Use upstream adslib directly from Beckhoff#508
Conversation
Make the AdvancedHandler less magical
- convert data values from dict into bytes, and write bytes accordingly in adsSumWrite() - unit test still fails since data structure is lost during writing
Needs to be filled with life
The use-case where no structure_def is passed is covered by the ads.write_list_by_name and read_list_by_name functions.
…st_by_name Structure support for write list by name
Auto-upload package to PyPi
First test of Github Actions
|
It would be easier for us to rely on the upstream directly. The downside is it will require more dependencies on the client system, in case we don't have a pre-compiled wheel for them. @stlehmann do you think this is acceptable? I myself have no real idea how complete our wheel coverage is right now. EDIT: Also relevant for editable installs from source. |
|
@RobertoRoos I'm in contact with Patrick Bruenn from Beckhoff discussing how to make adslib directly compilable to a shared library. So far it looks promising. |
Coverage Report for CI Build 25001451298Coverage decreased (-0.02%) to 93.389%Details
Uncovered ChangesNo uncovered changes found. Coverage RegressionsNo coverage regressions found. Coverage Stats
💛 - Coveralls |
|
@stlehmann what just happened in this repo? All PRs got closed, was that you? Those PRs (and this one too) suddenly show 900+ commits diff. Which is about the entire size of this repo. |
|
@RobertoRoos sadly I must admit that this was me, indeed. Yesterday I accidentally force-pushed another repository to pyads. I tackled by force-pushing the current version of pyads again. Unfortunately all PRs got closed. I haven't found a way to deal with this, yet. Any help is welcome. I'm sorry for the mess 😒. |
|
Oof 😅 I fetched locally without pulling and it doesn't show any differences with the upstream, so all commit references, tags, etc. should still work. We should re-open the closed PRs, but maybe put them in 'Draft' because I think they would require a touchup before they could be merged. |
|
Tell me ... 😮💨 Yes, I would like to re-open the PRs again. Don't know really how to do this because they refuse to enable the "REOPEN" button: 
If someone has a good idea ... |
|
Oh dear, I thought something was up. I feel like we must be able to recover from this somehow, did it rewrite history? If not we may be able to rewrite it back. |
|
@chrisbeardy , the history is still identical, I checked that yesterday. See my comment above. I was experimenting with trying to re-open #419 . I just rebased and force-pushed that branch, but the existing PR still shows this This seems to be a known situation, see for example: https://gist.github.com/robertpainsi/2c42c15f1ce6dab03a0675348edd4e2c EDIT: @stlehmann are you allowed to re-open #419 now? I reset my branch back to the last commit of |
|
Conversation continues in #511 |
20 participants
Resolves #507