Shouldn't that be 'gimp-image-get-item-position' ?

According to the PDB, 'gimp-image-get-layer-position' is a depreciated procedure, which doesn't support layer groups.

Thanks for pointing out the issues with Layer Groups. They are handled properly now.


Well, in this script, there are only 3 lines of code that need to be changed for GIMP 2.6.x support (see comments). I tried to keep 2.6 in mind while coding and only used procedures that were compatible with both GIMP versions. Only "layer group" specific commands are the ones that need changing. I will release two versions of the script, so as not to leave out GIMP 2.6.x users who might want to take advantage of this script.

Thanks for pointing out the issue with adding inner glow to the border of a non-resized image. I'll ensure that is handled in the next release.

Also, thanks for the contributions on layer naming. I ended up using a little different layer naming scheme in the script because of the layer positioning/merge methods I'm using.

The new version RC 0.6 of the script has been released. See the first post of this thread and/or grab the new version from GIMP Scripts.

I've added support for GIMP 2.8 layer groups, as well as naming shadow layers with selected shadow options. If the shadow name is blank (default), the script uses the selected layer name. On layer merge, the script maintains the selected layer name.



I've released versions for GIMP 2.6.x and GIMP 2.8.x so everyone can take advantage of the script.

Thanks to everyone who has contributed to the development of this script!

I do like the layer naming. Thanks, GnuTux!

Thanks, O. I appreciate and value your input!


Actually, I really appreciate your suggestions and contributions, Graechan. While I might not use your exact coding changes, which is simply a matter of coding style, your input has been extremely helpful, so has Saulgoode's and everyone else who has contributed and offered suggestions to improve the script.

Well, heck yeah. I didn't realize the GIMP version could be determined from within Script-Fu. I had asking (or complained) about this before and I seem the remember the answer was that there was no reliable technique to determine the GIMP version.

If this works, it will be great and I can recombine the two shadow scripts, as well as use it on other scripts. I'll give it a test run.

Thanks so much!

Another option for making your 2.8 script work with GIMP 2.6 is to define the new procedures at the beginning of your script.

@Graechan - Take a look at SG's code. He's checking to see if a GIMP 2.8 procedure has been defined. If it hasn't, you know that it's an earlier (or perhaps later) version of GIMP.

Very nice, SG!

I wasn't aware the "defined?" function existed. That makes it pretty easy to determine the GIMP version.

Is there a list somewhere of all the available functions?

I seem to remember now that I was inquiring about determining the OS version being used (from within Script-Fu), not the GIMP version being used.

Thanks so much. Now I don't have to maintain two versions.

If you specify an outer shadow, your "Allow resizing" toggle always increases the size of the image. It appears that you are performing a 'gimp-layer-resize-to-image-size' before offsetting the shadow layer and that this necessarily places the shadow layer outside the image. In my opinion, this is undesirable behavior because there is no need to increase the image size if the resulting shadow should fall within the boundaries of the original image.

To rectify this, I would propose two modifications to your script. First, only increase the size of the shadow layer enough to ensure that the blurring will not get "chopped off" as discussed previously. Increasing the layer size on all four sides by the amount of the blur should accomplish this.


Then, after you've resized, positioned, and blurred your shadow layer adjust the image size (if necessary) to accommodate the new layer. The following code fragment shows how this could be done.

Try the following:

@ Graechan - That's an interesting routine and it could be useful but why use it when you could just check to see if a GIMP 2.8 only procedure has been defined, as SG demonstrated? If it hasn't, you know that the version of GIMP is earlier than 2.8.x.

@SG - I recognized that there was a resizing deficiency in the script that needed to be addressed. The image just keeps getting bigger and bigger the more layers that are added. After a couple of resizes, I was just toggling off "Allow resize".

Thanks for the code snippets. I'll incorporate those into the next release.

Graechan, I found a list of functions that SG had previously posted. The "defined?" function is in the list.

http://www.gimpchat.com/viewtopic.php?f=8&t=5127#p64036

You should be able to do a check like..


If that statement is true, you know the version of GIMP being used is earlier than V2.8.x. Actually, any procedure with "item" in it is new for GIMP 2.8.x.

I tried this and wilbur said 'no way' hang on I had a typo, it's very dull today and I can't see keyboard properly. got it

I'm Stoked that we don't have to have 2 lots of scripts or leave 2.6 out

Script-fu is for the most part an implementation of version 5 of the Revised Report on the Algorithmic Language Scheme (AKA R5RS). It implements almost all of the functions described in that report -- except that it does not implement either complex arithmetic or the syntax rules macro facility (Script-fu implements macros in a manner similar to the older R4RS-based Schemes).

Script-fu also has two additional procedures, 'strbreakup' and 'unbreakupstr', that are holdovers from earlier versions of Scheme and that are very useful for dividing up strings (one example is Graechan's 'gimp-version-meets?' procedure discussed earlier).