[Qgis-user] announcing time series plot plugin (experimental)

[josef k]

Hi Anita,

I definitely see your point that implementing other data formats may
attract other interested users and possible developers.

At the moment, the code simply sends a few simple sql-statements to the
sqlite table with time series data so for that reason there should not be
too much of a work to change data format. I guess those sql statements
could be addressed to a post-GIS database as well. The major holdback for
such a change is probably my own lack of time and lack of coding skills.

But I may try to learn som post-GIS basics some day, just that much I can
let the plugin read that data format.

It is just too bad that adding a csv table to qgis results in string values
only (or have I misunderstood?). Otherwise it might be the quick-fix to let
the plugin be useful with more widely spread data formats.

Thank you for your valuble feedback, I really appreciate it!

Best regards
Josef

2011/11/28 Anita Graser <anitagraser at gmx.at>

> Hi Josef,
>
> On Thu, Nov 24, 2011 at 8:34 AM, josef k <groundwatergis at gmail.com> wrote:
>
>> Hi Anita,
>> I am glad you find some use for it!
>> I did choose SpatiaLite since I deal with quite a lot of data from
>> several sites at the same time.
>>
>
> Are you using any of SpatiaLite's functions? Would it be difficult to
> adjust the code so it could handle arbitrary vector data sources?
>

>
>> It turned out that spaialite is a great data container for me and my
>> colleagues since we only have to deal with one single (and portable) db
>> file for each site. Furthermore, since we are mainly hydrogeologists and
>> for sure no db gurus, spatialite seemed lika a much easier way to start
>> organizing things than e.g. post-GIS.
>>
>
> There are some reasons why I don't use Spatialite much, e.g. the
> combination of QGIS + Spatialite has some issues if you define database
> views and try to import them into QGIS. The data types are screwed up
> often, which in turn makes it impossible to visualize the features by their
> (numerical) attributes.
>
> Once PostGIS is set up, I've found it to be much more reliable and
> predictable than Spatialite. e.g. Spatialite's weak typing lead to
> unexpected results.
>
> Therefore, I'd suggest implementing support for other vector formats
> available in QGIS before building new features. A broader setup will
> attract more interested users and possible developers.
>
> Best wishes, keep up the good work!
>
> Anita
>
>
>
>> I also hope that I (or anyone else that finds it interesting) will get
>> time to extend the plugin to include more ways of visualising other
>> hydrogeo data from the same spatialite db. Additional functions would be
>> stratigraphy plots (similar to ARPAT), time change visualisation  (as your
>> Time Manager), gesection profiling, some R statistics etc. (Could be a
>> lightweight alternative to the core functionality of ArcHydro GW data model
>> and tools.)
>>
>> regards
>> josef
>>
>> 2011/11/23 Anita Graser <anitagraser at gmx.at>
>>
>>> Looks great! I'd have quite a lot of use cases for such a plugin.
>>> Any specific reason why it's limited to Spatialite as a data source?
>>>
>>> Best wishes,
>>>
>>>  Anita
>>>
>>>
>>> On Wed, Nov 23, 2011 at 10:18 PM, josef k <groundwatergis at gmail.com>wrote:
>>>
>>>> Fellow QGIS-Users!
>>>>
>>>> I just finished my first QGIS python plugin in an experimental version.
>>>>
>>>> The plugin uses pyspatialite and matplotlib to load data from a sqlite
>>>> database and plot it by matplotlib plot command. The intended use is for
>>>> e.g. hydrogeologists (i.e. me) to quickly select features that corresponds
>>>> to observation wells, stream gauges or other measuring stations with data
>>>> to be plotted and then quickly (yes I am in a  hurry) have a window with
>>>> time series plot for zooming, panning and visually examine the data (I know
>>>> it is not very scientific but I do perform some serious data analysis at a
>>>> later stage...).
>>>>
>>>> Anyway, the plugin is named TimeSeriesPlot, marked as experimental, and
>>>> found in QGIS User-Contributed Python Plugin Repository.
>>>> Screenshot:
>>>> Instructions are found in the readme.txt file shipped with the plugin.
>>>> Download some sample sqlite data:
>>>> And a (silent) video:
>>>>
>>>> Any feedback is appreciated!
>>>> /joska
>>>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20111128/7f56f444/attachment.html>