DBCFeeder Authorization Support #74
Conversation
d8aafcd to
8c9d418
Compare
|
@lukasmittag @SebastianSchildt - for information - this adapts dbcfeeder to authorization. The refactoring here could possibly also be reused by other python-based feeders like ddsfeeder, i.e. that they just re-use the wrappers. Not ready for final review yet |
4fc26bd to
2c64f90
Compare
9891c97 to
aeb7eb5
Compare
Also major refactoring to move client-specific code out of main file.
aeb7eb5 to
81ec953
Compare
| Old mapping format no longer supported. |
There was a problem hiding this comment.
what is here different?
There was a problem hiding this comment.
Github is not that good at showing subtle differences. If you use for example gitk you will see that a newline has been added.
Feeder refactored and new mapping format based on VSS introduced, see [documentation](mapping.md).
-Old mapping format no longer supported.
\ No newline at end of file
+Old mapping format no longer supported.
|
|
||
| # TODO add data for root cert if using TLS and if given | ||
|
|
||
| self._kuksa = KuksaClientThread(self._client_config) |
There was a problem hiding this comment.
do we want to us threaded API here?
There was a problem hiding this comment.
This file is for KUKSA.val server (i.e. websocket) and has not been changed in this PR - the server-specific code has just been put in separate file to make dbcfeeder.py unaware of what APIs that actually are used. What other options do we have?
There was a problem hiding this comment.
Ah okay, I missed that. fought it was in the databrokerclientwrapper. Then its fine
| @@ -0,0 +1,16 @@ | |||
| name: pre-commit | |||
There was a problem hiding this comment.
@SebastianSchildt is this okay from an open source legal perspective? Tp run automatically run a code formatter before every commit? We had the dicussion regarding auto generation of licenses too.
There was a problem hiding this comment.
It actually does not fix anything when run as part of CI. See the dummy PR in erikbosch#2 where I have added a blank in a README. The check will fail, it will not fix it.
This can be compared to if you run it locally. There it will fix the file, but it will not do a git add and it will not do a git commit --amend. So also when running locally you must "accept" the changes (by git add) and then try git commit again.
erik@debian3:~/kuksa.val.feeders$ pre-commit run --file README.md
Trim Trailing Whitespace.................................................Failed
- hook id: trailing-whitespace
- exit code: 1
- files were modified by this hook
Fixing README.md
Fix End of Files.........................................................Passed
Check Yaml...........................................(no files to check)Skipped
Check for added large files..............................................Passed
flake8...............................................(no files to check)Skipped
There was a problem hiding this comment.
Ok, it read like it would fix it. But then also fine.
This PR does:
It uses the kuksa-client pre-release https://pypi.org/project/kuksa-client/0.4.0a1/. Later as part of release process we should make sure that it uses an official version instead.