Jump to content

Search the Community

Showing results for tags 'ftrack-connect'.

  • Search By Tags

    Type tags separated by commas.
  • Search By Author

Content Type


Forums

  • Company news
    • General announcements
  • ftrack Studio (main app)
    • General discussion
    • Feature requests
    • API
    • Help
    • Actions and Widgets
    • ftrack Go
  • ftrack Review
    • General discussion
    • Feature requests
    • Help
  • Connect and plugins
    • General discussion
    • Connect package releases
    • Feature requests
    • Help
    • Connect plugins releases
  • Advisory Forum
  • Tech preview
    • API
    • Connect 2.0 beta
    • New integrations

Categories

There are no results to display.


Find results in...

Find results that contain...


Date Created

  • Start

    End


Last Updated

  • Start

    End


Filter by number of...

Joined

  • Start

    End


Group


Website URL


Location


Interests

Found 6 results

  1. Hi all, I was testing ftrack integration with RV and encountered some difficulties to handle the data between them. Hope that someone knows any workarounds. Before anything, the long story short: Is it possible to query real entityId and entityType from tempdata? To explain what I tested and what are my thoughts, I made an simple graph showing a flow chart of how ftrack-RV integration works. It is important to say that those tests are based on how RV works in Windows 10 environment with an normal RV installation. --- Windows 10 RV 7.5.1 ftrack-connect 1.1.2 --- ,─────, ( ftrack ) '─────' ├───────────────────────────────────────────────────────┐ ↓ ↓ ┌──────────┐ ┌─────┐ │ play in RV button │ │ Actions │ └──────────┘ └─────┘ │ ↓ ↓ ┌───────────┐ ┌─────────┐ │ ftrack Connect -> RV │ │ RVLINK protocol │ └───────────┘ └─────────┘ ↓ │ ,────────────────────────────────────────, ↓ ( "C:\Program Files\Shotgun\RV-7.5.1\bin\rv.exe" rvlink://baked/...(assetVersion) ) defined in ftrack_connect_rv_hook.py ┌──────────────────────────────┐ '────────────────────────────────────────' │ call registry HKEY_CLASSES_ROOT\rvlink\shell\open\command │ └──────────────────────────────┘ │ ↓ ┌──────────────────────────────────────────────────┐ │ "C:\Program Files\Shotgun\RV-7.5.1\bin\rvpush.exe" -tag rvlink url "rvlink://baked/...(tempdata)" │ or whatever is in the registry └──────────────────────────────────────────────────┘ │ / \ < rvpush could find a rv.exe instance with networkTag rvlink already running? \ / │ ├───────────────┐ │ │ │ YES │ NO ↓ │ ,─────────────────, │ ( connect to already running rv.exe )│ '─────────────────' │ / \ < RVPUSH_RV_EXECUTABLE_PATH is defined? \ / │ ├─────┐ │ │ │ YES │ NO │ ↓ │ ,────────────────────────────────────────────────────, │ ( "C:\Program Files\Shotgun\RV-7.5.1\bin\rv.exe" -networkTag rvlink -network rvlink://baked/...(tempdata) ) ↓ '────────────────────────────────────────────────────' ,────────────────, ( call RVPUSH_RV_EXECUTABLE_PATH ) '────────────────' When a user selects an AssetVersion and pushes the "play selection in RV button", a rvlink protocol is called, rvpush.exe runs and rv.exe is launched. In this chain of executions, the rvlink url(passed to rv.exe as an argument) includes a tempdata information, instead of the information from selected entity. This tempdata info is read by RV, passed to FtrackMode.generateUrl, which generates and returns an temporary url. something like below: https://DOMAIN.ftrackapp.com/widget?view=freview_nav_v1&itemId=freview&theme=dark&auth_token=AUTHTOKEN&entityId=ENTITYID&entityType=tempdata This url is accessible from any browser, but I couldn't query entityId nor entityType of the item selected in the ftrack's webUI. On the other hand, when launching RV using Actions and ftrack-connect, real entityId and entityType are passed to RV, allowing me to handle those information easily. However, it has its cons: a.) to use it, I need an running instance of ftrack-connect b.) it always launches a new instance of RV About b.), customizing ftrack_rv_hook.ApplicationLauncher._getApplicationLaunchCommand function seems to solve the problem. Here is a sample code that shows how I implemented a custom launch process. def _getApplicationLaunchCommand(self, application, context=None): command = None if sys.platform in ('win32', ): # command = [application['path']] # ======================================================================================== # # CUSTOM LAUNCH PROCESS environment = self._getApplicationEnvironment(application, context) server = environment.get('FTRACK_SERVER') event = environment.get('FTRACK_CONNECT_EVENT') event_decode = base64.b64decode(event) event_obj = json.loads(event_decode) selection = event_obj.get('selection', '') rvlink_str = "-flags ModeManagerPreload=ftrack ftrackUrl=%s params='%s'" % ( server, selection ) rvlink_encoded = codecs.encode(rvlink_str, 'hex') # do not use normal command that calls rv.exe # instead, call an instance of cmd, start a rvlink protocol and call rvpush command = [ "cmd.exe", "/c", 'start', "", "rvlink://baked/%s" % rvlink_encoded ] # ======================================================================================== # else: self.logger.warning( 'Unable to find launch command for {0} on this platform.' .format(application['identifier']) ) return command With this code, I could call rvlink protocol using ftrack-connect RV launcher, instead of launching new RV instances every time. About a.), the only way I could solve was editing HKEY_CLASSES_ROOT\rvlink\shell\open\command in a way that launches RV with required environment variables. This allows me to launch a RV that mimics a RV instance launched from ftrack-connect, without the need of a running ftrack-connect instance. The problem is that, when calling RV from "play Selection in RV" button, I only have tempdata. As so, solving a.) does not solves my problem at all. I have the same problem as when I launched a ftrack instance without ftrack-connect, indeed. So I return to my first question: Is it possible to query real entityId and entityType from tempdata? I've already checked other posts in the forum: https://forum.ftrack.com/topic/987-ftrack-rv-integration https://forum.ftrack.com/topic/26-installation-guide-for-rv And some helps pages: https://help.ftrack.com/en/articles/1040451-review https://help.ftrack.com/en/articles/1040536-reviewing-in-hiero-player-or-rv Maybe I'm missing something or misunderstanding how integration works. I'll be glad if anyone could correct me or show me any workaround to reach entityId when using ftrack-RV integration without ftrack-connect's Actions. Thanks in advance. Best Regards, Honda
  2. Hi, Does anyone know what happend with the Flame integration plug which was on bitbucket ? Any plans to make a proper flame plugin ? Thanks,
  3. At the moment we can start as many instance of Ftrack Connect as we want. In this case in the Actions window in Ftrack Web UI you will get the same actions multiple times. Is it possible to restrict one instance per URL?
  4. Hi everyone, We have an issue trying to publish assets with maya2017 batch using ftrack-connect-0.7.6 & custom location (it worked with ftrack-connect-0.7.5). We started to dig and we noticed that the plugin is not selecting the custom location for publishing, even if it has the highest priority. Also, it's very strange that it works in maya for the same location. We printed out some information when location is selected by publisher (ftrack_connect/connector/base.py - pickLocation() method): We noticed that candidateLocation.accessor is not symbol.NOT_SET is different for the same location in maya and mayabatch: False in mayabatch and True in maya. Is there anything we need to set up for our custom location ? Also the list of locations is different in maya compared with what we get in mayabatch: Any suggestions for debugging/fixing the issue are welcome! Thank you! Best, Laura
  5. Hey there, we are recently creating an executable file (.exe) for windows from the sources at https://bitbucket.org/ftrack/ftrack-connect.git. Having together all packages needed, this works great using Python 2.7.14, but there seems to be a major drawback. When trying to start Nuke via ftrack-connect-action, we get an error on hashlib saying: Nuke 10.5v4, 64 bit, built May 4 2017. Copyright (c) 2017 The Foundry Visionmongers Ltd. All Rights Reserved. Traceback (most recent call last): File "C:/Program Files/Nuke10.5v4/plugins/init.py", line 22, in <module> import nukescripts.ViewerProcess File "C:/Program Files/Nuke10.5v4/plugins\nukescripts\__init__.py", line 5, in <module> from autobackdrop import * File "C:/Program Files/Nuke10.5v4/plugins\nukescripts\autobackdrop.py", line 3, in <module> import nuke, random File "c:\python2.7.14\Lib\random.py", line 49, in <module> import hashlib as _hashlib File "c:\python2.7.14\Lib\hashlib.py", line 138, in <module> _hashlib.openssl_md_meth_names) AttributeError: 'module' object has no attribute 'openssl_md_meth_names' This can be fixed via editing c:\python2.7.14\Lib\hashlib.py on line 137, 138, by commenting these lines: algorithms_available = algorithms_available.union( _hashlib.openssl_md_meth_names) So we would omit the union of 'opensl_md_meth_names' then... Is here anybody having the same or a similar problem? I'd be greatfull for some insight. Thanks in advance Tim
  6. Hi everyone, I am currently trying to use nuke studio to prep plates and various assets. I though I could use the ftrack plugin to publish all these assets into our ftrack project but I have troubles writing a custom template. it seems the only data against I can run an expression is the shot name on the timeline, is this correct ? Is there any way to get sequence/shot/version from anywhere else (clip name or media file path would be great) ? Thanks in advance, Jerome
×
×
  • Create New...