Azure allows the use of Python, Django, Postgres database and even Postgres add-ons such as PostGIS for working with spatial data.
Any webapp should be able to work with spatial data and it's not uncommon. However, libraries typically require the use of external dependencies e.g. GDAL. Please see this page for reference as to what libraries are required:
https://docs.djangoproject.com/en/2.2/ref/contrib/gis/install/geolibs/
It seems like quite an oversite to not think about working with GIS data.
If running a Linux VM, this is trivial as we can run the shell command:
sudo apt-get install binutils libproj-dev gdal-bin,
as indicatedd in the docs referenced above.
However, when using an Appservice, it is unclear as to how we achieve this.
Any guidance on how we run the shell command mentioned above? Where would the associated packages be installed?
@oi-eddyoj Thank you for the valuable feedback,we are investigating the issue.
@Adam-Smith-MSFT I have had to go down the route of creating containers, which has taken me 3 days to learn so far. Docker is a monumental waste of time for small business functions to develop applications with, but it makes sense for much bigger organizations.
The functionality of being able to run custom commands in a linux appservice is so critical in keeping app development cheap. It is very tempting to leave Azure just for this issue.
@oi-eddyoj We apologize for the frustration that you have encountered. Our blessed images are locked down a bit more than a IaaS VM image. As a result, there's a bit of a different workaround. Please see the below steps to achieve this.
# Installing spatial dependencies
apt-get update -qq && apt-get install binutils libproj-dev gdal-bin -yqq
We will now proceed to close this thread. If there are further questions regarding this matter, please tag me in your reply. We will gladly continue the discussion and we will reopen the issue.
Most helpful comment
@oi-eddyoj We apologize for the frustration that you have encountered. Our blessed images are locked down a bit more than a IaaS VM image. As a result, there's a bit of a different workaround. Please see the below steps to achieve this.
We will now proceed to close this thread. If there are further questions regarding this matter, please tag me in your reply. We will gladly continue the discussion and we will reopen the issue.