Image processor cs 4 millimeter support

I am a new user javascript and i want to ask the following. Is it possible to change the image processor script to use millimeters instead pixel?

The short answer is yes. But the easy answer is download Dr. Brown's 1-2-3 Process. It is basically an extended version of Image Processor. It allows you to set the units for resize to any unit Photoshop supports.

Similar Messages

Image Processor not working in CS 5

I have installed CS5 Design Premium on my Mac Pro at work. (Had installed the education demo, but it turns out my school doesn't fit Adobe's critereon for eligibility - and doesn't take it back after telling you that) although we have qualified for years, and we education people in 14 countries are are the top school in the world on our subject. No biggie, the upgrade's only $50 more with the NAPP discount.
But at home it installed just fine on my Mac Pro and works great. But at work, where CS4 works flawlessly, I can't get it to run Image Processor from Bridge, or import photos into Photoshop Layers, it does do Photomerge, but it also does not do HDR PRo.
Before the settings wouldn't stick in image processor, and I fixed the issue by changing permissions on the Photoshop settings folder in my user account. Now settings stick for Image Processor, but when I tell it to process a set of images, it keeps coming up with a dialog saying "Sorry, I could not process the following files..." and then gives a list of each file with it's the path.
I'm wondering if the javascrpits running this process are broken, or have permission problems. (Image Processor, HDR Pro, etc.)
If in Bridge, I seelct Tools->Photoshop->Batch, it gives me an error dialog: "FatalError: General Photoshop error occurred. This functionality may not be available in this version of Photoshop. <No additional Information Available>"
I rely on Photoshop for a very large portion of my work. And that includes processing thousands of images. I cannot see how this problem is unique to me. Is anyone else having this problem? Is there a fix? Am I stuck with CS4 for automation? BTW, I can use Applescripts to automate Photoshop CS5. How wierd is that? Scripts I wrote work, but built-in ones appear to be broken on this machine.
Additional information: I'm on a Windows network using Active Directory and my company user account while this is happening. Where it works at home, it's just my own Mac running OS X 10.6.4. I have all the latest CS5 updates installed.

But all the
files that are in my home directory (Snow Leopard 10.6.5) give that error.
Do you have a beta of Snow Leopard, to my best knowledge 10.6.4 is the
latest version??
I'm going to format
the drive on Monday and restore my system from Backup and see what that does.
First try the uninstall CS5 suite using the uninstaller that is in each
application folder. Also erase your serial number from computer and run the
clean script:
http://www.adobe.com/support/contact/cs5clean.html
It is useless to reformat and after that reinstall from Back Up when you
also re install the current preferences, highly likely you will end up with
the same problem because you copied the old prefs again

Image Processor Not Working

Hi,
I used the Image Processor feature to resize and save in a separate folder all the time in Bridge CS3. I recently upgraded to CS4 and I can't get the Image Processor to work. When I select the feature, it opens photoshop but nothing happens...it just opens the photoshop window but does nothing. Can anyone help? Does the feature not function in the same way for CS4? How do I properly use Image Processor in Bridge CS4? Thank you!

But all the
files that are in my home directory (Snow Leopard 10.6.5) give that error.
Do you have a beta of Snow Leopard, to my best knowledge 10.6.4 is the
latest version??
I'm going to format
the drive on Monday and restore my system from Backup and see what that does.
First try the uninstall CS5 suite using the uninstaller that is in each
application folder. Also erase your serial number from computer and run the
clean script:
http://www.adobe.com/support/contact/cs5clean.html
It is useless to reformat and after that reinstall from Back Up when you
also re install the current preferences, highly likely you will end up with
the same problem because you copied the old prefs again

Image Processor Pro won't run in PS6 using Window 7 OS.

Trying to run IPP on a Windows 7 OS in PS6. I select my images in bridge and run the script IPP. Photoshop pops up and a blank white screen appears and nothing else happens.

Scripting was an extra down for PS7 and the first version of Photoshop to have scripting support PS6 has no scripting support. The Image Processor Pro script uses newer scripting features like Plug-in support that was added in CS3 and I think Image Processor Pro needs newer then that like it may only works with CS5, CS6 and CC. I use it with CS6 and CC... I do not usr the installer I extract the zxp file and install manually. There are a couple of minor issues the X is working on Alpha Channels in PSD files and some TGA and BMP support... Adobe has pull Russell Brown off of Photoshop and X is feed up with Adobe.. Make sure you have the latest version of the script... I use Windows 7 Pro OS...
CS5/CS6 Script:
Dr. Brown’s Services 2.3.1
Dr. Brown’s Services is a collection of time-saving utilities for designers and photographers. Use the new CS6 Extension Manager to install both the Mac and PC versions. If the Extension Manager does not work on your system, use the manual installation process instead. Installation will require that you be in the ADMINISTRATOR mode on a PC.
This installation includes:
Image Processor Pro
Stack-A-Matic
Place-A-Matic 8-bit
Place-A-Matic 16-bit
Trouble Shooting
Having trouble with any of my scripts? Then be sure and try my "JUST FIX IT" http://www.russellbrown.com/scripts.html#Fix_Itscript that will remove all the damaged preference files. This could to the trick to solve your issues.
IMAGE PROCESSOR PRO
STACK-A-MATIC
PLACE-A-MATIC
Adobe Photoshop CS6 Installer
Adobe Photoshop CS5 Installer

Image Processor Jpg Files Cause LR to Stop Working

Photoshop CS3 and Lightroom 1.4.1
Vista Ultimate 64 / 8 Gig Ram / Lots of Hard Drive space on Scratch Drive
nVidia 8800 GTS w/640 MB Ram
Image files processed in Photoshop
Exported via: File > Scripts > Image Processor (IP)
IP settings are default except the jpg output is set to 12/max
From the same PSD file, no modifications in PS
One file will cause LR to crash ("A problem caused the program to stop working correctly) the next does not.
Both jpg files open and display without a problem in...
- Photoshop
- Adobe Camera Raw
- Microsoft Office Picture Viewer
- Apple Quicktime Picture Viewer
- Windows Explorer (thumbnails)
I have saved two otherwise identical files, one crashes LR every time, the other opens in LR fine, every time.
This has been occurring in about 3 out of every 100 files saved via IP.
The next time LR is opened it crashes again if the 'bad' file is not renamed or deleted due to the automatic file updating process.
Trying to import 100 files requires importing them one at a time when this happens because there is nothing to indicate which file cause LR to break during the batch import.
If Adobe Tech Support would like the two 'identical' files they can email me and I'll send them along.

*push*

Bridge Image Processor always creates JPEG or TIFF folder, how do I change that behavior?

I have searched the web, Bridge forums, and Photoshop forums, and come up empty-handed. (The strange thing is, I know there used to be a post that answered this question a few years ago.) Anyway, I use Image Processor quite a bit, and it works great except for the annoying behavior of always creating a new folder to store the newly-created images.
I remember the solution is to change some line in a (XML?) file from "JPEG" to "". For the life of me, I can't find which file that is! I think it used to located down inside Application Support. I think.
Thanks for help on this, it's driving me crazy!
Jeff

Thanks for replying! I'm running CS 5.1. Oh, now I see...I've never run Image Processor from within Photoshop, just from Bridge. You're right, it does offer different options than Bridge.
Anyway, I found it what I was looking for.
Open up (don't run!) the Image Processor.jsx file. I opened it up in ExtendScript Toolkit. The path is /Applications/Adobe Photoshop CS5.1/Presets/Scripts/Image Processor.jsx
I'm on line 1574 (in other versions, the line number may vary)
This is what the line says:
var subFolderText = inFolderLocation + "/JPEG/";
Change to:
var subFolderText = inFolderLocation + "//";
Problem solved. Repeat as necessary for other folders you don't want created (TIFF, PSD)
Back to work!

Unable to get Image Processor Pro to run on Laptop. Creative CC suite.

Hi, All,
now i been using the excellent plugin from Russel Brown on my work PC for as long as i can remember. I had to do some work from home on Windows 8.1 Laptop 64Bit.
However this strange issue as show as image below just unable to get past this point. Seems like something is stuck. I can close it only. left it for ages. I even tried to trace the system with Microsoft Systemmonitor with no luck.
I have re-installed the creative cc bridge now three times along with photoshop, cleaning out every reference to data settings etc..Still with same results.
Even from outside bridge just trying to run the script from Photoshop directly gives the same result.
Anything have any great suggestion what possibly could be reason. I got a ton of batch conversions i need to save PNGs otherwise i would just just the normal image process.
Thanks in advance.

Its Not Adobe Code and its not shipped with Photoshop. Photoshop versions also has bugs in ScriptUI which is used by scripts with a dialog. Image Processor Pro 2.3.1 has some bugs. Some files formats its saves may have problems.
Russell Brown works for Adobe but he has been pulled away from Photoshop. Xbytor is the programmer that wrote Image Processor Pro he does not work for Adobe and he is feed up with Adobe for not fixing bugs in Photoshop scripting. He stated he will support Image Processor Pro for some time.
Russell web site has this. The Open Source code is now available for many of my script and panels
Because of new projects and a changing demand for my time, it will be necessary for me to move on to new projects at Adobe. For this reason I will no longer be able to support, or update many of my scripts and panels. Never fear! – I am releasing the source code for many of my programs and if you are a programmer then you can use this code to modify and change my scripts to meet your needs. If you are not a programmer then I recommend working with anyone on the list of programmers that I have supplied below. However! Let me make it clear that his code is not to be sold. This is an open and free source to be used for the good of humankind and Adobe Photoshop users around the world
===========================
I do not have a Mac or a Windows 8 machine to test on. On my Windows 7 PC Image Processor Pro 2.3.1 and newer versions run on CS6, CC and CC 2014.. X is still working on it

Image Processor from Bridge CS5

Lately, I have been getting this error message when trying to access the Image Processor (PS CS5)from Bridge CS5: "Sorry something major happened and I can't continue". I called Adobe support and was told that it might be related to third party plugins. The puzzling part is that I have been using all of the above with no problems for over a year . . I am using a Mac Pro, OS 10.6.4, with 10 G of RAM, and have the latest updates to PS and Bridge. Any suggestions?

One way around the problem would be to use a different processor..
There is the New Image Processor Pro and Dr Browns 1-2-3 processor @
http://www.russellbrown.com/tips_tech.html
Also Picture Processor @ http://www.scriptsrus.talktalk.net/index.htm

Scripts/Image Processor Error Message

When using Image Processor on Photoshop CS5, what does the following error message mean and how can I fix this issue? ReferenceError: Window does not have a constructor : 327

Contact Adobe support, you hit a bug. I never seen that one pop up before.
But in general what it is saying is it is trying to create a window and to do so it needs to reference a constructor.
That constructor should have been automatically created or a programmer creates one from scratch, that depends on how the code was created. Since that bug normally does not appear, I would say a some code pointed to the wrong location in memory hence the word reference. (In generic terms reference means to point to a location in memory)
Try it again and see if it repeats itself. I think Adobe will tell you that you need to reinstall photoshop, but its better that they tell you that as I could easily be misinterpreting that error message.

Image processor always on top?

When I use image processor (launched by Photoshop or Bridge) Photoshop go in front (top window) each new picture processed, means every seconds.
It's pretty annoying cause I can't even keep a browser in front...
Does anybody have the same problem?
It comes with CS4 and CS5, on 2 different machines.
How can I fix that?

The Image Processor resize is a fit image type resize not an Image Size dialog type resize. Let say you have a 3:2 aspect ratio camera. Landscale image have 3:2 aspect ratio and portrait have 2:3 aspect ratio. You want the resize your large camera images for the web. You want the longest side to be 600 px. You would set the width and height values 600 px. Your image would be resize constraine to retain their aspect ratios. So landscape images would resize to 600px by 400 px and Portraits would resize to 400 px by 600 px. You can also resize Image fot a device like your 4:3 UXGA display set 1600 in for the width and 1200 for the height Portrait would still be 2:3 with a 1200 px height and Landscape will still be 3:2 with a 1600 px width. When displayed on you 4:3 display the whole images would fit on your display and fil the width or hight depending in the images orientation ther woul be a border on two sides.
Photoshop also shipts with a FitImage Plug-in so you can record using the plug-in in as action. When you do the values you set in the fitimage dialog while recording the action are recorded into the action step. When the action plays back the values are passed to fitimage and the dialog for the step is not diaplayed. Unless you thurn on that steps dialog. If you do you will see the recorded values are set into the dialog. So you don't need to write a script to handle resize for different orientations. It just so happens to be that since CS3 FitImage is a script. So if you want to pass some values to a script. Write the script as a plugin and record as action to pass recorded values. Plug-in scripting support was added in CS3. The script needs to record its dialog setting into the action step and also not display it dialog if it being run from a recorded action step. To learn more read FitImage.jsx in Photoshop's Presets\Scripts folder.

Photoshop Cs 4 Image Processor

Unable to use image processor feature in CS 4. Error message indicates "Cannot find javascript plug-in. This feature was working just fine until a few weeks ago. Nothing unusual has changed. Thanks

the folder must have moved or something...
here it is for now copy it to ... applications/adobe photoshop cs4/presets/scripts : and it will be a good time to see if that folder is even there.
Regards
G
// (c) Copyright 2006-2007. Adobe Systems, Incorporated. All rights reserved.
// Photomerge in ExtendScript.
// Translated from the original C++ automation & filter plugins
// John Peterson, Adobe Systems, 2006
// Adobe Patent or Adobe Patent Pending Invention Included Within this File
@@@BUILDINFO@@@ Photomerge.jsx 3.0.0.1
// BEGIN__HARVEST_EXCEPTION_ZSTRING
<javascriptresource>
<name>$$$/JavaScripts/Photomerge/Menu=Photomerge...</name>
<about>$$$/JavaScripts/Photomerge/About=Photomerge^r^rCopyright 2006-2008 Adobe Systems Incorporated. All rights reserved.^r^rCombines several photographs into one continuous image.</about>
<menu>automate</menu>
</javascriptresource>
// END__HARVEST_EXCEPTION_ZSTRING
// on localized builds we pull the $$$/Strings from a .dat file
$.localize = true;
// Put header files in a "Stack Scripts Only" folder. The "...Only" tells
// PS not to place it in the menu. For that reason, we do -not- localize that
// portion of the folder name.
var g_StackScriptFolderPath = app.path + "/"+ localize("$$$/ScriptingSupport/InstalledScripts=Presets/Scripts") + "/"
                                        + localize("$$$/Private/Exposuremerge/StackScriptOnly=Stack Scripts Only/");
$.evalFile(g_StackScriptFolderPath + "LatteUI.jsx");
$.evalFile(g_StackScriptFolderPath + "StackSupport.jsx");
$.evalFile(g_StackScriptFolderPath + "CreateImageStack.jsx");
$.evalFile(g_StackScriptFolderPath + "Geometry.jsx");
$.evalFile(g_StackScriptFolderPath + "PolyClip.jsx");
// debug level: 0-2 (0:disable, 1:break on error, 2:break at beginning)
// Must leave at zero, otherwise trapping gFileFromBridge fails on QA's debug builds.
$.level = 0; // (Window.version.search("d") != -1) ? 1 : 0;
// debugger; // launch debugger on next line
if (typeof(PMDebug) == 'undefined')
    var PMDebug = false;
// photomerge routines
// Debug - write the trapezoid in Matlab format
function dumpTrap( name, corners )
    var i;
    if (! PMDebug) return;
    $.write( name + "= [" );
    for (i in corners)
        $.write( ((i > 0) ? "; " : "" ) + corners[i].fX + " " + corners[i].fY );
    $.writeln( "];");
StackElement.prototype.dumpMLCorners = function()
    // Weed out file suffix (chokes matlab)
    dumpTrap( this.fName.match(/([^.]+)/)[1], this.fCorners );
// Set the fCorners of the layer to the bounds of the Photoshop layer.
StackElement.prototype.setCornersToLayerBounds = function( stackDoc )
    if (typeof(stackDoc) == "undefined")
        stackDoc = app.activeDocument;
    var bounds = stackDoc.layers[this.fName].bounds;
    this.fCorners = new Array();
    this.fCorners[0] = new TPoint( bounds[0].as("px"), bounds[1].as("px") );
    this.fCorners[2] = new TPoint( bounds[2].as("px"), bounds[3].as("px") );
    this.fCorners[1] = new TPoint( this.fCorners[2].fX, this.fCorners[0].fY );
    this.fCorners[3] = new TPoint( this.fCorners[0].fX, this.fCorners[2].fY );
// Add the corner data to the string of per-stackElement information
// that gets passed to the filter plugin
StackElement.prototype.addPieceData = function()
    if (typeof(this.fCorners) != "undefined")
        // Add corners in place of trailing '\n'
        this.fString = this.fString.slice(0,-1) + "fCorners=";
        for (j = 0; j < 4; j++)
            this.fString += " " + this.fCorners[j].fX.toString() + " " + this.fCorners[j].fY.toString();
        this.fString += "\t";
        if (typeof(this.fScale) != "undefined")
            this.fString += ("fScale=" + this.fScale.toString() + "\t");
        if ((typeof(this.fConnectedTo) != "undefined") && this.fConnectedTo)
            this.fString += "fConnectedTo=" + this.fConnectedTo.fLayerID + "\t";
        if (typeof(this.fLayerID) != "undefined")
            this.fString += "fLayerID=" + this.fLayerID.toString() + "\t";
        this.fString += "\n";
    else
        debugger;    // Corner data missing!
StackElement.prototype.overlapArea = function( other )
    if (other == this)
        return TPoint.polygonArea( this.fCorners );
    var overlapBounds = TRect.intersection( this.fBoundsCache, other.fBoundsCache );
    if (overlapBounds.isEmpty())
        return 0.0;
    var clipPoly = TPoint.intersectConvexPolygons( this.fCorners, other.fCorners );
    if (! clipPoly)
        return 0.0;
    else return TPoint.polygonArea( clipPoly );
// Find the points where the two quadrilaterals intersect (yes, eight is a theoretical max)
// Note situations where one piece intersects the same edge twice - special
// case handled by SoftEdgeBlend
StackElement.prototype.findQuadIntersections = function( other, intersections )
    var i, j;
    var curIntersections, numIntersections = 0;
    var innerEdgeIntersections = [ 0, 0, 0, 0 ];
    var outerEdgeIntersections = [ 0, 0, 0, 0 ];
    var thisEdge = false;
    var otherEdge = false;
    for (i = 0; i < 4; i++)
        var eb0 = other.fCorners[i];
        var eb1 = other.fCorners[(i + 1) > 3 ? 0 : i + 1];
        curIntersections = numIntersections;
        for (j = 0; j < 4; j++)
            var ed0 = this.fCorners[j];
            var ed1 = this.fCorners[(j + 1) > 3 ? 0 : j + 1];
            var cross = TPoint.lineSegmentIntersect( eb0, eb1, ed0, ed1 );
            if (cross != TPoint.kInfinite)
                intersections[numIntersections++] = cross;
                innerEdgeIntersections[j]++;
        outerEdgeIntersections[i] = numIntersections - curIntersections;
    if (numIntersections == 2)
        for (i = 0; i < 4; i++)
            if (innerEdgeIntersections[i] == 2) thisEdge = true;
            if (outerEdgeIntersections[i] == 2) otherEdge = true;
    return {"numIntersections":numIntersections, "thisEdge":thisEdge, "otherEdge":otherEdge};
// Look for a point of this that's inside the corners of other.
StackElement.prototype.findSingleInsidePoint = function( other )
    var i;
    var result = new Object();
    result.numFound = 0;
    for (i = 0; i < 4; i++)
        if (this.fCorners[i].pointInQuad( other.fCorners ))
            result.numFound++;
            result.insidePt = this.fCorners[i];
    return result;
// "Adobe patent application tracking # B349, entitled 'Method and apparatus for Layer-based Panorama Adjustment and Editing', inventor: John Peterson"
// Because we're dealing with quads, we can't use the simple scheme
// to generate blend rects that the rectangles use - black stuff
// from the rects will seep in. So instead, we use various heuristics to
// figure out how the quads themselves intersect. This isn't too hard
// when there's just two intersection points (usual case), but there are
// some pathological cases where there are many more intersections (the octogon
// from one square 45 deg. off from another is the classic example).
StackElement.prototype.softEdgeBlend = function( other, blendRad )
    var i, j;
    var intersections = new Array();    // Worst case is a square inside a 45 degree rot square
    // If there's no distortion, do the blend strictly on the rectangles
/*    if ((!IsQuadMapped() && !other->IsQuadMapped())
        || (IsRectilinear() && other->IsRectilinear()))
        Assert_( other->fWarpedRaster );
        fWarpedRaster->SoftEdgeBlendRasters( *(other->fWarpedRaster), blendRad );
        return;
    var bounds = this.getBounds();
    // Find the points where the two quadrilaterals intersect
    var intResult = this.findQuadIntersections( other, intersections );
    var numIntersections = intResult.numIntersections;
    var thisEdgeTwice = intResult.thisEdge;
    var otherEdgeTwice = intResult.otherEdge;
    var thisInsidePt = this.findSingleInsidePoint( other );
    var otherInsidePt = other.findSingleInsidePoint( this );
    // If quads don't overlap, just bail
    if (numIntersections == 0)
        return;
    // Handle cases where just one point overlaps the other piece
    var insidePoint;
    if (otherEdgeTwice || ((numIntersections == 2) && (thisInsidePt.numFound == 1)))
        insidePoint = thisInsidePt.insidePt;
        this.makeBlendTrapezoid( intersections[0], intersections[1], insidePoint, blendRad, false );
        return;
    if (thisEdgeTwice /*|| ((numIntersections == 2) && (otherInsidePt.numFound == 1))*/)
        insidePoint = otherInsidePt.insidePt;
        this.makeBlendTrapezoid( intersections[0], intersections[1], insidePoint, blendRad, false );
        return;
    // ...More than two corners overlap, apply heuristics to find reasonable blending
    $.bp( numIntersections > 8 );
    // If we got more than two points, pick the two furthest apart
    if (numIntersections > 2)
        var max0, max1;
        var maxDist = -1.0;
        for (i = 0; i < numIntersections - 1; i++)
            for (j = i + 1; j < numIntersections; j++)
                var dist = (intersections[i] - intersections[j]).vectorLength();
                if (dist > maxDist)
                    max0 = intersections[i];
                    max1 = intersections[j];
                    maxDist = dist;
        $.bp( maxDist <= -1.0 );
        intersections[0] = max0;
        intersections[1] = max1;
        numIntersections = 2;
    TPoint.clipLineToRect( bounds, intersections[0], intersections[1] );
    // The point furthest away from the cut line on the "other" image
    // is the one that we blend toward (the "dark" corner).
    var maxPoint = 0, maxDist = -1;
    for (i = 0; i < 4; i++)
        var dist = Math.abs( other.fCorners[i].distanceToLine( intersections[0], intersections[1] ) );
        if (dist > maxDist)
            maxDist = dist;
            maxPoint = i;
    this.makeBlendTrapezoid( intersections[0], intersections[1], other.fCorners[maxPoint], blendRad, true );
// "Adobe patent application tracking # B349, entitled 'Method and apparatus for Layer-based Panorama Adjustment and Editing', inventor: John Peterson"
// Create a layer mask that fades out from the edge0,edge1 cutline towards "insidePt"
// If two points are inside, construct the mask fade from the edge of the image.
//                       insidePt
//             |            *           |
//             |           / \          |
//             |          /   \         |
// insetLine-> |       +-/-----\-+      | --
//             |       |/       \|      |   | radius
// baseLine-> +-------*---------*------+ --
//                    /edge0      \edge1
StackElement.prototype.makeBlendTrapezoid = function( edge0, edge1, insidePt, radius, useCorners )
    function sgn(x) { if (x < 0) return -1; if (x > 0) return 1; return 0; }
    function wrap4(i, next)
        i += next;
        if (i > 3) return i % 4;
        if (i < 0) return 3;
        return i;
    // Create a layer mask
    selectOneLayer( app.activeDocument, this.fName );
//    app.activeDocument.activeLayer = app.activeDocument.layers[this.fName];    // Broken if multiple layers selected.
    createLayerMask();    // Does nothing if the layer already has a mask
    // Create vector perpendicular to edge towards insidePt
    var edgeDir = edge1 - edge0;
    var blendDir = new TPoint( -edgeDir.fY, edgeDir.fX );
    blendDir /= blendDir.vectorLength();    // Make unit length
    // Make a polygonal selection covering the area
    var dist = insidePt.distanceToLine( edge0, edge1 );
    var blendOffset = blendDir * dist;
    var blendBox;
    if (! useCorners)
        blendBox = [edge0, edge0+blendOffset, edge1+blendOffset, edge1, edge0];
    else
        // If the cutline slices across the image (two points on each side), then
        //        - Look for the edge edge0 is on
        //        - Figure out which corner is on the same side as "insdePt"
        //        - Construct the blendBox from that.
        var i, nextPtInd;
        for (i = 0; i < 4; i++)
            if (edge0.distanceToLine( this.fCorners[i], this.fCorners[wrap4(i,1)] ) < 0.0001 )
                if (this.fCorners[i].sideOf( edge0, edge1 ) == sgn( dist ))
                    nextPtInd = -1;
                else
                    nextPtInd = 1;
                blendBox = new Array();
                blendBox[0] = edge0;
                blendBox[1] = (nextPtInd < 0) ? this.fCorners[i] : this.fCorners[i+1];
                blendBox[2] = (nextPtInd < 0) ? this.fCorners[wrap4(i, -1)] : this.fCorners[wrap4(i, 2)];
                blendBox[3] = edge1;
                break;
        $.bp( i == 4 );    // Never found edge0?
    createPolygonSelection( blendBox );
    // Fill it.
    var midPoint = (edge0 + edge1) * 0.5;
    gradientFillLayerMask( midPoint, midPoint + blendDir * radius * sgn(dist) );
    app.activeDocument.selection.deselect();
// Photomerge base class
const kPhotomergeAdvancedBlendingFlag = app.stringIDToTypeID( "PhotomergeAdvancedBlendingFlag" );
photomerge = new ImageStackCreator( localize("$$$/AdobePlugin/Shared/Photomerge/Process/Name=Photomerge"),
                                          localize('$$$/AdobePlugin/Shared/Photomerge/Auto/untitled=Untitled_Panorama' ), null );
// For now, alignment is turned OFF, because we want to
// invoke it independantly.
photomerge.useAlignment            = false;    // We do the alignment, not PS
photomerge.hideAlignment        = true;
photomerge.mustBeSameSize        = false;
photomerge.mustBeUnmodifiedRaw = false;
photomerge.mustNotBe32Bit        = ! app.featureEnabled(localize("$$$/private/32BitLayersFeature=32-Bit Layers"));
photomerge.radioButtons = ["_LOauto", "_LOperspective", "_LOcylindrical", "_LOspherical", "_LOcollage", "_LOnormal", "_LOinteractive"];
photomerge.interactiveFlag        = false;
photomerge.alignmentKey            = "Auto";    // Defaults to perspective
photomerge.compositionFile        = null;
photomerge.advancedBlending        = true;
photomerge.lensCorrection        = false;
photomerge.removeVignette        = false;
try {
// We want to steer people to the advanced blending option,
// so have it be on by default, rather than sticky.
//    var desc = app.getCustomOptions("PhotomergeFlags001");
//    photomerge.advancedBlending = desc.getBoolean( kPhotomergeAdvancedBlendingFlag );
catch (e)
// Get the bounds of all of the stackElements.
photomerge.getBounds = function()
    var i;
    for (i in this.stackElements)
        if (i == 0)
            this.fBounds = this.stackElements[i].getBounds();
        else
            this.fBounds.extendTo( this.stackElements[i].getBounds() );
    return this.fBounds;
// Align selected layers by content (uses SIFT registration in Photoshop core)
// This just returns the alignment data, it does not actually transform the layers
// unless doTransform is true
photomerge.getAlignmentInfo = function( stackDoc, doTransform )
    selectAllLayers(stackDoc, 1);
    const kMargin = 10;
    function offsetGroup( delta, group )
        group.bounds.offset( delta );
        var k;
        for (k = 0; k < group.layers.length; ++k)
            group.layers[k].offset( delta );
            if (doTransform)
                selectOneLayer( stackDoc, group.layers[k].fName );
                // Translate gets broken when document DPI isn't 72 DPI...(PR 1417264)
//                activeDocument.activeLayer.translate( UnitValue( delta.fX, "px" ), UnitValue( delta.fY, "px" ) );
                translateActiveLayer( delta.fX, delta.fY );
    var i, j, alignInfo;
    var alignmentFlags = [];
    if (this.lensCorrection) alignmentFlags.push(kradialDistortStr);
    if (this.removeVignette) alignmentFlags.push(kvignetteStr);
    alignInfo = getActiveDocAlignmentInfo( this.alignmentKey, doTransform, alignmentFlags );
    // If the alignment fails completely, fake up a plan B...
    // For now, just set the images side by side.
    if (! alignInfo)
        alert(localize("$$$/AdobePlugin/Shared/Photomerge/alignbad=Some images could not be automatically aligned"));
        var xpos = 0;
        for (i in this.stackElements)
            this.stackElements[i].setCornersToSize();
            this.stackElements[i].offset( new TPoint( xpos, 0 ) );
            xpos += this.stackElements[i].getBounds().getWidth() + kMargin;
        this.fGroups = null;
    else
        var layerList = alignInfo.layerInfo;
        this.fGroups = new Array();
        for (i = 0; i < layerList.length; ++i)
            // Note we depend on stackElement's order matching
            // the document's sheet list!
            var curGroup = layerList[i].groupNum;
            if (!doTransform && (layerList[i].corners.length > 0))
                this.stackElements[i].fCorners = layerList[i].corners;
            else
                this.stackElements[i].setCornersToLayerBounds( stackDoc );
            this.stackElements[i].fAlignGroup = curGroup;
            this.stackElements[i].fBaseFlag = layerList[i].baseFlag;
            if (typeof(this.fGroups[curGroup]) == "undefined")
                this.fGroups[curGroup] = new Object();
                this.fGroups[curGroup].hasCorners = layerList[i].corners.length > 0;
                this.fGroups[curGroup].bounds = this.stackElements[i].getBounds();
                this.fGroups[curGroup].layers = new Array();
                this.fGroups[curGroup].xformType = layerList[i].xformType;
            else
                this.fGroups[curGroup].bounds.extendTo( this.stackElements[i].getBounds() );
            this.fGroups[curGroup].layers.push( this.stackElements[i] );
        // Now move the groups into place
        // Note carefully: if the corners were given, then the group is already
        // transformed into the proper spot, and we just need to move the corners to
        // match. So shut off moving the layer pixels around from here on out.
        offsetGroup( -this.fGroups[0].bounds.getTopLeft(), this.fGroups[0] );
        for (i = 1; i < alignInfo.numGroups; ++i)
            var spacing = Math.round(this.fGroups[i-1].bounds.getHeight() / 10.0);
            offsetGroup( -this.fGroups[i].bounds.getTopLeft() + new TPoint(0, spacing + Math.round(this.fGroups[i-1].bounds.fBottom)),
                           this.fGroups[i] );
    this.getBounds();
// The original Photomerge plugin needs to have the "connectivity" of the
// pieces when in perspective mode, i.e., a pieces distortion is based
// on the distortion of the one it overlaps most. This takes the
// "base" piece information from the PS core and uses overlap area to
// determine this.
photomerge.setupConnectivity = function()
    var i;
    // See if stackElem is connected to the "base". If "without"
    // is given, then the path to the base must not use "without"
    function isConnectedToBase( stackElem, without, dbg_count )
        if (typeof(dbg_count) == "undefined")
            dbg_count = 0;
        if (typeof(without) == "undefined")
            without = null;
        $.bp( dbg_count > 150 );    // oops, got stuck in a loop...
        if (stackElem == null)
            return false;
        if (stackElem == without)
            return false;
        if (stackElem.fConnectedTo == stackElem)
            return true;        // Already at base
        return isConnectedToBase( stackElem.fConnectedTo, without, dbg_count + 1 );
    // Initialize
    for (i in this.stackElements)
        this.stackElements[i].fBoundsCache = this.stackElements[i].getBounds();
        this.stackElements[i].fLayerID = i;
        // Bases connect to themselves.
        this.stackElements[i].fConnectedTo = this.stackElements[i].fBaseFlag ? this.stackElements[i] : null;
        this.stackElements[i].fNeighborOverlap = 0;
    // Create a connection table based on the overlap of the pieces
    var g, i, j, baseInd = -1;
    if (this.fGroups)
        for (g in this.fGroups)
            var group = this.fGroups[g];
            var connections = new Array();
            var numLayers = group.layers.length;
            // Ignore orphan images
            if (numLayers < 2)
                group.layers[0].fBaseFlag = false;
                group.layers[0].fConnectedTo = null;
                continue;
            // If there's no perspective, there's no connections ("0" is kProjective in UAlignment.h)
            if (group.xformType != 0)
                for (i = 0; i < numLayers; ++i)
                    group.layers[i].fBaseFlag = false;
                    group.layers[i].fConnectedTo = group.layers[i];
                continue;
            for (i = 0; i < numLayers; ++i)
                if (group.layers[i].fBaseFlag)
                    baseInd = i;
                connections[i] = new Array();
                group.layers[i].fGroupIndex = i;
                if (i > 0)
                    for (j = 0; j < i; ++j)
                        connections[i][j] = group.layers[i].overlapArea( group.layers[j] );
                        connections[j][i] = connections[i][j];    // table is symentric
            $.bp( baseInd == -1 );    // Never found the base?
            // Debug - dump the connection table
    /*        for (i = 0; i < numLayers; ++i)
                var s = "";
                for (j = 0; j < numLayers; ++j)
                    s += ", " + Math.floor(connections[i][j]);
                $.writeln(s);
            // Connect everything to the base that's connected.
            for (i = 0; i < numLayers; ++i)
                if ((i != baseInd) && (connections[baseInd][i] > 0))
                    group.layers[i].fConnectedTo = group.layers[baseInd];
                    group.layers[i].fNeighborOverlap = connections[baseInd][i];
            // Walk the cconnectivity table and make sure everything is
            // "optimally" connected.
            var changes = false;
            do {
                changes = false;
                for (i = 0; i < numLayers; ++i)
                    if (i != baseInd)
                        var curLayer = group.layers[i];
                        for (j = 0; j < numLayers; ++j)
                            if (((j != baseInd) && (j != i))
                                && (((connections[i][j] > curLayer.fNeighborOverlap)
                                        && isConnectedToBase( group.layers[j], curLayer ))))
                                curLayer.fConnectedTo = group.layers[j];
                                curLayer.fNeighborOverlap = connections[i][j];
                                changes = true;
            } while (changes);
    //        for (i = 0; i < numLayers; ++i)
    //            $.writeln( group.layers[i].fName + " is connected to " + (group.layers[i].fConnectedTo ? group.layers[i].fConnectedTo.fName : "??") );
photomerge.offsetStack = function( delta )
    for (i in this.stackElements)
        this.stackElements[i].offset( delta );
    this.fBounds.offset( delta );
photomerge.scaleStack = function( s )
    for (i in this.stackElements)
        this.stackElements[i].scale( s );
    this.getBounds();
// This gets executed before a filter plugin is invoked. "desc"
// allows passing parameters to the filter.
photomerge.customPluginArguments = function( desc )
    var f = new File(this.stackElements[0].fFullName);
    var path = File.encode( f.parent.fsName ) + (File.fs == "Windows" ? "\\" : "/");
    desc.putString( app.charIDToTypeID('PMfp'), path );
    if (this.compositionFile)
        desc.putString( app.charIDToTypeID('PMrf'), this.compositionFile.fsName );
        desc.putString( app.charIDToTypeID('PMcf'), File.encode( this.compositionFile.fsName ) );
photomerge.callInteractivePlugin = function( stackDoc )
    // Scale the results to fit the screen
/*        var scaleFactor = 1.0, screenSize = primaryScreenDimensions() * 0.75;
    if (this.fBounds.getHeight() > this.fBounds.getWidth())
        if (this.fBounds.getHeight() > screenSize.fY)
            scaleFactor = screenSize.fY / this.fBounds.getHeight();
    else
        if (this.fBounds.getWidth() > screenSize.fX)
            scaleFactor = screenSize.fX / this.fBounds.getWidth();
    const kMaxPieceSize = 1024;        // Must match value in PhotomergeUI.cpp
    var i;
    // The old plugin insists on eight bit data.
    if (stackDoc.bitsPerChannel != BitsPerChannelType.EIGHT)
        stackDoc.bitsPerChannel = BitsPerChannelType.EIGHT;
        this.stackDepthChanged = true;
    if (this.compositionFile == null)
        // Make sure the quad coordinates coorespond to the scale used by the UI plugin
        var maxPieceSize = 0;
        for (i in this.stackElements)
            maxPieceSize = Math.max( Math.max( this.stackElements[i].fWidth, this.stackElements[i].fHeight ), maxPieceSize );
        var mipLevel = 0;
        while (maxPieceSize >> mipLevel > kMaxPieceSize)
            mipLevel++;
        var imageReduction = 1.0 / (1 << mipLevel);
        this.offsetStack( -this.fBounds.getCenter() );
        this.scaleStack( imageReduction );
        this.offsetStack( -this.fBounds.getTopLeft() );
        this.setupConnectivity();
        // Add the additional per-piece metadata to pass to the filter plugin
        for (i in this.stackElements)
            this.stackElements[i].addPieceData();
    // Make the result layer active
    app.activeDocument.activeLayer = app.activeDocument.layers[app.activeDocument.layers.length -1];
    // Note: we need an "unmodified" flag, so if no
    // changes are made we skip the data recovery step...
    var result, err;
    try {
        result = this.invokeFilterPlugin( "AdobePhotomergeUI4SCRIPT", DialogModes.ALL );
    catch (err)
        result = null;
    if (result == null)        // Cancelled / bombed out
        stackDoc.close(SaveOptions.DONOTSAVECHANGES);
        return null;
    // Extract the data from the plugin
    var modifiedPieceInfo = result.getString( app.charIDToTypeID('PSpc') ).split('\n');
    for (i in modifiedPieceInfo)
        // If we loaded a composition (.pmg) file, we won't have corners yet.
        if (typeof(this.stackElements[i].fCorners) == "undefined")
            this.stackElements[i].fCorners = new Array();
        var j, pieceData = modifiedPieceInfo[i].split('\t');
        for (j in pieceData)
            var k, pair = pieceData[j].split('=');
            if (pair[0] == 'fUsed')
                this.stackElements[i].fUsed = eval(pair[1]);
            if (pair[0] == 'fCorners')
                var coords = pair[1].split(/\s+/).slice(1);
                for (k = 0; k < 4; k++)
                    this.stackElements[i].fCorners[k] = new TPoint( Number(coords[k*2]), Number(coords[k*2+1]) );
    // Remove unused photos
    for (i = 0; i < this.stackElements.length; ++i)
        if (! this.stackElements[i].fUsed)
            stackDoc.layers[(this.stackElements.length-1)-i].remove();
            this.stackElements.splice(i,1);
            i--;
    // Hey...it could happen.
    if (this.stackElements.length < 2)
        return null;
    // If we run the UI, we're restricted to an eight bit stack.
    // If the source images were higher, we need to reload the image stack.
    if (this.stackDepthChanged)
        stackDoc.close(SaveOptions.DONOTSAVECHANGES);
        stackDoc = this.loadStackLayers();
    this.getBounds();        // Update w/new corner data
    return stackDoc;
// "Adobe patent application tracking # B349, entitled 'Method and apparatus for Layer-based Panorama Adjustment and Editing', inventor: John Peterson"
// Use the geometry of the overlapping pieces to create
// simple rectangular blend masks.
photomerge.quickBlend = function()
    var i, j;
    var blendRadius = Math.round(Math.min(this.stackElements[0].fWidth, this.stackElements[0].fHeight) / 10.0);
    if (PMDebug)
        for (i in this.stackElements)
            this.stackElements[i].dumpMLCorners();
    // Set up progress bar for blending
    // The progress bar doesn't work - there's know way in ScriptUI to force it to update.
/*    var progressWindow = latteUI( g_StackScriptFolderPath + "PMBlendingProgress.exv" );
    var num = this.stackElements.length;
    var progressBar = progressWindow.findControl('_progress');
    progressBar.maxvalue = (num * (num + 1)) / 2;
    num = 0;
    progressWindow.center();
    progressWindow.show();
    // Blend the i'th piece against the 0..i-1 pieces below it
    for (i = this.stackElements.length-1; i > 0; --i)
        for (j = i-1; j >= 0; j--)
//            num++;
//            progressBar.value = num; // ScriptUI bug - there's no way to force this to update.
            this.stackElements[i].softEdgeBlend( this.stackElements[j], blendRadius );
//    progressWindow.close();
// Wrap the advancedBlend in a try/catch so errors (i.e., user cancel)
// just stop the blend process.
photomerge.advancedBlend = function( stackDoc )
    try {
        selectAllLayers(stackDoc, 1);
        advancedMergeLayers();
    catch (err)
// With the stack elements specified, this
// portion actually creates the Panorama.
// Returns boolean indicating success/failure
photomerge.doPanorama = function()
    var i, stackDoc = null;
    function primaryScreenDimensions()
        var i;
        for (i in $.screens)
            if ($.screens[i].primary)
                return new TPoint( $.screens[i].right - $.screens[i].left,
                                      $.screens[i].bottom - $.screens[i].top );
    function resizeCanvasToFitPano()
        // Extend the canvas to hold the panorama
        var w = UnitValue( photomerge.getBounds().getWidth(), "px" );
        var h = UnitValue( photomerge.getBounds().getHeight(), "px" );
        app.activeDocument.resizeCanvas( w, h, AnchorPosition.TOPLEFT );
    if (this.interactiveFlag)
        // Filter must have eight bit depth and RGB color space
        stackDoc = this.loadStackLayers( BitsPerChannelType.EIGHT );
        stackDoc.changeMode(ChangeMode.RGB);
    else
        stackDoc = this.loadStackLayers();
    if (! stackDoc)
        return false;
    // Remove spurious last layer (not needed by Photomerge)
    if (app.activeDocument.layers[app.activeDocument.layers.length-1].name == this.pluginName)
        app.activeDocument.layers[app.activeDocument.layers.length-1].remove();
    // The UI needs everything in the top left corner
    if (this.interactiveFlag)
        for (i = 0; i < stackDoc.layers.length; ++i)
            var xoff = stackDoc.layers[i].bounds[0].as("px");
            var yoff = stackDoc.layers[i].bounds[1].as("px");
            if ((xoff != 0) || (yoff != 0))
                stackDoc.layers[i].translate( UnitValue( -xoff, "px" ), UnitValue( -yoff, "px" ) );
    // Sort out exactly what operations we want to do.
    if (! this.interactiveFlag)
        selectAllLayers(stackDoc, 1);
        this.getAlignmentInfo( stackDoc, true );
        resizeCanvasToFitPano();
        if (this.advancedBlending)
            stackDoc.changeMode( ChangeMode.RGB );    // Auto-blend requires RGB
            this.advancedBlend( stackDoc );
// The Advanced blending works so well that there's little point
// in having the rectangular gradient blends anymore. Uncomment the
// following two lines if you still want them (see similar code below
// for the interactive case).
//        else
//            this.quickBlend();
        purgeHistoryStates();
        return true;
    // Interactive happens here
    if (this.compositionFile == null)
        this.getAlignmentInfo( stackDoc, false );
        // With the corners computed by getAlignmentInfo,
        // find the bounds and use that to slide the images
        // over so their bounds has origin 0,0 (top left)
        this.offsetStack( -this.fBounds.getTopLeft() );
    if (this.interactiveFlag)
        stackDoc = this.callInteractivePlugin( stackDoc );
    if (stackDoc == null)
        return false;
    resizeCanvasToFitPano();
    // Now apply the transformation to the pieces
    for (i in this.stackElements)
        this.stackElements[i].transform();
    if (this.advancedBlending)
        selectAllLayers( stackDoc, 1 );
        stackDoc.changeMode( ChangeMode.RGB );    // Auto-blend requires RGB
        advancedMergeLayers();
// The new "advanced blending" works well enough that there's
// little point in invoking the rectangular gradient blends anymore.
// If you really want them, you can uncomment the two lines below.
//    else
//        this.quickBlend();
    purgeHistoryStates();
    return true;
// Extra group breaks the main dialog's radio buttons,
// so we manually implement it here (ScriptUI lossage)
// NOTE: When called, this is a member function of radioControl,
// -not- photomerge. JavaScript voodoo.
photomerge.radioClick = function()
    var w = this.window;
    var i;
    // Some of the transforms don't allow lens correction...
    var allowLensCor = ((this.button_id != '_LOnormal')
                        && (this.button_id != '_LOcollage')
                        && (this.button_id != '_LOinteractive'));
    w.findControl('_useLensCorrection').enabled = allowLensCor;
    w.findControl('_removeVignette').enabled = allowLensCor;
    // Be aggressive about it...
    if (! allowLensCor)
        w.findControl('_useLensCorrection').value = false;
        w.findControl('_removeVignette').value = false;
    for (i in photomerge.radioButtons)
        var b = w.findControl(photomerge.radioButtons[i]);
        if (b != this)
            b.value = false;
// Callback when "Load Composition" is clicked.
// NOTE: When called, this is a member function of buttonControl,
// -not- photomerge. JavaScript voodoo.
photomerge.loadCompositionClick = function()
    function MacPMGType( f )
        if (f.type == 'PhMg')
            return true;
        var match = f.name.match(/.*[.](.+)$/);
        var suffix = match != null ? match[1].toLowerCase() : "";
        if (suffix == "pmg")
            return true;
        if (f instanceof Folder)
            // If the item is an app or a bundle it will be an
            // instance of folder. If we return true it will
            // appear as an enabled item. While the OS will not
            // let the user navigate into it, it is better to
            // have it appear as disabled.
            if (suffix.match(/app|bundle/i) != null)
                // Do not navigate folders that end in .app or .bundle
                return false;
            // navigate any other folder
            return true;
        // some unknown file type/suffix
        return false;
    var fileType = File.fs == "Windows" ? localize("$$$/AdobePlugin/Shared/Photomerge/Auto/Win=Photomerge Compositions:*.pmg")
                                    : MacPMGType;
    photomerge.compositionFile = File.openDialog( localize("$$$/AdobePlugin/Photomerge/LoadComp=Load Photomerge Composition"), fileType );
    if (photomerge.compositionFile == null || !photomerge.compositionFile.open("r"))
        return;
    var line = photomerge.compositionFile.readln();
    if (! line.match(/^VIS/))
        alert( this.pluginName + localize("$$$/AdobePlugin/Photomerge/BadComp=The Composition file is corrupt"), this.pluginName, true );
        return;
    // Read through the composition file and extract the file paths in it.
    var mergeFiles = new Array();
    while (! photomerge.compositionFile.eof)
        line = photomerge.compositionFile.readln();
        var f = line.match(/^\s*PATH\s+<([^>]+)/);
        if (f)
            // If no file path delimiters, image paths are assumed relative to the composition file
            var relPath = (f[1].indexOf( (File.fs == "Windows") ? "\\" : "/" ) < 0);
            f = (relPath ? photomerge.compositionFile.path + "/" : "") + f[1];
            mergeFiles.push( new StackElement( new File(f) ) );
    if (mergeFiles.length < 2)
        alert( this.pluginName + localize("$$$/AdobePlugin/Photomerge/BadComp=The Composition file is corrupt"), this.pluginName, true );
        return;
    photomerge.stackElements = mergeFiles;
    photomerge.interactiveFlag = true;
    this.window.close(kFilesFromPMLoad);
// Set up the radio buttons
photomerge.customDialogSetup = function( w )
    var i, button;
    for (i in this.radioButtons)
        button = w.findControl(this.radioButtons[i]);
        button.onClick = this.radioClick;
        // Flag with name so we can identify it in radioClick
        button.button_id = this.radioButtons[i];
    // Missing feature: We should query the selected file's metadata and
    // automatically turn on the '_useLensCorrection' checkbox if the
    // file has the proper support for it.
    w.findControl("_loadcomp").onClick = this.loadCompositionClick;
    // Julie didn't like the intro line; nuke it here because stackDialog looks for it.
    w.findControl("_intro").parent.remove(['_intro']);
    var size = w.findControl("_fileList").size;
    size[1] += 20;
    w.findControl("_LOauto").value = true;        // Set default
    w.findControl("_advancedBlend").value = this.advancedBlending;
    // If the PhotomergeUI or ADM plugins aren't there, don't display the option for it.
    if ((app.systemInformation.search(/PhotomergeUI/) < 0) || (app.systemInformation.search(/ADM /) < 0))
       w.findControl("_PMInteractive").hide();
       w.findControl("_loadcomp").hide();
// Called by the dialog on closing to collect the results
photomerge.customDialogFunction = function( w )
    if (w.findControl("_LOinteractive").value)
        this.interactiveFlag = true;
        this.alignmentKey = 'interactive';
    else
        var i, d = [{k:"_LOauto",v:"Auto"},{k:"_LOnormal",v:"translation"},{k:"_LOperspective",v:"Prsp"},{k: "_LOcylindrical",v:"cylindrical"},{k:"_LOspherical",v:"spherical"},{k:"_LOcollage",v:"scen eCollage"}];
        for (i in d)
            if (w.findControl(d[i].k).value)
                this.alignmentKey = d[i].v;
                break;
    this.advancedBlending = w.findControl("_advancedBlend").value;
    this.lensCorrection = w.findControl("_useLensCorrection").value;
    this.removeVignette = w.findControl("_removeVignette").value;
// "Main" execution of Photomerge from the menu
photomerge.doInteractivePano = function ()
    // Because of the ",true", the dialog is pre-loaded with any bridge files.
    this.getFilesFromBridgeOrDialog( localize("$$$/Private/Photomerge/PMDialogexv=PMDialog.exv"), true );
    try {
        if (this.stackElements && this.doPanorama())
            fitViewOnScreen();
            var flagDesc = new ActionDescriptor();
            flagDesc.putBoolean( kPhotomergeAdvancedBlendingFlag, photomerge.advancedBlending );
            app.putCustomOptions( "PhotomergeFlags001", flagDesc, true );
    catch (err)
        if (this.stackDoc)
            this.stackDoc.close(SaveOptions.DONOTSAVECHANGES)
// Use this version to call Photomerge from a script.
photomerge.createPanorama = function(filelist, displayDialog)
    this.interactiveFlag = (typeof(displayDialog) == 'boolean') ? displayDialog : false;
    if (filelist.length < 2)
        alert(localize("$$$/AdobePlugin/Shared/Photomerge/AtLeast2=Photomerge needs at least two files selected."), this.pluginName, true );
        return;
    var j;
    this.stackElements = new Array();
    for (j in filelist)
        var f = filelist[j];
        this.stackElements.push( new StackElement( (typeof(f) == 'string') ? File(f) : f ) );
    if (this.stackElements.length > 1)
        this.doPanorama();
if ((typeof(runphotomergeFromScript) == 'undefined')
    || (runphotomergeFromScript==false))
    photomerge.doInteractivePano();

Image Processor error message

Issue:
On Mac 10.9.2, running Adobe CC, I attempt to convert a selection of PSD files using Image Processor through Bridge. I get an error stating "Sorry, I could not process the following files".
Image Processor supports conversion from PSD files as far as I know. Why is this not working?
I see on the forums that this error message sometimes comes up when you are using the "Run Action" function, but I am not using this.

After a stunningly unhelpful chat session with Adobe Support, it seems that the issue stems from the fact that the PSD files were originally EPS.
You see, I actually need to end up with TIFF files, but Image Processor doesn't accept EPS.
So I thought I would have to manually go through and save each EPS file as a TIFF in Photoshop--but it doesn't even give me that option! Did some testing and this seems to be because the EPS files are in Duotone mode.
I can only choose from Photoshop, Large Document Format, Photoshop EPS, Photoshop PDF, and Photoshop Raw.
As a workaround, I decided to manually convert to PSD, thinking that then I could run them through Image Processor and get my TIFFs.
But now Image Processor is rejecting the PSD files.
What to do?

Photoshop Scripts/Image Processor / Action Error

CS3:
My image processor has been working just fine with this action.. but for some reason out of the blue it says "sorry I could not process files:" and lists all the files that was processed. now its funny because it works just fine, the action is applied and it saves just fine but it doesn’t close the current window any longer and it gives that error.
the action is pretty complex but it only copies and rename layers and applies one filter.
again it worked just fine not too long ago.. I'm not sure what happened...
here is a copy of the action ... it uses portraiture plug-in filter in the beginning...
http://www.fileden.com/files/2006/8/29/194220/File%20start%20action.atn
any help would be great!
Thanks!!

Contact Adobe support, you hit a bug. I never seen that one pop up before.
But in general what it is saying is it is trying to create a window and to do so it needs to reference a constructor.
That constructor should have been automatically created or a programmer creates one from scratch, that depends on how the code was created. Since that bug normally does not appear, I would say a some code pointed to the wrong location in memory hence the word reference. (In generic terms reference means to point to a location in memory)
Try it again and see if it repeats itself. I think Adobe will tell you that you need to reinstall photoshop, but its better that they tell you that as I could easily be misinterpreting that error message.

Image Processor Pro v3.0 beta available

The Beta for the next rev of Image Processor Pro is now available and can be found here:
https://sourceforge.net/projects/ps-scripts/files/Image%20Processor%20Pro/Betas/.
For questions, support, etc... contact me directly ([email protected]) or here:
http://www.ps-scripts.com/bb/viewtopic.php?f=17&t=5845&sid=12f1f22f7f9fbc5669a27f432b56130 dwww.ps-scripts.com/bb/viewtopic.php?f=17&t=5845
The current Beta rev is 1_107.
CHANGES
- A number of bugs were fixed, especially in the BMP and TGA save
panels.
- A new check box, 'Keep Original File', was added to Step 2 in the
UI. It is accessible when 'Save in Same Location' is checked and Save
in Subfolder' is unchecked. In this situation, if you uncheck 'Keep
Original File' the script will overwrite the original files.
- A new check box, 'Preserve filename in XMP Metadata' has been added
to Step 4 in the UI. If this is checked, the original name of the file
is saved with the property name 'PreservedFileName' in the XMP Media
Management Properties section of the new files XMP Metadata.
- A new flag ImageProcessorOptions.OVERWRITE_TARGET_FILES has been added
to the script. When set to true IPP will always overwrite target files
unless 'Save In Same Location' and 'Keep Original File' are checked.
The default is false. You must edit the file with a text editor to
change this flag. I may eventually work this into the UI.

I tries from Photoshop and had script errors for users options and IPP log. Will try from bridge to see if the will create them. View image in new tab so you can read them.
Just tried from bridge fails there too.
renamed the old IPP files in user AppData\Roaming and the script now works...

Resizing large quantites of images using image processor - STUCK

I have upgraded to Photoshop CC. I am attempting to resize large quantites of images using the image processor. I need both the width and height set to an exact number for web use. When I enter the pixel size into the "resize" boxes and click run, the sizes i get are correct on one dimension but slightly off on the other dimension. When I just do one image at a time to these same dimensions it resizes correctly? It appears it only does when selecting large qunatites. Any suggestions?die

Since you do not want to crop your images to a square 1:1 aspect ratio changing the canvas to be square will not make your images square they will retain their Aspect Ratio and image size will be changer to fit within your 1020 px square. There will be a border or borders on a side or two borders on opposite sides. You do not need a script because Photoshop ships with a Plug-in script to be used in Actions. What is good about Plugins is the support Actions. When you record the action the plug-in during action recording records the setting you use in its dialog into the actions step. When the Action is played the Plug-in use the recorded setting an bypasses displaying its dialog. So the Action can be Batch. The Action you would record would have two Steps. Step 1 menu File>Automate>Fit Image... in the Fit Image dialog enter 1020 in the width and height fields. Step 2 Canvas size enter 1020 pixels in width and height not relative leave the anchor point centered it you want even borders on two sides set color to white in the canvas size dialog. You can batch the action.
The above script will also work. Its squares the document then re-sizes to 1020x1020 the action re-sizes the image to fit with in an area 1020 x 1020 then add any missing canvas. The script like the action only process one image so it would also need to be batched. Record the script into and action and batch the action. As the author wrote. The script re size canvas did not specify an anchor point so the default center anchor point is uses like the action canvas will be added to two sides.

Image processor cs 4 millimeter support

Similar Messages

Maybe you are looking for