I have a huge collection of .pdf files I want to compress to .zip files according to information found in the filename of the original files.
Example: I want all files that has 20161228 in their filename, like BUAB_20161228_1_01_007.pdf , to be appended to a .zip file with the name BUAB_20161228_1.zip.
One possible way would be to move or copy the file to folders named with parts of filename prior to compressing them.
I suppose this can be achieved using variables in VisualCron. Can anyone point me in the right direction.
Thanks

If I am understanding your request and you are looking to match "20161228" in filenames that end in ".pdf" here's a regular expression that will help:
{REGEX(Match|BUAB_20161228_1_01_007.pdf|^.*20161228.*\.pdf)}
Of course the actual filename would be replaced by a variable of the current filename as you loop through the files. The regular expression looks for the string "20161228" in the filename of a .pdf file. You can get more descriptive and tighten up the matching if the files always follow the pattern you gave in the example to make sure you don't grab a file by accident just because it's name contains the same string.
i.e. this regular expression: ^[A-Z]{4}_20161228_\d_\d{2}_\d{3}\.pdf
Matches the beginning of the line, followed by a character class of an uppercase letter with an indicator showing we want 4 of the previous type of character (so 4 uppercase characters), followed by a literal underscore, followed by the literal 20161228, an underscore, a digit, an underscore, 2 digits, etc.

Hope this helps.

Oh note that in this example the regex matches lowercase "pdf". VisualCron's regex engine is non-standard in that normally one can indicate an OR by using a pipe symbol. I tried using a group of (pdf|PDF) to match on upper or lowercase pdf but it failed since VisualCron reserves the pipe for the argument separator and the paren for the function closing symbol, even when escaped. I could not seem to find any info on alternate characters to use for this which limits the regex functionality somewhat. I suppose before looping through the files you could do a rename in the directory and ensure all extensions are lowercase or something if necessary.

You can do this (if all target file names should be cropped the 11 last characters in the zip file name including extension):

NB: Test thoroughly. There might be bugs.

Create two tasks in the job:
1: A "List file(s)" task for listing the relevant files (enter details in [Folder] and [Include file mask])
2: An "Archive - compress" task that iterates through the list with a "For each x in y" loop

In your particular case you can:

In [Archive settings]:
a: set the [Archive path] by removing the last 11 chars of the file name including extension:
<path to zip-file>\{STRING(Replace,{PATH(GetFileName|{LOOP(CurrentValueXLine)})},{STRING(Right|{PATH(GetFileName|{LOOP(CurrentValueXLine)})}|11)},)}.zip
b: set [Append to archive]

In [Content]:
a: set [Location]/[Folder] = {PATH(GetDirectoryName|{LOOP(CurrentValueXLine)})}
b: set [Location]/[Include file mask] = {STRING(Replace,{PATH(GetFileName|{LOOP(CurrentValueXLine)})},{STRING(Right|{PATH(GetFileName|{LOOP(CurrentValueXLine)})}|11)},)}*.pdf
(get all pdf's having the same name until the last 11 chars including extension)
(c): in [Item settings] you can click [Delete source file after]

Loop:
a: On the job page click the [Tasks] tab
b: Click [Add loop]
c: Set the above archive task in both start and end
d: In [Loop settings] choose [For each x in y]
e: Set [For each row x in] = {TASK(PrevTask,StdOut)}


There might be a more clever and stable way to crop the file name, but I couldn't find more suitable variables for that.