Customising tag expressions

It is possible to customise the tag expressions used when matching the clip names to work with your own naming convention. This can be done in two different ways, either per session or persisted in Connect.

Modify for a single session

If you temporarily want to change the expression used when matching the clip names you can do it from within Nuke Studio.

Open the tag window, Window ‣ Tags and navigate to the ftrack folder and double click on the tag you want to modify to bring up the edit window.

Note

Changes made in the tag window will be restored once Nuke Studio is restarted.

../_images/edit_tag.png

In the edit window you need to edit the re attribute in the data list.

The regular expression must define a named group called value which will be used as the value when the expression matches. The expression is used in a search, so use anchors if appropriate for exact matches.

Note

You will need to clear and re-add the tags to the clips in the timeline after the tags are modified.

Persist changes in hook

To persist changes to the tag expressions between Nuke Studio sessions you will need to modify the default hook that loads the tags.

The file to modify is called context_tags_hook.py located in different locations depending on how you are running the Nuke Studio plugin.

If you are using a built ftrack connect package application, the file can be found in the following locations:

Platform Path
OS X /Applications/ftrack-connect.app/Contents/MacOS/resource/ftrack_connect_nuke_studio/application_hook
Windows C:/Program Files/ftrack-connect package/resource/ftrack_connect_nuke_studio/application_hook
Centos <Installation directory>/ftrack-connect-package/resource/ftrack_connect_nuke_studio/application_hook

If running from source the hooks can be found in resource/application_hook/ in the plugin project folder.

Once you’ve found the file, open it in your favorite text editor. The file should look something like this:

# :coding: utf-8
# :copyright: Copyright (c) 2015 ftrack

import logging

import ftrack


class ContextTags(object):
    '''Return context tags for Nuke Studio.'''

    def __init__(self, *args, **kwargs):
        '''Initialise context tags hook.'''
        self.logger = logging.getLogger(
            __name__ + '.' + self.__class__.__name__
        )

        super(ContextTags, self).__init__(*args, **kwargs)

    def launch(self, event):
        '''Return context tags.

        Should be list with tags using the format:

            ('tag_id', 'ftrack_type_id', 'regexp')

        '''

        self.logger.debug('Loading context tags from hook.')


        return [
            ('project', 'show', None),
            ('episode', 'episode', 'EP(?P<value>\d+)|(?P<value>.+)'),
            ('sequence', 'sequence', 'SQ(?P<value>\d+)|(?P<value>.+)'),
            ('shot', 'shot', 'SH(?P<value>\d+)|(?P<value>.+)')
        ]

    def register(self):
        '''Register hook.'''
        ftrack.EVENT_HUB.subscribe(
            'topic=ftrack.connect.nuke-studio.get-context-tags',
            self.launch
        )


def register(registry, **kw):
    '''Register hooks for context tags.'''

    # Validate that registry is instance of ftrack.Registry, if not
    # return early since the register method probably is called
    # from the new API.
    if not isinstance(registry, ftrack.Registry):
        return

    plugin = ContextTags()
    plugin.register()

The part you need to focus on is the one returning the actual tags:

return [
    ('project', 'show', None),
    ('episode', 'episode', r'EP(?P<value>\d+)|(?P<value>.+)'),
    ('sequence', 'sequence', r'SQ(?P<value>\d+)|(?P<value>.+)'),
    ('shot', 'shot', r'SH(?P<value>\d+)|(?P<value>.+)')
]

Edit the value for each context type to desired expression, below is a modified example with the example expression:

return [
    ('project', 'show', None),
    ('episode', 'episode', r'EP(?P<value>\d+)|(?P<value>.+)'),
    ('sequence', 'sequence', r'(?P<value>.+)_'),
    ('shot', 'shot', r'_(?P<value>.+)')
]

Example expressions

Sequence and shot separated with underscore

If your shot names contain both the name of the shot and the sequence separated by an underscore you can use an expression for shot and sequence looking like this:

Context Expression
Sequence (?P<value>.+)_
Shot _(?P<value>.+)

Given three shots named 001_A010, 001_B010 and 002_010 would generate a hierarchy like:

../_images/example_expression.png