bulk method for attaching source media file to thousands of imported clips via tab delimated text file

[Bryson Jones]

Check the FAQ on the Squarebox site. It lays out how to do this import with Excel. Take your text into Excel and you should be able to get paths in that way.

We do this all the time but it sometimes takes a few tries to get the format right the first time.

bryson

bryson “at” northshoreautomation.com

northshoreautomation.com

[Rolf Howarth]

It should work if your tab separated text has a column called either Media File or Media Filename.

There are other ways also:

One way is to create a CatDV XML batch file, as that way it will import the media you specify at the same time.

If the clips are named the same as the file and the files are all in the same folder you should be able to bulk attach media by locating the first one and have it do the others automatically (certainly that’s how Update Media Location works, which updates the path of clips that already have media attached, I’m not 100% sure offhand whether Attach Media does the same thing).

Finally, another approach (if each entry in the text file matches precisely one file) is to import all the files in one window and import your tab-separated text file containing the metadata in the other, then copy and paste the metadata from the second window onto the clips in the first. This relies on the two windows having the same clips in the same order.

[Chris Plapp]

I still can’t get this to work via Tab delimited.

We have also attempted to recreate the CatDV XML file format as well, but when I attempt to import the new XML file using File -> Import As -> CatDV XML, we get the following error every time:

“CatDV XML Batch File… – org.xml.sax.SAXParseException: Document root element is missing.”

Here is a link to the file we are attempting to import:

5078_cvodcatdvxmltestcp.xml.zip

[Bryson Jones]

i was able to peel out the first 7 into a smaller file and the import got me 2 clips.

I did notice that a lot of the clips refer to the same asset. We’ve never brought in sub-clips, only full assets on a main import.

Rolf can you check that subclips of the same clip come in via an XML batch?

We’ve always done this with full clips.

bryson

bryson “at” northshoreautomation.com

northshoreautomation.com

[Chris Plapp]

Thanks Bryson.

I know our workflow is a little unorthodox, but we are basically attempting to reverse engineer video clips from metadata.

You see all the sub-clips in my file because we are attempting to create the MP4 sub-clips using CatDV. Each full video has at least 5 sub-clips, sometimes more.

Can we change the format of the XML file I am ingesting to trick CatDV into thinking they are all full videos and still accomplish what I am trying to do?

[Rolf Howarth]

Your file contains a byte order mark. You need to remove that for the XML parser to read the file. (Do Save As and select a format that doesn’t include a BOM).

[Chris Plapp]

We removed the BOM and the file will now import, but it ONLY imported the 1st clip for every video.

It also changed all of the IN/OUT points and ignored what is included in the attached XML file.

For example, 10381_01 imports with the duration being the length of the source media file, 16 minutes. It ignores the 16 second start time and 3:18 end time.

[Bryson Jones]

Rolf, do you actually allow for INs and OUTs to be imported? I’ve never done that, only brought in bulk clips as we’re normally working from another MAM or a bulk list from some other database.

Chris are these clips from an NLE? Where were they created?

bryson

bryson “at” northshoreautomation.com

northshoreautomation.com

[Chris Plapp]

We can correctly see the IN and OUT points in CatDV if I import a simple tab delimited text file.

I don’t understand why the CatDV XML format does not work in the same manner.

The problem with this method for us is that we would have to manually attach the correct source media file for every batch of clips we want to create.

We are not creating the clips in CatDV, we are importing predetermined in and out points from a database of ours, that have been represented on the web as embedded cue points. We are trying to take those virtual segments and turn them into physical video clips.

[Rolf Howarth]

I think you’re correct and the XML import currently imports only one clip per file. That feature is primarily designed to import assets into CatDV together with metadata (for example, when you archive media files you can choose to save a copy of the CatDV metadata as an XML file with the media, then if you lose the CatDV catalog but reimport the media from the archive you will get the metadata back as well). It isn’t currently very well suited to importing subclips but that seems like an unnecessary limitation, so we’ll see if we can improve that in a future version.