Use uuid4 instead of uuid1 to generate trace uid #1236
Conversation
|
Thanks for the PR @erichiggins! Our CI tests aren't running right now due to some issues with GitHub, but once they come back up I'll pull this in. |
|
It's not clear from the tracebacks in the CI tests failed due to my change or something else. It doesn't appear to be related at first glance. Looks like other PRs are failing on them as well. |
* Add layout.template codegen and validation logic * Updated codegen to add support for elementdefaults properties. e.g. `layout.template.layout.annotationdefaults` * Added template acceptance/validation tests * Implementation of plotly.io.templates configuration object to supports registering/unregistering templates and setting default template * Added plotly.io.template tests * Added plotly.io.to_templated function. This inputs a figure and outputs a new figure where all eligible properties have been moved into the new figure's template definition * Added plotly.io.templates.merge_templates utility function * Support specifying flaglist of named templates to be merged together. e.g. fig.layout.template = 'template1+template2' * Added 5 built-in themes: 'ggplot2', 'seaborn', 'plotly', 'plotly_white', and 'plotly_dark' * Added 'presentation' template that can be used to increase the size of text and lines/markers for several trace types, and 'xgridoff' template to remove x-grid lines * Update orca tests to only compare EPS images. Something changed in CircleCI mid-development that broke the reproducibility of other image formats.
|
Hi @erichiggins , I think the tests are fixed on master now (but we're still experiencing some issues since the GitHub problems yesterday). Please merge master and push when you have a chance. |
|
@jonmmease Done! Looks like that fixed it. Syncing and merging forks is a bit outside my usual workflow, so the merge/rebase could probably have been cleaner. Sorry if it introduces any noise into the commit history. |
|
Thanks @erichiggins , I'm not quite sure why this PR is showing "368 Files Changed" with respect to master. I think GitHub may be a bit behind. In any case I'll take it from here 🙂 |
This PR addresses #1235 .
I didn't see any existing tests which specifically test this functionality. Since this issue and fix has more to do with core Python, I'm not entirely convinced that they would be that useful.