Parse data from scanned checks?

I have to download a ton of checks for work, rename them, etc. I save all the checks to a folder called "preprocess"

Right now, when I download the check, I do the following through Hazel:

0 If kind = pdf
0.1 if name does not contain "(OCR)"

Then...

1. Rename the file to YYMMDD check.pdf (the bank's file names are useless)
2. Open the file with Drop2Read (Read Iris's droplet that OCRs the check.)
3. Append "(OCR)" to prevent hazel from processing the check again.

I'd like it if Hazel could do the following:

1. Read the text adjacent to "Account name:"," Check number", "Date posted" and "Amount" and fill in variables
2. Read the text adjacent to "Pay to the order of" and "memo" and fill in variables
3. Append the file name with the check's account, number, posting, amount, payable to and memo

I've been reading MacSparky and my brain caught fire. I'm not adverse to hiring a professional to get me the basic structure, and taking over from there.

Any ideas?

Dave
dlleon@swbell.net

Search the help for "match patterns" to get started. And yes, read the relevant MacSparky articles. This one in particular: http://macsparky.com/blog/2013/6/hazel- ... e-matching

Give it a shot and report back with specific problems you run into. Or if you are set on hiring someone, maybe someone here would be up for that.

Is there a sample rule set? I've been working with the MacSparky video, but I'm just not getting it.

Also, what do the people there charge to write a custom rule? I think once I get some samples to work, I'll be fine.

There is no sample rule set for this as it is different for everyone. Is there a particular part of the MacSparky video that you are having trouble with?