Worker Node Expressions and import issue

[Bryson Jones]

I’m always a fan of separate jobs, just so you can troubleshoot easier. And yes, I believe you’re getting one “analysis” pass.

The Reg Ex stuff is the hardest part of the Worker. Can I ask why you aren’t using the $1, $2 and $3 variables?

bryson

bryson “at” hidefcowboy.com

hidefcowboy.com

[David White]

Hi Bryson

Thanks for your reply, I’ve now configured two separate jobs, one for transcode and the second to add to the database.

I thought I was using the $1, $2, $3 variables as shown above….
My understanding of that regex is extract the text between the start and the 1st _ into $1 and then that will place it in the jobid field.
Maybe I’m missing something in your reply?

Regards

David

[Bryson Jones]

$1 is enough, there’s no need for a regex if your file is named as you show.

$1 = 0001

$2 = snowman

$3 = peter

The Worker just does that as if my magic. 😉

bryson

bryson “at” hidefcowboy.com

hidefcowboy.com

[David White]

Hi Bryson

It’s not the file that’s named in that format, but the folder, sorry if that wasn’t clear from the original post.

ie
/ingest/0001_snowman_peter/file1.mov
/ingest/0001_snowman_peter/file2.mov

hence the need to do a regex on $q, the parent folder name.

[David White]

Hi, I’ve found a regex that works, and is giving me the the names including spaces:

$q{s,^.*([a-zA-Z0-9s]+)_([a-zA-Z0-9s]+)_([a-zA-Z0-9s]+)$,$1,}

The strange thing is that even though the folder is named
AB001_Project Name_Client Name

I get project and client in $2 $3 respectively, but $1 only ever returns a number.
In the above example it returns 1

I’m really starting to lose my mind with this, and would appreciate any insights!

Thanks

[Bryson Jones]

Man, to be honest, regex is a bear, no matter where you do them. We never do them as part of a “fixed scope”. If a client wants regex, We put a guy on it and we see how long it takes to sort.

One way we deal with them is to break the action out in to a shell script so that you have more control over the action and not be subject to the Worker Node interpretation and then hand the data back to the Worker.

bryson

bryson “at” hidefcowboy.com

hidefcowboy.com

[Rolf Howarth]

You don’t need the ^.* part in your regex above. That means consume any number of characters at the start of the text, but you don’t need that. Instead, try

$q{s/^([^_]*)_.*/$1/}

to pull out everything up to the first underscore

^ = start of line
[^_] = any character other than _
_ = the first underscore
.* anything after the underscore

[David White]

Hi Rolf

Thanks for that, I’ll give it a try later today.

Regards

David