PMGS StampFix

PostmasterGS I upgraded from v1.20 to 1.46 just now, and I seem to have run into a minor glitch. While the stamp detection worked perfectly out of the box, it will not extract and save them. When I click on Wxtract to file, I initially get an hourg;ass, and the button flickers, showing it's been activated, but then nothing. Any ideas what I need to do?

Here is what I have been doing:

I reinstalled v1.46 this time downloading the full package, in case it was an issue with JRE. No change.

I downloaded/installed v1.34. No change.

I downloaded v1.22. It now saves, but it also extracts 34 stamps from my 14-stamp page.

Sure. It will be 20 minutes or so before I can.

Put a number (100) in the Fine Tuning Threshold blank and see if it works.

That blank should have had a default value upon installation, so if you didn't delete it, I'll need to patch that so it's not blank by default.

That box was blank by default.

I noticed on a scan I just completed that 2 of the stamps got rotated 90 deg clockwise.
Thanks again for an awesome program.

1.46

I’ve just uploaded a new version of StampFix, v1.54. As usual, download is here.

This version includes two new features, both by user request.

First, a change to the order in which individual stamps are extracted from a larger image when using the Batch from Image feature. Often, when scanning stamps, a user will place them on the scanner in rows and expect that the resulting images will be numbered in the order they were on the scanner, from top-left to bottom-right. Until now, the stamps were ordered in the order the software detected them, which is roughly bottom to top, without a lot of regard for left to right. (See example order in v1.51 screenshot below). This led to a lot of renaming to get the stamps named in the proper order.



With v1.54, the stamps are numbered in a more logical order, from left to right in rows from top to bottom. For this to work best, the stamps need to be in rows. If the stamps are scattered all over the image, or if there’s a random stamp in one row that’s much taller than the rest so that it intrudes on the row below, it will cause some odd results because it throws off the row detection.

Here's the same image as above in v1.54.



Second, this version includes the ability to further customize the filenames when batch processing using the Batch Folder/Files/Image processes. Filename customization is Pandora’s Box, and it gets very complicated very fast. There’s a reason there are entire apps devoted just to renaming files. But, here’s the logic behind this addition. If you’re going to use this feature, please read carefully, because the renaming behavior isn’t always intuitive or obvious.

There’s a new block of inputs in the left sidebar.



This block includes blanks for the prefix, core filename, and suffix. It also includes blanks for separators to go between the prefix/core and core/suffix.

The checkboxes at left of the prefix/core/suffix blanks will cause those fields to iterate. Iteration works as follows:

1. If iterate is checked and the box is blank, it will insert integers starting at 1. (ex. first stamp would be 1, next is 2, etc.)
2. If iterate is checked and the box contains only an integer (ex. 100), the integer will iterate starting from that number. (ex. first stamp is 100, next is 101, next is 102, etc.)
3. If iterate is checked and the box contains mixed text — IOW, not a pure integer — (ex. 3A), it will append an integer iterating from 1 (ex. first stamp would be 3A1, second would be 3A2, etc.)

One caveat — when using Batch from Image (but not Batch Files or Folder), new unique filenames must be generated since you’re creating multiple files from one. In the past, these new files automatically received integer filenames iterating from 1. This is still the case unless you override this behavior using the new options. If you leave all renaming checkboxes/blanks empty, the app will automatically check the box to iterate the core so it can create core filenames iterating from 1.

Also, the feature that allows you to prevent overwriting of source files is still available. Because of this, if that option is checked, you may still see additional dashes/integers (Ex. -2) appended to filenames as necessary to avoid overwriting files of the same name in your output folder.

These new renaming options are also available on the Edit Individually screen when using Batch from Image.



The renaming works the same way here. If you’d like to watch it in action without generating the files, you can go here and play with the options, hitting Apply to watch the filenames generate.

With an update this large, I suspect there may be a few bugs here and there. If you find one, please let me know. If the numbering or renaming aren’t behaving as you think they should or if you just can’t figure out how to make it work, please don’t hesitate to let me know.

Also, I haven’t yet updated the detailed instructions on my website to reflect these additions. I’ll try to knock that out tomorrow.