KimPatch
commented
5 years ago @ethanbutler I'm documenting the logic we worked out to do this in a simple, practical way, and I've gone a bit further: (this is low priority, but would increase efficiency and ease of use for content creators)
- If the supporting content has timecodes, ignore and preserve them.
- Start at first line with text on the first instance, and after the first instance the first line following a blank line (ignoring timecode lines if they are present).
- Distinguish between TITLE and DESCRIPTION: assume that the first bit of text following a blank line is a title, and the second line is a description unless it is a URL.
- Distinguish between external and internal URLs: if it doesn't have a number at the end it is an external URL. If it does have a number at the end the user is alerted and chooses whether it's an internal or external URL. This is to prevent false positives because external URLs can have numbers.
- Distinguish between URLs and pictures or files – pictures and files are always going to have wp-content as part of the label.
Note that this does not account for every type of supporting content, but accounts for by far the most common types, and is meant to be checked.
Give the content provider a button to add labels to content in the supporting content .vtt tab. The logic would be if TITLE doesn’t appear as the first word of the first line below the timecode, add TITLE, if DESCRIPTION doesn’t appear as the first word of the next paragraph, add DESCRIPTION, if URL doesn’t appear as the first word of the next paragraph, add URL, then find the next timecode etc. (What we are doing now is using a Word macro to do this. What I’m looking to do is cut out the extra step of putting this in Word and running the macro by adding that ability into the editor.)