 463 or 
i...@linz.govt.nz) and destroy the original message. LINZ accepts no 
responsibility for changes to this email, or for any attachments, after its 
transmission from LINZ. Thank You.
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Report 2 - QGIS Symbology Sharing Tools

[Nathan Woodrow]

Using the raw end point sounds fine to me for now. Easy to change later if
needed.

On Thu, 9 Jun 2016 8:14 am Akbar Gumbira  wrote:

> Hi Ale,
>
> I just tried Dulwich. There are three ways that I see to checkout a single
> file: git sparse checkout, git archive, or shallow clone (still cloning all
> files though).
>
> With git sparse checkout, Dulwich does not provide the porcelain-level
> API. The author says that it's possible though (
> https://github.com/jelmer/dulwich/issues/405#issuecomment-224579678). I
> am not sure how much work to implement it dealing with the git object,
> tree, and blob itself at the plumbing level.
>
> With git archive, unfortunately Github doesn't allow us to do this (
> https://twitter.com/GitHubHelp/status/322818593748303873). WIth shallow
> clone left, I feel that I would over-engineer it forcing to use git just to
> be able to do this. I think it's more simple if for now we use the raw
> endpoint for each hosts that we know (Github, Bitbucket, etc), and we add
> more supported hosts as users put the collection in other hosts as long as
> we know the URL to the raw metadata file.
>
> What do you think?
>
> Cheers
>
> On Wed, Jun 8, 2016 at 9:51 AM, Akbar Gumbira 
> wrote:
>
>> Wasn't that a requisite? If we are going to use git as a storage we need
>>> a git client (pure python seems to exist, could be used as a fallback in
>>> case git is not installed in the system)
>>> https://github.com/jelmer/dulwich .
>>>
>>
>>
>>> BTW: if you do not want to deal with git in this first task, you could
>>> used the zip endpoint, and assume that the metadata will be available at a
>>> known location in the remote http repo and just wget it.
>>
>>
>> Ok. I was just trying to find solution that could work out well without
>> having git (as from the last discussion with Martin, it's better to avoid
>> using git for now). I will try pygit2 and dulwich and I'll report you back
>> what I thought using those two.
>>
>> On Wed, Jun 8, 2016 at 9:41 AM, Alessandro Pasotti 
>> wrote:
>>
>>> 2016-06-08 9:21 GMT+02:00 Akbar Gumbira :
>>>
 Yes, I have tried it. We can do it with sparse checkout, but, it
 requires git on the client.

>>>
>>>
>>> Wasn't that a requisite? If we are going to use git as a storage we need
>>> a git client (pure python seems to exist, could be used as a fallback in
>>> case git is not installed in the system)
>>> https://github.com/jelmer/dulwich .
>>>
>>> BTW: if you do not want to deal with git in this first task, you could
>>> used the zip endpoint, and assume that the metadata will be available at a
>>> known location in the remote http repo and just wget it.
>>>
>>>
>>>

 On Wed, Jun 8, 2016 at 9:18 AM, Alessandro Pasotti 
 wrote:

>
> 2016-06-08 9:10 GMT+02:00 Akbar Gumbira :
>
>> Hi Akbar,
>>> The most flexible installation tool that I know is probably python
>>> pip.
>>> pip can install software from a zip file, from git and from other
>>> sources too.
>>> I'd suggest you to have a look to pip implementation of the install
>>> functionality, maybe there is some interesting for you.
>>
>> I just read pip code base. I think I can pick something from there
>> for downloading the resources. But the problem I have right now is to get
>> only the metadata file from the repository. Or are you suggesting that 
>> when
>> users add a repository connection, it also downloads the repository
>> directly?
>>
>
>
> If you are working on the git repos you can probably fetch just one
> file without cloning the whole repo, I did not test it but here are
> probably some pointers:
>
>
> http://stackoverflow.com/questions/2466735/how-to-checkout-only-one-file-from-git-repository
>
> http://stackoverflow.com/questions/1125476/retrieve-a-single-file-from-a-repository
>
>
>
>
>>
>> On Mon, Jun 6, 2016 at 8:39 AM, Alessandro Pasotti <
>> apaso...@gmail.com> wrote:
>>
>>> 2016-06-05 10:13 GMT+02:00 Richard Duivenvoorde >> >:
>>>
 On 05-06-16 09:02, Akbar Gumbira wrote:

 > *Are you blocked on anything?*
 > ... In Github or Bitbucket they provide a direct link to
 > the raw file. But I think I should look at more general approach
 without
 > manipulating the URL depending on the host. If you have some
 input, I
 > would be happy to assess it.

 Thanks Akbar,

 I did some googling:

 http://stackoverflow.com/questions/14405782/git-fetch-single-file-from-remote-repository-programatically
 If you really want to keep it git, it looks like a shallow
 clone/copy is
 the only way? That post also talks about some 

Re: [Qgis-developer] Report 2 - QGIS Symbology Sharing Tools

[Akbar Gumbira]

Hi Ale,

I just tried Dulwich. There are three ways that I see to checkout a single
file: git sparse checkout, git archive, or shallow clone (still cloning all
files though).

With git sparse checkout, Dulwich does not provide the porcelain-level API.
The author says that it's possible though (
https://github.com/jelmer/dulwich/issues/405#issuecomment-224579678). I am
not sure how much work to implement it dealing with the git object, tree,
and blob itself at the plumbing level.

With git archive, unfortunately Github doesn't allow us to do this (
https://twitter.com/GitHubHelp/status/322818593748303873). WIth shallow
clone left, I feel that I would over-engineer it forcing to use git just to
be able to do this. I think it's more simple if for now we use the raw
endpoint for each hosts that we know (Github, Bitbucket, etc), and we add
more supported hosts as users put the collection in other hosts as long as
we know the URL to the raw metadata file.

What do you think?

Cheers

On Wed, Jun 8, 2016 at 9:51 AM, Akbar Gumbira 
wrote:

> Wasn't that a requisite? If we are going to use git as a storage we need a
>> git client (pure python seems to exist, could be used as a fallback in case
>> git is not installed in the system) https://github.com/jelmer/dulwich .
>>
>
>
>> BTW: if you do not want to deal with git in this first task, you could
>> used the zip endpoint, and assume that the metadata will be available at a
>> known location in the remote http repo and just wget it.
>
>
> Ok. I was just trying to find solution that could work out well without
> having git (as from the last discussion with Martin, it's better to avoid
> using git for now). I will try pygit2 and dulwich and I'll report you back
> what I thought using those two.
>
> On Wed, Jun 8, 2016 at 9:41 AM, Alessandro Pasotti 
> wrote:
>
>> 2016-06-08 9:21 GMT+02:00 Akbar Gumbira :
>>
>>> Yes, I have tried it. We can do it with sparse checkout, but, it
>>> requires git on the client.
>>>
>>
>>
>> Wasn't that a requisite? If we are going to use git as a storage we need
>> a git client (pure python seems to exist, could be used as a fallback in
>> case git is not installed in the system)
>> https://github.com/jelmer/dulwich .
>>
>> BTW: if you do not want to deal with git in this first task, you could
>> used the zip endpoint, and assume that the metadata will be available at a
>> known location in the remote http repo and just wget it.
>>
>>
>>
>>>
>>> On Wed, Jun 8, 2016 at 9:18 AM, Alessandro Pasotti 
>>> wrote:
>>>

 2016-06-08 9:10 GMT+02:00 Akbar Gumbira :

> Hi Akbar,
>> The most flexible installation tool that I know is probably python
>> pip.
>> pip can install software from a zip file, from git and from other
>> sources too.
>> I'd suggest you to have a look to pip implementation of the install
>> functionality, maybe there is some interesting for you.
>
> I just read pip code base. I think I can pick something from there for
> downloading the resources. But the problem I have right now is to get only
> the metadata file from the repository. Or are you suggesting that when
> users add a repository connection, it also downloads the repository
> directly?
>


 If you are working on the git repos you can probably fetch just one
 file without cloning the whole repo, I did not test it but here are
 probably some pointers:


 http://stackoverflow.com/questions/2466735/how-to-checkout-only-one-file-from-git-repository

 http://stackoverflow.com/questions/1125476/retrieve-a-single-file-from-a-repository




>
> On Mon, Jun 6, 2016 at 8:39 AM, Alessandro Pasotti  > wrote:
>
>> 2016-06-05 10:13 GMT+02:00 Richard Duivenvoorde 
>> :
>>
>>> On 05-06-16 09:02, Akbar Gumbira wrote:
>>>
>>> > *Are you blocked on anything?*
>>> > ... In Github or Bitbucket they provide a direct link to
>>> > the raw file. But I think I should look at more general approach
>>> without
>>> > manipulating the URL depending on the host. If you have some
>>> input, I
>>> > would be happy to assess it.
>>>
>>> Thanks Akbar,
>>>
>>> I did some googling:
>>>
>>> http://stackoverflow.com/questions/14405782/git-fetch-single-file-from-remote-repository-programatically
>>> If you really want to keep it git, it looks like a shallow
>>> clone/copy is
>>> the only way? That post also talks about some undocumented feature,
>>> but
>>> I would not depend on that?
>>>
>>> Personally I would be ok when both Github and Gitlab/Gog would work
>>> (as
>>> both a closed source and open source member of the git-web-world)...
>>>
>>> Or: a script running somewhere on our server, (shallow) 

[Qgis-developer] R: Duplicate plugins: proposal

[Niccolò Marchi]

(unfortunately) No suggestions for improvement except for setting that sentence 
as first and highlighting it as much as possible :)



The additional page was absolutely not meant to be on your shoulders but up to 
each developer with valid access, as a sort of “blackboard for notes”. When 
someone is testing more plugins dealing with a specific topic and finds similar 
functionalities, he can just add his notes as a reminder (open to the public). 
Pretty “informal”, let’s say, but useful to cluster plugins according to the 
effort of everyone who wants to contribute, instead of basing everything on 
your exclusive knowledge of each plugin.



Thanks for your time,



Niccolò







Da: Paolo Cavallini
Inviato: mercoledì 8 giugno 2016 18:11
A: Niccolò Marchi; 
qgis-developer@lists.osgeo.org
Oggetto: Re: Duplicate plugins: proposal



Hi Niccolò,
thanks for your suggestions:

Il 07/06/2016 18:55, Niccolò Marchi ha scritto:

> 1.   An advice to “check for improvements before creating new and
> get in contact with the author”,

it is already here: http://plugins.qgis.org/

does not duplicate of existing functionalities or plugin, unless there
is a good reason

Please suggest wording improement if you think i is not clear enough.

> 2.   And/or an additional page (read-only for users) dealing with
> “proposed merges” where , the qgis developers who are testing more
> plugins similar to each other, can resume which features can be
> collected and from where.
>
> In the latter case, after a while, it could be then useful to send a
> general email to the plugin devs to get in contact with each other
> according to the “clustered scheme” that dinamically will be created.
> Plugin developers will still be free to keep their own plugin but, at
> least, it can stimulate the more collaborative ones. A sort of
> coordination proposal about what the project “needs”.
>
> In the worst  case, it may work as a general framework for future funded
> (?) projects.

Maintaining such a page is quite time consuming, and I cannot afford it.
What I regularly do is to suggest, directly to the developer, a merge
before approving new plugins, any time I see a potential for it.
This has met some very nice success.

All the best.

--
Paolo Cavallini - www.faunalia.eu
QGIS & PostGIS courses: http://www.faunalia.eu/training.html
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS/Linux on high-resolution displays

[Barry Rowlingson]

On Wed, Jun 8, 2016 at 6:24 AM, Alessandro Pasotti  wrote:
> 2016-06-07 23:17 GMT+02:00 Barry Rowlingson :
>>
>> Ive just bought a Dell XPS13 which came with Ubuntu 14.04
>> pre-installed. The screen is 3200 pixels across. Getting applications
>> to scale properly is a nightmare.
>>
>> QGIS (and I guess all Qt apps) seems to be the worst. The icons are
>> microscopic, and the text and other interface elements overlap. It is
>> unusable. Same for Gimp and Inkscape, but I mostly use QGIS.
>
>
>
> QGIS works pretty well (with some minor glitches) in KDE on 4K and retina
> screens, see the tips and the links in this article
> http://www.itopen.it/qgis-and-qt-getting-ready-for-hidpi-screens/

Hmm I had more than minor glitches. The problem is that there were so
many possible things to tweak I could never be sure what was helping!

> No, try setting:
> - the dpi resolution in the display configuration of your OS (system
> settings), that is usually a wrapper for xrandr
> - for QT5, the env vars  QT_DEVICE_PIXEL_RATIO=2 and/or
> QT_SCREEN_SCALE_FACTORS=2
> - icon size in QGIS can be (at least in part, for the toolbar) configured in
> the QGIS options dialog
>
> See also: http://doc.qt.io/qt-5/highdpi.html
>

Since those are all Qt5 things they're no good with QGIS at the
moment, yes? Or is there a QGIS built against Qt5.6+ that I could be
using?

 I've got the icons scaled up now, but there's still some tiny UI
elements - the visibility toggles on layers, and the little plus signs
that expand trees are almost invisible...

thanks for the tips

Barry
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS Server performance deterioration

[René-Luc Dhont]

Hi Devs,

I have created a PR: https://github.com/qgis/QGIS/pull/3186

QgsEditorWidgetRegistry connects to appropriate signals from map layers 
to load and save style but never disconnects its. In the context of QGIS 
Server, layers are connecting to appropriate signals at each request.


Thanks to review it.

Regards,

Le 08/06/2016 17:50, René-Luc Dhont a écrit :

Hi Devs,

I found which object is connected to QgsMapLayerRegistry through 
layerWasAdded slot in QGIS Server, and it's QgsEditorWidgetRegistry !


The method mapLayerAdded is apply to the layer added:
```
void QgsEditorWidgetRegistry::mapLayerAdded( QgsMapLayer* mapLayer )
{
  QgsVectorLayer* vl = qobject_cast( mapLayer );

  if ( vl )
  {
connect( vl, SIGNAL( readCustomSymbology( const QDomElement&, 
QString& ) ), this, SLOT( readSymbology( const QDomElement&, QString& 
) ) );
connect( vl, SIGNAL( writeCustomSymbology( QDomElement&, 
QDomDocument&, QString& ) ), this, SLOT( writeSymbology( QDomElement&, 
QDomDocument&, QString& ) ) );

  }
}
```

Could a method call `mapLayerWillBeRemoved` be added ?


Le 07/06/2016 15:58, René-Luc Dhont a écrit :

Hi devs,

David Marteau, our colleague in 3Liz, has found some serious 
deteriorations of performance in QGIS Server.


The first one concerne QgsExpressionContext which is not clean up in 
QgsServer::handleRequest


Each call to QgsMapLayer::draw add items to QgsExpressionContext scope.
The scope is not cleaned between request and thus the scope list to 
grow indefinitely.


Having the list growing indefinitely deteriorate seriously 
performances in the long run. It also increase the memory footprint 
but in a less noticiable way.


This behavior has been checked by monitoring reponse time of several 
ten of thousand identical requests during server long runs.


This has been fixed in master and in LTR:
https://github.com/qgis/QGIS/commit/5c3aa51e80887de6ff76682a42814041bb32cd5b 

https://github.com/qgis/QGIS/commit/50181eee00709e3f7e29b2b53282f28bdedfe265 



Now we encounter an other issue with deteriorationof performance in 
QGIS Server.


This issue is due to the slot 'layerWasAdded' in QgsMapLayerRegistry. 
The duration of the emit layerWasAdded grows with the number of 
drawing. Do you know what it is done when 'layerWasAdded' is emitted ?


Regards,
René-Luc D'Hont
3Liz




___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS Server performance deterioration

[René-Luc Dhont]

Hi Devs,

I found which object is connected to QgsMapLayerRegistry through 
layerWasAdded slot in QGIS Server, and it's QgsEditorWidgetRegistry !


The method mapLayerAdded is apply to the layer added:
```
void QgsEditorWidgetRegistry::mapLayerAdded( QgsMapLayer* mapLayer )
{
  QgsVectorLayer* vl = qobject_cast( mapLayer );

  if ( vl )
  {
connect( vl, SIGNAL( readCustomSymbology( const QDomElement&, 
QString& ) ), this, SLOT( readSymbology( const QDomElement&, QString& ) ) );
connect( vl, SIGNAL( writeCustomSymbology( QDomElement&, 
QDomDocument&, QString& ) ), this, SLOT( writeSymbology( QDomElement&, 
QDomDocument&, QString& ) ) );

  }
}
```

Could a method call `mapLayerWillBeRemoved` be added ?


Le 07/06/2016 15:58, René-Luc Dhont a écrit :

Hi devs,

David Marteau, our colleague in 3Liz, has found some serious 
deteriorations of performance in QGIS Server.


The first one concerne QgsExpressionContext which is not clean up in 
QgsServer::handleRequest


Each call to QgsMapLayer::draw add items to QgsExpressionContext scope.
The scope is not cleaned between request and thus the scope list to 
grow indefinitely.


Having the list growing indefinitely deteriorate seriously 
performances in the long run. It also increase the memory footprint 
but in a less noticiable way.


This behavior has been checked by monitoring reponse time of several 
ten of thousand identical requests during server long runs.


This has been fixed in master and in LTR:
https://github.com/qgis/QGIS/commit/5c3aa51e80887de6ff76682a42814041bb32cd5b 

https://github.com/qgis/QGIS/commit/50181eee00709e3f7e29b2b53282f28bdedfe265 



Now we encounter an other issue with deteriorationof performance in 
QGIS Server.


This issue is due to the slot 'layerWasAdded' in QgsMapLayerRegistry. 
The duration of the emit layerWasAdded grows with the number of 
drawing. Do you know what it is done when 'layerWasAdded' is emitted ?


Regards,
René-Luc D'Hont
3Liz


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS Layer Search Plugin

[C Hamilton]

With these discussions I want to make it clear what the LayerSearch plugin
use is intended for and what it is not.


What it is:

1. It provides the user the ability to search all layers and all fields
against a particular search string. It may be that the user already knows
the layer he is interested in, but that is not always the case. It is
possible that the user has a number of layers that contain similar data. I
really do intend for it to search multiple layers, but LayerSearch has the
ability to select a single layer and single column if desired.

2. It does not require any pre configuration other than the search feature
and whether you want an exact match, a match at the beginning of a data
field, or a match anywhere in the data.

3. The resulting matches display what layer and column the match was found
in and by clicking on the match QGIS will zoom to that feature.

4. While searching on a large data set you can click an explore the found
features even while it is still searching.



What it is not:

The fastest search engine in the world. It is not meant to be used to
search a reference data set like Quick Finder does. I would not want it to
do any pre indexing or pre configuration. That is not its purpose. I just
tried a geonames data set with 2.2 million records,  LayerSearch completed
the task in 45 seconds on my laptop. That is acceptable to me. In the mean
time I was able to look at the matches it was finding. Even though I say
this, I am interested in optimizing it to be as fast as possible, but I
don't what the user to have to index the data or do any pre configuration
first.



BTW when you talk about moving to "Processing" what are you talking about?
Are you talking about the Processing toolbox or something else? I am new to
the development side of QGIS.

All the best,

Calvin



On Wed, Jun 8, 2016 at 2:20 AM, Denis Rouzaud 
wrote:

> Dear all,
>
> Indeed, the layer search offers something Quickfinder doesn't: searching
> on all layers without any configuration, and the use of operators.
> Although I think choosing an operator in a global search is not very
> required (why doing price < 1000 on all layers, there is a pretty good
> chance you know what layer you're looking for).
>
> From this, I think Michael's idea is pretty good.
> Make QuickFinder just a search bar in the application toolbar. Then you
> can register any search process to it, which would be a processing alg:
> might be OSM, FTS or Layer search.
>
> I would definitely be ok to do so for QuickFinder.
> But the question is when... It would take me a bit of time to reorganize
> the plugin, make FTS generation and search as processing algs (never done
> that before). I won't be able to achieve this before August.
>
> So...is this fair to ask someone to wait that long for a dev and then ask
> to modify its code to fit in? I have no answer, it's a real question.
>
>
> Best wishes,
> Denis
>
>
> PS: yes, QuickFinder FTS search handles multiple columns as it supports
> expressions (you can concatenate the columns).
>
>
>
>
>
>
>
> On 06/07/2016 03:27 PM, kimaidou wrote:
>
> Hi all,
>
> I think it is not really profitable to have multiple plugins. In my
> opinion, efforts must be made to merge plugins when possible. In this case,
> we have
>
> * Quick Finder : Full Text Search is used here, which allows fuzzy search,
> and is very efficient compared to scanning all features for a match. A
> drawback, you must fill in the "FTS vectors" for each layer you want to be
> able to search within. I think it is not so bad because the process is
> fast. I also think there is nothing preventing to add more than one field
> in the search ( it is possible with FTS in PostgreSQL, must also be the
> case with sqlite)
>
> * Layer search : Great to have a quick way to search among all the fields
> of all loaded layers, with no configuration. Drawback : I think this can be
> very slow for big datasets, and it is not "fuzzy", as regexp are used while
> iterating on each feature. Another drawback : for external vector sources
> like PostgreSQL, all features must been fetched !
>
> IMHO, the way to go could be
>
> * Add the FTS pregenerator of Quick Finder in Processing, as an alg
> available for anyone to use
> * Add the FTS search of Quick Finder as a Processing Alg
> * Add the "layer search" worker as an Alg in Processing
> * Have a combined plugin which let the user choose the search type : FTS
> with pregeneration, or direct search with features iteration. This will
> then run the corresponding alg
>
> Cheers,
> Michaël
>
> 2016-06-07 7:16 GMT+02:00 Jeremy Palmer :
>
>>
>> >
>> > I would personally prefer keeping it as a separate plugin, but what do
>> you
>> > think? Should it be integrated with another plugin?
>> >
>>
>> It seems like it has value over the other plugins. Another issue I’ve had
>> is searching all layer fields and pushing the query filter down the
>> provider (e.g PostGIS or WFS) 

Re: [Qgis-developer] Build issue on Ubuntu 16.04 xenial with release-2_14 : crssync target failed

[Yves Jacolin]

On Wednesday, June 08, 2016 13:49:53 kimaidou wrote:
> Hi Bas,
> 
> Thanks for your answer.
> It appears I built gdal on year ago to check GeoJson write capabilities in
> QGIS.
> 
> Now, I have indeed just upgraded to last Ubuntu xenial, and I am trying to
> get rid of old GDAL build and reinstall GDAL from repositories. So I
> removed all symlinks made earlier today, I remove old Gdal build, remove
> all lines concerning GDAL in /etc/ld.so.conf.d/. I ran ldconfig
> 
> Then uninstalled qgis and gdal-bin + libgdal and others, restarted my
> computer
> 
> It still does not work.
> mdouchin@mdouchin-Lenovo-U41-70:~$ ldd /usr/local/bin/gdalinfo | grep "not
> found"
> libnetcdf.so.7 => not found
> libgif.so.4 => not found
> 
> I will do some more googling
If you install Gdal from package, why do you have such path to your gdal: 
/usr/local/bin/gdalinfo 

How did you clean up gdal build? "make clean" don't remove gdal installation, 
you should remove the directory by hand.

Y.
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Build issue on Ubuntu 16.04 xenial with release-2_14 : crssync target failed

[kimaidou]

Hi all,

Setting the compilation variables via ccmake to point to libgdal 1.11.3
worked like a charm. QGIS 2.14 has successfully been built against gdal
1.11.3

Bas, you are right, I inverted them in my previous email. Here are the vars
used in ccmake to build QGIS successfully on my machine

GDAL_CONFIG  /usr/bin/gdal-config
GDAL_INCLUDE_DIR /usr/include/gda
GDAL_LIBRARY /usr/lib/libgdal.so

Thanks again, both Bas and Yves, for you kind help

Cheers
Michaël

2016-06-08 14:36 GMT+02:00 Bas Couwenberg :

> On 2016-06-08 14:17, kimaidou wrote:
>
>> It appears I had 2 gdal installed :
>>
>> one by qgis from qgis.org repository ( 1.11) which corresponds to
>> /usr/lib/libgdal.so
>> and one installed via apt-get install gdal-bin from repositories (2.1)
>> which corresponds to : /usr/local/lib/libgdal.so
>>
>
> You got this the wrong way around, the package gdal is installed in
> /usr/lib, you custom build in /usr/local/lib.
>
> The /usr/local hierarchy is reserverd for the local administrator, /usr
> for the system.
>
>
> Kind Regards,
>
> Bas
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS Server performance deterioration

[Luca Manganelli]

On Wed, Jun 8, 2016 at 8:06 AM, Luca Manganelli  wrote:
> On Tue, Jun 7, 2016 at 5:52 PM, René-Luc Dhont  wrote:
>> [...]

I compiled a debug version of QGIS Server. It seems that there's a
memory leak when it reads the QGS project file, after some project
switching in QWC project list:

Program received signal SIGSEGV, Segmentation fault.
0xb6f80d11 in QBasicAtomicInt::ref (this=0x7fff) at
/usr/include/qt4/QtCore/qatomic_i386.h:120
120 : "memory");
(gdb) bt
#0  0xb6f80d11 in QBasicAtomicInt::ref (this=0x7fff) at
/usr/include/qt4/QtCore/qatomic_i386.h:120
#1  0xb6f82ce4 in QString::QString (this=0xafc1b2c, other=...) at
/usr/include/qt4/QtCore/qstring.h:726
#2  0xb7238b33 in QgsMapLayer::id (this=0xab1fc20) at
/usr/local/src/qgis-2.14.3/src/core/qgsmaplayer.cpp:107
#3  0xb7093f85 in QgsLayerTreeLayer::QgsLayerTreeLayer
(this=0xafc1b10, layer=0xab1fc20)
at /usr/local/src/qgis-2.14.3/src/core/layertree/qgslayertreelayer.cpp:27
#4  0xb709462e in QgsLayerTreeLayer::readXML (element=...) at
/usr/local/src/qgis-2.14.3/src/core/layertree/qgslayertreelayer.cpp:111
#5  0xb70a724d in QgsLayerTreeNode::readXML (element=...) at
/usr/local/src/qgis-2.14.3/src/core/layertree/qgslayertreenode.cpp:56
#6  0xb7091d85 in QgsLayerTreeGroup::readChildrenFromXML
(this=0xaf28218, element=...)
at /usr/local/src/qgis-2.14.3/src/core/layertree/qgslayertreegroup.cpp:293
#7  0xb70917c0 in QgsLayerTreeGroup::readXML (element=...) at
/usr/local/src/qgis-2.14.3/src/core/layertree/qgslayertreegroup.cpp:259
#8  0xb70a7200 in QgsLayerTreeNode::readXML (element=...) at
/usr/local/src/qgis-2.14.3/src/core/layertree/qgslayertreenode.cpp:54
#9  0xb7091d85 in QgsLayerTreeGroup::readChildrenFromXML
(this=0xada4e48, element=...)
at /usr/local/src/qgis-2.14.3/src/core/layertree/qgslayertreegroup.cpp:293
#10 0xb70917c0 in QgsLayerTreeGroup::readXML (element=...) at
/usr/local/src/qgis-2.14.3/src/core/layertree/qgslayertreegroup.cpp:259
#11 0x080f9da2 in QgsServerProjectParser::findLegendGroupElements
(this=0xafc7df0)
at /usr/local/src/qgis-2.14.3/src/server/qgsserverprojectparser.cpp:1209
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Build issue on Ubuntu 16.04 xenial with release-2_14 : crssync target failed

[Bas Couwenberg]

On 2016-06-08 14:17, kimaidou wrote:

It appears I had 2 gdal installed :

one by qgis from qgis.org repository ( 1.11) which corresponds to
/usr/lib/libgdal.so
and one installed via apt-get install gdal-bin from repositories (2.1)
which corresponds to : /usr/local/lib/libgdal.so


You got this the wrong way around, the package gdal is installed in 
/usr/lib, you custom build in /usr/local/lib.


The /usr/local hierarchy is reserverd for the local administrator, /usr 
for the system.


Kind Regards,

Bas
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Build issue on Ubuntu 16.04 xenial with release-2_14 : crssync target failed

[kimaidou]

Hi Bas,

Thanks for your answer.
It appears I built gdal on year ago to check GeoJson write capabilities in
QGIS.

Now, I have indeed just upgraded to last Ubuntu xenial, and I am trying to
get rid of old GDAL build and reinstall GDAL from repositories. So I
removed all symlinks made earlier today, I remove old Gdal build, remove
all lines concerning GDAL in /etc/ld.so.conf.d/. I ran ldconfig

Then uninstalled qgis and gdal-bin + libgdal and others, restarted my
computer

It still does not work.
mdouchin@mdouchin-Lenovo-U41-70:~$ ldd /usr/local/bin/gdalinfo | grep "not
found"
libnetcdf.so.7 => not found
libgif.so.4 => not found

I will do some more googling
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Build issue on Ubuntu 16.04 xenial with release-2_14 : crssync target failed

[Bas Couwenberg]

On 2016-06-08 11:46, kimaidou wrote:
Here are the command lines used to make gdalinfo happy (quite hackish 
... )


ln -s /usr/lib/x86_64-linux-gnu/libgif.so
/usr/lib/x86_64-linux-gnu/libgif.so.4
ln -s /usr/lib/x86_64-linux-gnu/libnetcdf.so
/usr/lib/x86_64-linux-gnu/libnetcdf.so.7
echo "/usr/lib/x86_64-linux-gnu/" >> /etc/ld.so.conf.d/qgis.conf
ldconfig


Please don't ever create your own library symlinks, this is always 
wrong.


You need to rebuild your GDAL with the currently installed dependencies, 
as it looks like you upgraded the system packages but never rebuilt your 
custom GDAL after you upgraded your Ubuntu system to xenial.


Kind Regards,

Bas


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Build issue on Ubuntu 16.04 xenial with release-2_14 : crssync target failed

[kimaidou]

Ok, I think I found the issue. I have a problem with GDAL

gdalinfo --formats

returns

gdalinfo: error while loading shared libraries: libnetcdf.so.7: cannot open
shared object file: No such file or directory

Here are the command lines used to make gdalinfo happy (quite hackish ... )

ln -s /usr/lib/x86_64-linux-gnu/libgif.so
/usr/lib/x86_64-linux-gnu/libgif.so.4
ln -s /usr/lib/x86_64-linux-gnu/libnetcdf.so
/usr/lib/x86_64-linux-gnu/libnetcdf.so.7
echo "/usr/lib/x86_64-linux-gnu/" >> /etc/ld.so.conf.d/qgis.conf
ldconfig

Now gdalinfo does not complain anymore, but the build does not succeed
either, see
https://paste.sh/l2Jdh0Bt#5wrhkbMFV3nUwfnsHdLhfZ5U



2016-06-08 11:14 GMT+02:00 kimaidou :

> I just tested if with one proc only, which helps to see the errors. Here
> they are
>
> https://paste.sh/24JYwaQj#WXMylljn7ATP3KB7UM2TdbhZ
>
>
>
> 2016-06-08 10:45 GMT+02:00 kimaidou :
>
>> Hi all
>>
>> I am trying to build QGIS release-2_14 in my Ubuntu 16.04. I have
>> installed dependencies listed in the INSTALL.md file, and tried it with a
>> clean build directory and ccache emptied
>>
>> I use QGIS official repo
>> deb   http://qgis.org/debian xenial main
>> deb-src http://qgis.org/debian xenial main
>>
>>
>> Here is my ccmake cl :
>>
>> cd /home/mdouchin/src/qgis/build/2_14/
>> ccmake -DCMAKE_INSTALL_PREFIX=/home/mdouchin/src/qgis/build/2_14
>> -DCMAKE_BUILD_TYPE=Release -DENABLE_TESTS=False
>> -DWITH_INTERNAL_QWTPOLAR=True -DWITH_SERVER=True -DWITH_GRASS=False
>> -DWITH_GRASS7=False ../../source/QGIS/
>>
>> Here are the errors during the build process
>>
>> src/crssync/CMakeFiles/crssync.dir/build.make:112 : la recette pour la
>> cible « output/bin/crssync » a échouée
>>
>> I have no really other helpfull erros here. Has anyone had problem with
>> crssync with last ubuntu ?
>>
>>
>> Thanks in advance
>> Michaël
>>
>
>
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Build issue on Ubuntu 16.04 xenial with release-2_14 : crssync target failed

[kimaidou]

I just tested if with one proc only, which helps to see the errors. Here
they are

https://paste.sh/24JYwaQj#WXMylljn7ATP3KB7UM2TdbhZ



2016-06-08 10:45 GMT+02:00 kimaidou :

> Hi all
>
> I am trying to build QGIS release-2_14 in my Ubuntu 16.04. I have
> installed dependencies listed in the INSTALL.md file, and tried it with a
> clean build directory and ccache emptied
>
> I use QGIS official repo
> deb   http://qgis.org/debian xenial main
> deb-src http://qgis.org/debian xenial main
>
>
> Here is my ccmake cl :
>
> cd /home/mdouchin/src/qgis/build/2_14/
> ccmake -DCMAKE_INSTALL_PREFIX=/home/mdouchin/src/qgis/build/2_14
> -DCMAKE_BUILD_TYPE=Release -DENABLE_TESTS=False
> -DWITH_INTERNAL_QWTPOLAR=True -DWITH_SERVER=True -DWITH_GRASS=False
> -DWITH_GRASS7=False ../../source/QGIS/
>
> Here are the errors during the build process
>
> src/crssync/CMakeFiles/crssync.dir/build.make:112 : la recette pour la
> cible « output/bin/crssync » a échouée
>
> I have no really other helpfull erros here. Has anyone had problem with
> crssync with last ubuntu ?
>
>
> Thanks in advance
> Michaël
>
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Build issue on Ubuntu 16.04 xenial with release-2_14 : crssync target failed

[kimaidou]

Hi all

I am trying to build QGIS release-2_14 in my Ubuntu 16.04. I have installed
dependencies listed in the INSTALL.md file, and tried it with a clean build
directory and ccache emptied

I use QGIS official repo
deb   http://qgis.org/debian xenial main
deb-src http://qgis.org/debian xenial main


Here is my ccmake cl :

cd /home/mdouchin/src/qgis/build/2_14/
ccmake -DCMAKE_INSTALL_PREFIX=/home/mdouchin/src/qgis/build/2_14
-DCMAKE_BUILD_TYPE=Release -DENABLE_TESTS=False
-DWITH_INTERNAL_QWTPOLAR=True -DWITH_SERVER=True -DWITH_GRASS=False
-DWITH_GRASS7=False ../../source/QGIS/

Here are the errors during the build process

src/crssync/CMakeFiles/crssync.dir/build.make:112 : la recette pour la
cible « output/bin/crssync » a échouée

I have no really other helpfull erros here. Has anyone had problem with
crssync with last ubuntu ?


Thanks in advance
Michaël
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Its visual changelog time! QGIS 2.16

[Tim Sutton]

Hi

> On 08 Jun 2016, at 10:02, Richard Duivenvoorde  wrote:
> 
> On 07-06-16 23:25, Tim Sutton wrote:
>> 
>> I have started a new changelog for 2.16. If you have written a cool new
>> feature for 2.16 could you help me by writing up a little entry for it
>> at: http://changelog.inasafe.org/en/qgis/version/2.16.0/
> 
> You mean
>   http://changelog.qgis.org/en/qgis/version/2.16.0/
> I think?
> ( I can log in in that one all empty )

Sorry about that, yes its the same thing, just different URL aliases. But yes 
it will be better to use the QGIS domain one in this context.

Regards

Tim

> 
> Richard
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer




---

Tim Sutton
QGIS Project Steering Committee Chair
t...@qgis.org




___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Its visual changelog time! QGIS 2.16

[Richard Duivenvoorde]

On 07-06-16 23:25, Tim Sutton wrote:
> 
> I have started a new changelog for 2.16. If you have written a cool new
> feature for 2.16 could you help me by writing up a little entry for it
> at: http://changelog.inasafe.org/en/qgis/version/2.16.0/

You mean
http://changelog.qgis.org/en/qgis/version/2.16.0/
I think?
( I can log in in that one all empty )

Richard
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Report 2 - QGIS Symbology Sharing Tools

[Akbar Gumbira]

>
> Wasn't that a requisite? If we are going to use git as a storage we need a
> git client (pure python seems to exist, could be used as a fallback in case
> git is not installed in the system) https://github.com/jelmer/dulwich .
>


> BTW: if you do not want to deal with git in this first task, you could
> used the zip endpoint, and assume that the metadata will be available at a
> known location in the remote http repo and just wget it.


Ok. I was just trying to find solution that could work out well without
having git (as from the last discussion with Martin, it's better to avoid
using git for now). I will try pygit2 and dulwich and I'll report you back
what I thought using those two.

On Wed, Jun 8, 2016 at 9:41 AM, Alessandro Pasotti 
wrote:

> 2016-06-08 9:21 GMT+02:00 Akbar Gumbira :
>
>> Yes, I have tried it. We can do it with sparse checkout, but, it requires
>> git on the client.
>>
>
>
> Wasn't that a requisite? If we are going to use git as a storage we need a
> git client (pure python seems to exist, could be used as a fallback in case
> git is not installed in the system) https://github.com/jelmer/dulwich .
>
> BTW: if you do not want to deal with git in this first task, you could
> used the zip endpoint, and assume that the metadata will be available at a
> known location in the remote http repo and just wget it.
>
>
>
>>
>> On Wed, Jun 8, 2016 at 9:18 AM, Alessandro Pasotti 
>> wrote:
>>
>>>
>>> 2016-06-08 9:10 GMT+02:00 Akbar Gumbira :
>>>
 Hi Akbar,
> The most flexible installation tool that I know is probably python pip.
> pip can install software from a zip file, from git and from other
> sources too.
> I'd suggest you to have a look to pip implementation of the install
> functionality, maybe there is some interesting for you.

 I just read pip code base. I think I can pick something from there for
 downloading the resources. But the problem I have right now is to get only
 the metadata file from the repository. Or are you suggesting that when
 users add a repository connection, it also downloads the repository
 directly?

>>>
>>>
>>> If you are working on the git repos you can probably fetch just one file
>>> without cloning the whole repo, I did not test it but here are probably
>>> some pointers:
>>>
>>>
>>> http://stackoverflow.com/questions/2466735/how-to-checkout-only-one-file-from-git-repository
>>>
>>> http://stackoverflow.com/questions/1125476/retrieve-a-single-file-from-a-repository
>>>
>>>
>>>
>>>

 On Mon, Jun 6, 2016 at 8:39 AM, Alessandro Pasotti 
 wrote:

> 2016-06-05 10:13 GMT+02:00 Richard Duivenvoorde :
>
>> On 05-06-16 09:02, Akbar Gumbira wrote:
>>
>> > *Are you blocked on anything?*
>> > ... In Github or Bitbucket they provide a direct link to
>> > the raw file. But I think I should look at more general approach
>> without
>> > manipulating the URL depending on the host. If you have some input,
>> I
>> > would be happy to assess it.
>>
>> Thanks Akbar,
>>
>> I did some googling:
>>
>> http://stackoverflow.com/questions/14405782/git-fetch-single-file-from-remote-repository-programatically
>> If you really want to keep it git, it looks like a shallow clone/copy
>> is
>> the only way? That post also talks about some undocumented feature,
>> but
>> I would not depend on that?
>>
>> Personally I would be ok when both Github and Gitlab/Gog would work
>> (as
>> both a closed source and open source member of the git-web-world)...
>>
>> Or: a script running somewhere on our server, (shallow) cloning all
>> registred repositories periodically, and making just the metadata.txt
>> files available via http/webserver? (maybe giving us some time to
>> check
>> the repo's on structure and (malicious?) content?
>>
>> Or else: a django app for the metadata...
>>
>> Regards,
>>
>> Richard
>>
>>
>>
>
> Hi Akbar,
>
> The most flexible installation tool that I know is probably python pip.
>
> pip can install software from a zip file, from git and from other
> sources too.
>
> I'd suggest you to have a look to pip implementation of the install
> functionality, maybe there is some interesting for you.
>
>
>
> --
> Alessandro Pasotti
> w3:   www.itopen.it
>



 --

 *---*
 *Akbar Gumbira *
 *www.akbargumbira.com *

>>>
>>>
>>>
>>> --
>>> Alessandro Pasotti
>>> w3:   www.itopen.it
>>>
>>
>>
>>
>> --
>>
>> *---*
>> *Akbar Gumbira *
>> *www.akbargumbira.com *
>>
>
>
>
> --
> Alessandro Pasotti
> w3:   www.itopen.it
>



-- 

*---*

Re: [Qgis-developer] Report 2 - QGIS Symbology Sharing Tools

[Alessandro Pasotti]

2016-06-08 9:21 GMT+02:00 Akbar Gumbira :

> Yes, I have tried it. We can do it with sparse checkout, but, it requires
> git on the client.
>


Wasn't that a requisite? If we are going to use git as a storage we need a
git client (pure python seems to exist, could be used as a fallback in case
git is not installed in the system) https://github.com/jelmer/dulwich .

BTW: if you do not want to deal with git in this first task, you could used
the zip endpoint, and assume that the metadata will be available at a known
location in the remote http repo and just wget it.



>
> On Wed, Jun 8, 2016 at 9:18 AM, Alessandro Pasotti 
> wrote:
>
>>
>> 2016-06-08 9:10 GMT+02:00 Akbar Gumbira :
>>
>>> Hi Akbar,
 The most flexible installation tool that I know is probably python pip.
 pip can install software from a zip file, from git and from other
 sources too.
 I'd suggest you to have a look to pip implementation of the install
 functionality, maybe there is some interesting for you.
>>>
>>> I just read pip code base. I think I can pick something from there for
>>> downloading the resources. But the problem I have right now is to get only
>>> the metadata file from the repository. Or are you suggesting that when
>>> users add a repository connection, it also downloads the repository
>>> directly?
>>>
>>
>>
>> If you are working on the git repos you can probably fetch just one file
>> without cloning the whole repo, I did not test it but here are probably
>> some pointers:
>>
>>
>> http://stackoverflow.com/questions/2466735/how-to-checkout-only-one-file-from-git-repository
>>
>> http://stackoverflow.com/questions/1125476/retrieve-a-single-file-from-a-repository
>>
>>
>>
>>
>>>
>>> On Mon, Jun 6, 2016 at 8:39 AM, Alessandro Pasotti 
>>> wrote:
>>>
 2016-06-05 10:13 GMT+02:00 Richard Duivenvoorde :

> On 05-06-16 09:02, Akbar Gumbira wrote:
>
> > *Are you blocked on anything?*
> > ... In Github or Bitbucket they provide a direct link to
> > the raw file. But I think I should look at more general approach
> without
> > manipulating the URL depending on the host. If you have some input, I
> > would be happy to assess it.
>
> Thanks Akbar,
>
> I did some googling:
>
> http://stackoverflow.com/questions/14405782/git-fetch-single-file-from-remote-repository-programatically
> If you really want to keep it git, it looks like a shallow clone/copy
> is
> the only way? That post also talks about some undocumented feature, but
> I would not depend on that?
>
> Personally I would be ok when both Github and Gitlab/Gog would work (as
> both a closed source and open source member of the git-web-world)...
>
> Or: a script running somewhere on our server, (shallow) cloning all
> registred repositories periodically, and making just the metadata.txt
> files available via http/webserver? (maybe giving us some time to check
> the repo's on structure and (malicious?) content?
>
> Or else: a django app for the metadata...
>
> Regards,
>
> Richard
>
>
>

 Hi Akbar,

 The most flexible installation tool that I know is probably python pip.

 pip can install software from a zip file, from git and from other
 sources too.

 I'd suggest you to have a look to pip implementation of the install
 functionality, maybe there is some interesting for you.



 --
 Alessandro Pasotti
 w3:   www.itopen.it

>>>
>>>
>>>
>>> --
>>>
>>> *---*
>>> *Akbar Gumbira *
>>> *www.akbargumbira.com *
>>>
>>
>>
>>
>> --
>> Alessandro Pasotti
>> w3:   www.itopen.it
>>
>
>
>
> --
>
> *---*
> *Akbar Gumbira *
> *www.akbargumbira.com *
>



-- 
Alessandro Pasotti
w3:   www.itopen.it
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Report 2 - QGIS Symbology Sharing Tools

[Alessandro Pasotti]

2016-06-08 9:10 GMT+02:00 Akbar Gumbira :

> Hi Akbar,
>> The most flexible installation tool that I know is probably python pip.
>> pip can install software from a zip file, from git and from other sources
>> too.
>> I'd suggest you to have a look to pip implementation of the install
>> functionality, maybe there is some interesting for you.
>
> I just read pip code base. I think I can pick something from there for
> downloading the resources. But the problem I have right now is to get only
> the metadata file from the repository. Or are you suggesting that when
> users add a repository connection, it also downloads the repository
> directly?
>


If you are working on the git repos you can probably fetch just one file
without cloning the whole repo, I did not test it but here are probably
some pointers:

http://stackoverflow.com/questions/2466735/how-to-checkout-only-one-file-from-git-repository
http://stackoverflow.com/questions/1125476/retrieve-a-single-file-from-a-repository




>
> On Mon, Jun 6, 2016 at 8:39 AM, Alessandro Pasotti 
> wrote:
>
>> 2016-06-05 10:13 GMT+02:00 Richard Duivenvoorde :
>>
>>> On 05-06-16 09:02, Akbar Gumbira wrote:
>>>
>>> > *Are you blocked on anything?*
>>> > ... In Github or Bitbucket they provide a direct link to
>>> > the raw file. But I think I should look at more general approach
>>> without
>>> > manipulating the URL depending on the host. If you have some input, I
>>> > would be happy to assess it.
>>>
>>> Thanks Akbar,
>>>
>>> I did some googling:
>>>
>>> http://stackoverflow.com/questions/14405782/git-fetch-single-file-from-remote-repository-programatically
>>> If you really want to keep it git, it looks like a shallow clone/copy is
>>> the only way? That post also talks about some undocumented feature, but
>>> I would not depend on that?
>>>
>>> Personally I would be ok when both Github and Gitlab/Gog would work (as
>>> both a closed source and open source member of the git-web-world)...
>>>
>>> Or: a script running somewhere on our server, (shallow) cloning all
>>> registred repositories periodically, and making just the metadata.txt
>>> files available via http/webserver? (maybe giving us some time to check
>>> the repo's on structure and (malicious?) content?
>>>
>>> Or else: a django app for the metadata...
>>>
>>> Regards,
>>>
>>> Richard
>>>
>>>
>>>
>>
>> Hi Akbar,
>>
>> The most flexible installation tool that I know is probably python pip.
>>
>> pip can install software from a zip file, from git and from other sources
>> too.
>>
>> I'd suggest you to have a look to pip implementation of the install
>> functionality, maybe there is some interesting for you.
>>
>>
>>
>> --
>> Alessandro Pasotti
>> w3:   www.itopen.it
>>
>
>
>
> --
>
> *---*
> *Akbar Gumbira *
> *www.akbargumbira.com *
>



-- 
Alessandro Pasotti
w3:   www.itopen.it
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Report 2 - QGIS Symbology Sharing Tools

[Akbar Gumbira]

>
> Hi Akbar,
> The most flexible installation tool that I know is probably python pip.
> pip can install software from a zip file, from git and from other sources
> too.
> I'd suggest you to have a look to pip implementation of the install
> functionality, maybe there is some interesting for you.

I just read pip code base. I think I can pick something from there for
downloading the resources. But the problem I have right now is to get only
the metadata file from the repository. Or are you suggesting that when
users add a repository connection, it also downloads the repository
directly?

On Mon, Jun 6, 2016 at 8:39 AM, Alessandro Pasotti 
wrote:

> 2016-06-05 10:13 GMT+02:00 Richard Duivenvoorde :
>
>> On 05-06-16 09:02, Akbar Gumbira wrote:
>>
>> > *Are you blocked on anything?*
>> > ... In Github or Bitbucket they provide a direct link to
>> > the raw file. But I think I should look at more general approach without
>> > manipulating the URL depending on the host. If you have some input, I
>> > would be happy to assess it.
>>
>> Thanks Akbar,
>>
>> I did some googling:
>>
>> http://stackoverflow.com/questions/14405782/git-fetch-single-file-from-remote-repository-programatically
>> If you really want to keep it git, it looks like a shallow clone/copy is
>> the only way? That post also talks about some undocumented feature, but
>> I would not depend on that?
>>
>> Personally I would be ok when both Github and Gitlab/Gog would work (as
>> both a closed source and open source member of the git-web-world)...
>>
>> Or: a script running somewhere on our server, (shallow) cloning all
>> registred repositories periodically, and making just the metadata.txt
>> files available via http/webserver? (maybe giving us some time to check
>> the repo's on structure and (malicious?) content?
>>
>> Or else: a django app for the metadata...
>>
>> Regards,
>>
>> Richard
>>
>>
>>
>
> Hi Akbar,
>
> The most flexible installation tool that I know is probably python pip.
>
> pip can install software from a zip file, from git and from other sources
> too.
>
> I'd suggest you to have a look to pip implementation of the install
> functionality, maybe there is some interesting for you.
>
>
>
> --
> Alessandro Pasotti
> w3:   www.itopen.it
>



-- 

*---*
*Akbar Gumbira *
*www.akbargumbira.com *
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS Layer Search Plugin

[Nathan Woodrow]

I'm not really sure I see what moving to processing gives us in this case?
Why even make the user make the index files?

If it's integrated into core you can just generate the index on project
open like i do in Roam.  I just spin up a thread and make the index.  Just
have some place the user can config the search options but that can right
in the search bar itself.

Regards,

On Wed, Jun 8, 2016 at 4:35 PM, Denis Rouzaud 
wrote:

>
>
> On 06/08/2016 08:23 AM, Paolo Cavallini wrote:
> > Il 08/06/2016 08:20, Denis Rouzaud ha scritto:
> >
> >> So...is this fair to ask someone to wait that long for a dev and then
> >> ask to modify its code to fit in? I have no answer, it's a real
> question.
> > Maybe cooperating on the same code base will make things much quicker
> > and more robust?
> > All the best, and thanks.
> Indeed asking to move the process to processing would a first step.
> Integrating this in QuickFinder shouldn't be too hard.
>
> I am available to discuss technical details, but I would need support
> for the processing part.
>
> Best wishes,
>
> Denis
>
>
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS Layer Search Plugin

[Denis Rouzaud]

On 06/08/2016 08:23 AM, Paolo Cavallini wrote:
> Il 08/06/2016 08:20, Denis Rouzaud ha scritto:
>
>> So...is this fair to ask someone to wait that long for a dev and then
>> ask to modify its code to fit in? I have no answer, it's a real question.
> Maybe cooperating on the same code base will make things much quicker
> and more robust?
> All the best, and thanks.
Indeed asking to move the process to processing would a first step.
Integrating this in QuickFinder shouldn't be too hard.

I am available to discuss technical details, but I would need support
for the processing part.

Best wishes,

Denis


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS Layer Search Plugin

[Paolo Cavallini]

Il 08/06/2016 08:20, Denis Rouzaud ha scritto:

> So...is this fair to ask someone to wait that long for a dev and then
> ask to modify its code to fit in? I have no answer, it's a real question.

Maybe cooperating on the same code base will make things much quicker
and more robust?
All the best, and thanks.

-- 
Paolo Cavallini - www.faunalia.eu
QGIS & PostGIS courses: http://www.faunalia.eu/training.html
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS Layer Search Plugin

[Denis Rouzaud]

Dear all,

Indeed, the layer search offers something Quickfinder doesn't: searching
on all layers without any configuration, and the use of operators.
Although I think choosing an operator in a global search is not very
required (why doing price < 1000 on all layers, there is a pretty good
chance you know what layer you're looking for).

From this, I think Michael's idea is pretty good.
Make QuickFinder just a search bar in the application toolbar. Then you
can register any search process to it, which would be a processing alg:
might be OSM, FTS or Layer search.

I would definitely be ok to do so for QuickFinder.
But the question is when... It would take me a bit of time to reorganize
the plugin, make FTS generation and search as processing algs (never
done that before). I won't be able to achieve this before August.

So...is this fair to ask someone to wait that long for a dev and then
ask to modify its code to fit in? I have no answer, it's a real question.


Best wishes,
Denis


PS: yes, QuickFinder FTS search handles multiple columns as it supports
expressions (you can concatenate the columns).







On 06/07/2016 03:27 PM, kimaidou wrote:
> Hi all,
>
> I think it is not really profitable to have multiple plugins. In my
> opinion, efforts must be made to merge plugins when possible. In this
> case, we have
>
> * Quick Finder : Full Text Search is used here, which allows fuzzy
> search, and is very efficient compared to scanning all features for a
> match. A drawback, you must fill in the "FTS vectors" for each layer
> you want to be able to search within. I think it is not so bad because
> the process is fast. I also think there is nothing preventing to add
> more than one field in the search ( it is possible with FTS in
> PostgreSQL, must also be the case with sqlite)
>
> * Layer search : Great to have a quick way to search among all the
> fields of all loaded layers, with no configuration. Drawback : I think
> this can be very slow for big datasets, and it is not "fuzzy", as
> regexp are used while iterating on each feature. Another drawback :
> for external vector sources like PostgreSQL, all features must been
> fetched !
>
> IMHO, the way to go could be
>
> * Add the FTS pregenerator of Quick Finder in Processing, as an alg
> available for anyone to use
> * Add the FTS search of Quick Finder as a Processing Alg
> * Add the "layer search" worker as an Alg in Processing
> * Have a combined plugin which let the user choose the search type :
> FTS with pregeneration, or direct search with features iteration. This
> will then run the corresponding alg
>
> Cheers,
> Michaël
>
> 2016-06-07 7:16 GMT+02:00 Jeremy Palmer  >:
>
>
> >
> > I would personally prefer keeping it as a separate plugin, but
> what do you
> > think? Should it be integrated with another plugin?
> >
>
> It seems like it has value over the other plugins. Another issue
> I’ve had is searching all layer fields and pushing the query
> filter down the provider (e.g PostGIS or WFS) to increase the
> performance and usability. Using the approach of iterating through
> all features client side with a large dataset can be a killer in
> terms of bandwidth, memory and time. Does your new plugin do that?
> If so then it would be really useful :)
>
>
>
> This message contains information, which may be in confidence and
> may be subject to legal privilege. If you are not the intended
> recipient, you must not peruse, use, disseminate, distribute or
> copy this message. If you have received this message in error,
> please notify us immediately (Phone 0800 665 463 or
> i...@linz.govt.nz ) and destroy the
> original message. LINZ accepts no responsibility for changes to
> this email, or for any attachments, after its transmission from
> LINZ. Thank You.
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org 
> List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
>
>
>
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer