instinct-vfx

Members
  • Content Count

    57
  • Joined

  • Last visited

  • Days Won

    7

instinct-vfx last won the day on October 3 2017

instinct-vfx had the most liked content!

About instinct-vfx

  • Rank
    Advanced Member

Profile Information

  • Gender
    Not Telling

Recent Profile Visitors

684 profile views
  1. If this is an on-prem installation you need to trust your internal SSL certs (e.g. by using CA_BUNDLE environment variable). If this is hosted by ftrack you should contact support about it.
  2. Thanks for the heads up (and sorry for the delay in replying, email notifications seem to be off, have to check). This is VERY welcome!
  3. Does this change lead to "way more" being supported but still limited? Or will this completely remove the limit?
  4. This is a front-end limitation, you can assign and unassign more through the API. We worked around it by managing tasks that need more than the supported user count in the UI in a separate local tool. It can seemingly be worked around also by changing a MariaDB setting. You may want to talk to support directly if you want to go down that route (we felt that managing that many users from the current legacy resource selector is not exactly fun anyways).
  5. Agreeing with this. Another thing broken in the new docs is that line breaks in code sections seem to be broken. e.g. https://help.ftrack.com/administering-ftrack/on-prem/configuring-the-database Generally i found the old format to be more accessible to browse. And having the section headers (like "Managing a local installation" on the same level as its contents seems pretty confusing) Cheers, Thorsten
  6. You are trying to create an assignment where i think you should be creating an allocation, no?
  7. Heya, i was wondering if there are any plans already to make the python api both 2.x and 3.x compatible? We are using the api in some of our server applications and would like to move towards python 3. Cheers, Thorsten
  8. @Mattias LagergrenThanks! @Martin Pengelly-Phillips agreed, that currently is not really an issue for me as i redeploy from scratch quite often on the testing instances. I did an ansible playbook automating the install, then inject our db snapshot.
  9. That's what i did here. Had a testing server with a database snapshot of our production server that was running 3.4.0.rc. I upgraded it to 3.3.38 and after killing my browser cache the new version was there.
  10. Is this with vanilla 3.3.38 and included or do we need to update to 3.3.38 and still pull the studio overview remotely and is the url still the same if this is the case?
  11. Hey there, not sure without actual queries/code, but i think this is the same limitation that i have hit. The reason is, that context is a dynamic (or polymorphic) entity that may be different entity types. Hence it does not contain all the attributes that an actual instance may or may not have. If this is what you are seeing then this was addressed in 3.3.35: NEW APIDeveloperAdded support for complex queries with projections and criteria containg attributes from concrete subclasses. See also: http://ftrack.rtd.ftrack.com/en/latest/developing/api/query_syntax.html#developing-api-query-syntax-queries-involving-relations-to-subclasses Cheers, Thorsten
  12. I'd like to add my vote here. The db is ever growing. We have not even really started publishing assets and versions in our main business lines and the db is getting pretty clunky already.
  13. Hey there, there was talk about docs. For now you can use the attributes and .keys properties/methods. Here is a snippet that prints a map of the available types and properties: import ftrack_api session = ftrack_api.Session( server_url='FTRACK_URL', api_key='API_KEY', api_user='USER' ) ftypes = session.types for ftype in sorted(ftypes): print ftype for attribute in sorted(ftypes[ftype].attributes.keys()): print "\t" + attribute Cheers, Thorsten