Xml encoding in xml_tools.py

[alceballosa]

Hi,

I was recently testing MAJA in a Docker image (in Windows, via WSL2) prepared as per the tutorial here (using the installer): #80

My miniconda Python environment consists of the following libraries: https://pastebin.com/LkY2XLDX

Most steps went smoothly, but during the xml_tools.py "can be opened?" validation, the lxml library could not open the XML files in the .SAFE folders so MAJA couldn't detect any assets, so I kept getting the "No input L1 product available" exception.

I fixed this by adding encoding='utf-8' to all instances of the open function in xml_tools.py; however, I am not sure if this would introduce bugs on non-WSL+Docker environments (maybe Windows is introducing some unwanted line endings?)

Should I make a pull request nonetheless?

Kind regards.

[petket-5]

Hi,

Thank you for the analysis as well as the proposed solution. As of today, Maja under WSL is not officially supported but we added this to the internal todo-list. Anyhow, we would be happy to test the utf-8 fix using our CI attached to the gitlab. Could you open a MR for this under the following link please: https://gitlab.orfeo-toolbox.org/maja/maja

@jbrossar @Julien-Osman : Given the recent talks we had about the topic of contributions, are third party MRs like this eligible for CI testing?

Kind regards,
Peter

[alceballosa]

Hi Peter!

Sorry for being late with this, the situation in my country (Colombia) has been delicate the last few weeks and I wasn't able to do the MR. In any case, just went ahead and did that so let me know if you are able to test the utf-8 fix.

Kindly,

Alberto.