Lorenzo Angeli

Hi @Alex Sknarin, Starting with ftrack 4.1, new accounts does not receive a API key by default and API keys can only be viewed when they are created. To generate a personal API key, you can press “Reset API Key” from My account. There is a slight difference though in the presentation in the UI: No API key set looks like: ******************** An API key set looks like: MXN5Y******************** Where the first characters are a readable identifier for the key. hope it helps. L.

Hi @Aaron Powell, depending on where you file lives (location type), you can use either : location = session.pick_location() location.get_filesystem_path(component) or location = session.pick_location() location.get_url(component) Hope it helps. L.

Hi @daans, so glad to hear you sorted it out! if you happen to have any other issue, please let us know! Cheers. L.

Hi @daans, Although I wouldn't suggest to install modules system wise (either installing on the user or in virtualenvs), ftrack api can totally be imported and put to the use from within Nuke and without the connect support. Looking at the error you are having : ImportError: No module named backports.functools_lru_cache seems some dependencies are not properly installed, or not picked up from Nuke itself. But I'd say the latter if you can make them work from pure python shell. As test I've been creating (under Linux Centos 7 , but should not make any difference for Ubuntu) a virtual environment as: python -m virtualenv api-test -p python2.7 (note, python2.7 is an alternative python built of 2.7.18 as C7 ships with 2.7.5 which is too old for the api to work) and then activating the virtualenv and installing the api with : python -m pip install ftrack-python-api started nuke (12.1v2) with and then from the Script editor: import sys sys.path.insert(0, '/home/ftrackwork/.venv/api-test/lib/python2.7/site-packages') import ftrack_api s = ftrack_api.Session() # note I've the my credentials in the environment variables print s.query('Project').all() > # Result: [<dynamic ftrack Project object 140663537353232>, <dynamic ftrack Project object 140663537353488>, <dynamic ftrack Project object 140663537353552>, <dynamic ftrack Project object 140663537353040>, <dynamic ftrack Project object 140663537353104>, <dynamic ftrack Project object 140663537353168>, <dynamic ftrack Project object 140663537353424>, <dynamic ftrack Project object 140663537353808>, <dynamic ftrack Project object 140663537353680>] hope it helps. L.

Thanks @Konstantin Maslyuk we'll go through them asap. Cheers. L.

hi @Konstantin Maslyuk thanks for reporting. We'll be having a closer look as soon as possible (or feel free as usual to open a PR in case )! L.

Hi @Aaron Powell have you tried using the 'launchArguments' rather than data.command ? you can see how it gets used in nuke (in this case from the discovery itself) https://bitbucket.org/ftrack/ftrack-connect-nuke/src/77c27edbab69316fc09a3b37f77852c9b8adadd7/resource/hook/ftrack_connect_nuke_hook.py#lines-276 and how gets interally used: https://bitbucket.org/ftrack/ftrack-connect/src/e9c26b4007b03825f1fcf5b0346a4260c38ad45b/source/ftrack_connect/application.py#lines-416 hope it helps. L.

Hi @intelyuxiang as first thing I would check if your machine provides the prerequisites: http://ftrack-connect-unity-engine.rtd.ftrack.com/en/latest/install.html When Unity start up correctly, you should be having a python interpreter opening in a shell like: and a python process running: Hope it helps. L.

Hi @Andrea Perisano, there's no way to customise the dropdown menu I'm sorry. The way to go is to provide an action which launches your review based on either lists, or components. Hope it helps. L.

Hi @chengshun neither does adding the notes in the projection ? entity_ins = self.session.query('select notes from AssetVersion where id is %s' % sel_id).first()

@chengshun we have been further looking into the reported issue. One thing you can try is to replace the session.cache.clear() call with session.populate(project, 'notes') Please let us know if does solve your problem. Cheers. L.

Hi @chengshun resetting the cache half way through the action is risky at least. Something you can try is to rework the query you are doing as such: entity_ins = self.session.query('select notes from AssetVersion where id is %s' % sel_id).first() here you can find one of our examples which shows you how to use the use of ftrack-action-handler and how to attach the result document to the job itself ,so you don't have to save it , hope it helps. if you are still having issues please let us know. L.

Hi @intelyuxiang, would you mind trying against the latest master on https://bitbucket.org/ftrack/ftrack-connect-unity-engine/src/master/ cone the repository with: git clone https://bitbucket.org/ftrack/ftrack-connect-unity-engine.git then from within the cloned folder run : python setup.py build_plugin this will create an up to date version of the integration. Please also ensure you Unity Hub has been stopped before re running the integration. Let us know how it goes. L. Ps. Also can you report back what Unity version are you using ? thanks.

Hi @intelyuxiang at the moment the only asset type allowed to be published are the image_sequences. We have now uploaded the technical documentation , please have a look a the trubleshooting page. Please rest assured we are actively looking on a potential fix. L.

Hi @Konstantin Maslyuk just saw the PR , thank for it. We'll review whenever possible . Cheers! L.

hi @Konstantin Maslyuk please feel free to fork the api and provide a PR against our repo ! more than interested to look into it. Thanks!

Hi @Spencer, have you been checking on our API help ? Hope it helps. L.

Hi @daansQt5 is currently not supported . We are working on that but there's no ETA atm I can provide I'm sorry. Unless you can find a way to install Qt4 on pop!_os I fear you are on a dead end for now. L.

hi @daans, my suggestion would be to try to install in a virtualenv ftrack-connect from sources. the process would be something along these lines: 1) ensure you have libssl and qt are installed and available on the system 2) create a virtualenv for python 2.7 (python > 2.7.9) 3) clone ftrack-connect repository https://bitbucket.org/ftrack/ftrack-connect 4) activate virtualenv 5) run : python setup.py install if everything went well, you should now be able to run : ftrack-connect from the shell. Let us know how it goes! L. P.s Differently from ftrack-connect-package this is just ftrack-connect and does not include the integrations. the easier way could be to copy the integration from the package or clone from bitbucket and build them with : python setup.py build_plugin (ensure you are in the virtualenv though)

@Andrea Perisano I've been checking an on the upcoming release we are already set to use Python for unity api 2.0 . Stay tuned for more info. L.

Hi Andrea, the unity integration is not officially out yet and we are still working on it. I'll be raising this api version issue internally and let you know. Thanks. L.

Hi @Konstantin Maslyuk seems most of the time is spent on the json encoder, interesting. Can you provide what python version are you using and some code to replicate the profiling ? Thanks. L.

Hi @John , if you want to remove the component and free up the space, you can use the <location>.remove_component(<component>) hope it helps. L.

RC6 is now available on pypi and can be installed with : pip install ftrack-python-api==2.0.0rc6 What this version fix or improve: Event server wait method will sometimes raise a connection error because of a race condition. Improve testing

Hi all, a new RC version (4) is out for testing.... this version introduce the fix for the issue @jen_at_floyd has been reporting above about SSL connection. You can install the new version as usual with : pip install ftrack-python-api==2.0.0rc4 Another step closer to the final release, please test and report if there's any issue! Thanks. L.