David Randall Stokes
David Randall Stokes
For consistency, the keyword arguments for `Schema.load()` and `Schema.loads()` should match. Specifically, `Schema.loads()` should have the `headers` keyword argument.
When exporting to XML, there are various cases in which it would be useful to specify how individual element values get converted to text, similar to the BinaryElement encoding added...
With the development of separate libraries (`idelib`, `endaq.device`), the schemata specific to Mide and/or enDAQ should be moved into them. This will establish which package's version is canon, and prevents...
At some point in Q1 2022, an update of [`flake8`](https://github.com/PyCQA/flake8) made some tests fail. It is likely indirect, caused by something in [`pytest-flake8`](https://github.com/tholo/pytest-flake8). A workaround was created by pinning `flake8
In `SCHEMA_PATH`, package-relative paths containing only the package name don't work; even though `idelib.schemata` is a valid, importable submodule, `"{idelib.schemata}"` doesn't work. It will work if followed by a slash...
I don't think it is worth supporting the old `python-ebml` schemata anymore; the project hasn't had any significant update since 2011. The schemata we use in our projects have long...
The little bit of help in the README isn't great. We should get the package marked up and rendering with Sphinx, and host it on readthedocs.io, either directly on the...
In schemas, we currently use default values in some of the `EBML` child elements to define version info (`DocType`, `DocVersion`, `ReadVersion`, etc.). This is not the best place for it;...
Some schemata have a header element, ``. The intention was to have it contain metadata, but was never fully developed; it's arguably useful to humans, but is currently ignored by...
Related to #12 and #39. In various places, the library raises exception types that don't seem entirely appropriate; for example, decoding an ID with a bad length raises an `IOError`,...