Gilles Dartiguelongue

looks sane. :+1:

Sounds like a good idea. Link to line in question: https://github.com/mardiros/pyshop/blob/master/pyshop/models.py#L301

Came for the same issue. I wanted to remove use of extra_javascript and extra_css but since I'm building our documentation in an isolated network, the build fails checking the URL...

Since our documentation has a CI/CD pipeline, it is easy ok for us to make it fail in case there is an error in the documentation (missing documents, broken links,...

Actually, if you have an API that returns `Content-Type: text/plain` but it actually returns UTF-8, it will return a badly (latin1) encoded string. Using the proper `Content-Type: text/plain; charset=UTF-8` instead...

This might not be exactly what you are expecting nor the best solution but for a similar case, in our Harbor API integration, we use the following code to deal...

For reference, here is the excerpt from Harbor spec: `$ jq '.paths["/projects/{project_name_or_id}"].delete' ~/harbor.json` ```json { "responses": { "200": { "$ref": "#/responses/200" }, "404": { "$ref": "#/responses/404" }, "403": { "$ref":...

It would be nice to allow building pex files without C extensions as some of my target environments don't have libpythonX.Y.so installed (weird I know).

To give a bit of context, I'm currently trying to match bank transaction statements with accounts to generate OFX file in woob project. The bank is generating 2 different types...