Riley,
I have about 200 documents a day that come in from remote sites via
fax-to-email services, scan-to-email, etc. They all come in as PDF's
and they are all associated to work orders in our job management system.
To make this manageable, I completely control the printing of them, so I
was able to apply bar codes to each that are relevant to their work
order numbers (ScriptMaster).
The requirement was to bring these documents back into the system as
attachments to each job (external container storage). These returning
documents tend to have installers' annotations which provide us with
insight into the details of a given project.
I tested QR, Code39 and Code128 bar coding. I noted the best decode
rate was attained with Code128 (especially for scans or faxes which are
crappy and tended to stretch the image and mess up the aspect ratio).
So... I used ScriptMaster (360Works) and a little Groovy to create a
plugin that creates a Code128 barcode for each printed document. It
also provides QR codes for some other applications we use here.
On the receiving side of things, I needed to explode the inbound PDF's
into separate pages, decode the bar codes and send them to their
associated records.
I created a UI where the guy who aggregates this paperwork can drag and
drop a PDF (containing one or more pages of work orders).
At this point, I needed to decode the barcodes. I tested a couple of
workable methods.
First, I tried MonkeyBread's plug-in which contains a whole load of
functions for manipulating PDF's and decoding plugins. Some are OSX,
some Windows and some are both. Although I was able to have quite good
control over the PDF's, my barcode hit rate was not so good.
Since my user's platform is Windows via Terminal Services, I ended up
using a separate executable utility named, "BarDecodeFiler"
(http://www.bardecode.com/en1/app/bardecodefiler/) that I could call via
windows cmd shell. It handles images in various formats as well as
PDF's. It also works with many types of barcodes and allows you to use
regex's in decoding to assist in the process.
From Filemaker, I programmatically export the target PDF into a specific
folder, and it explodes and renames the PDF's as I have configured it
(which is by the decoded WorkOrder #). Via scripting, these are brought
right back into my UI for review. Most can just be pushed into the
database, but a few which may have munged barcodes need the relationship
set manually... then they are pushed in as well.
This cut a three to four hour job down to 10 minutes.
regards...
Bob Minteer
I have a client that needs to be able to scan a thousand paper documents into a scanner, save the pdf scans to folder, then import the folder into FileMaker putting individual pages into container fields.
Each page has on it a QR code. After importing the images into FileMaker we need to read the QR codes.
We have tried the CNS-Barcode plugin. It is seems to work but gives too large a number of errors (failure to read the code). Do any of you have experience with a good way to accomplish this?
Riley Waugh
_______________________________________________
FMPexperts mailing list
http://lists.ironclad.net.au/listinfo.cgi/fmpexperts-ironclad.net.au
--
Bob Minteer
***@minteer.org