Thank you both for suggestions! I've tried Trevor's suggestion adding $.sleep(5000) but then the script just idles on that jsx line of code. Javascript is running on only one thread in the normal case, so I guess that makes sense. (Even though I thought the "new" CEP5/HTML5 UI approach was to be more loosely coupled than the older script interface approaches.) The HTML window can't be moved while the script is running etc... still no way to make a cancel button, and the the graphical interface is not updating. CEP/HTML5 doesn't seem quite ready. ScriptUI works better in CS6 than CEP/HTML5 does in CC. There is need for a "doevents" command or "update UI" command. Something that you can call for the user to interact with the interface, and to make the interface redraw. Back to the "Sleep(5000) test": After waiting for the script to produce 8 items*5 (seconds of sleep), in my InDesign document, the div tag with its progress bar is shown in the html interface. Doing one or a few items at a time with repeated calls from the js part would perhaps make a difference. Not sure. But perhaps it's the jsx part that is locking up everything in the UI, despite the fact that events from the jsx are handled in the js. Then the problem would not be altogether a "single thread" issue(?). Anyway, we have all database interaction in the jsx part, making it hard to break the loop up that is going on, and moving it to the js part. So I haven't tested this suggestion. Changing "sleep" into "alert" made the html interface redraw, so Loic's is right that there needs to be something that really requires a redraw for the interface to update. The ScriptUI progressbar however always showed up fine for me, and updated step by step in CC and earlier versions. I had no need to create a new window for every step in ScriptUI, as you write about Loic. "So why not use the ScriptUI progressbar – if you think it's working", someone might ask. Well, Adobe not being able to fix the ScriptUI treeview bug we rewrote our user interface in HTML5, and so I thought we shouldn't keep the old ScriptUI progressbar either – blending different techniques – but replace it with an HTML version. It's also no good practice to let anything show from the jsx part, where ScriptUI is situated. Showing interface stuff is the responsibility of the HTML/js part of the code. The enableRedraw setting is never changed by my Script. Nothing is seen on the document in InDesign CC until the script is completely done, despite the fact that Enable Redraw is set to True in the Scripts panel menu all the time. There is certainly plenty of room for improvements in the User Interface part of CEP/HTML5 extensions. Trying to find workarounds for things "not thought of" is not funny... and I had hopes that the change of technology would mean some improvements, a more loosely coupled interface and better UI redraw was one expectation that was not met. ... View more

If you use the command line tool ZXPSignCmd, then for CC 2015 you need to get the latest version of ZXPSignCmd.exe (from summer 2015), otherwise you will get the error: Error - the timestamp returned from the chosen TSA could not be verified, so the ZXP created is likely to be rejected by other tools. Please recreate your ZXP with a different trusted TSA. It doesn't matter what timestamp server you use. The updated version of ZXPSignCmd.exe can be found on this address: https://github.com/Adobe-CEP/CEP-Resources/tree/master/ZXPSignCMD ... View more

Hi Bob! For this question I'm not interested in portability. It's a Windows only environment, with MIF files being produced by our windows software, and the users are working in FrameMaker on the same Windows computers that they use to export the MIF files. You wrote that I should "Stay relative path"... We've been using absolute paths to images for many years. In the MIF file we use ImportObFile on its own (no ImportObFileDI) when referencing the images. All the images at the customer are on a server, with a mapped drive letter – some years they've had the images locally as well, to try and get faster document response / loading times, but the path is still absolute, since the MIF file is written by our program code. If I remember correctly, local vs network placed images were not a big difference. Would changing to relative paths make FrameMaker respond quicker or work faster, when loading the document ("turning" pages in it) and other things that involve FrameMaker processing the pages (and images), such as print to pdf? The original question concerns the strange thing that the only thing that seems to create a valid image reference, in MIF, to a Windows UNC path, is by adding the ImportObFileDI command with an Apollo only "h" code. The documentation might just be wrong here, and missing something about UNC paths being solved in that way... Thanks, Andreas Jansson ... View more

Yes, I've tried that.  it doesn't matter how many slashes I add (I tried to duplicate them). ... View more

Hi! I just had a silimar case. The user had just got a new computer with Windows 8.1, with the same InDesign version installed as on his old one (InDesign CS5). When running our javascript (i.e. "extendscript") that use VBScript to get the user name in Windows and an applescript to get the user from OSX, he got this error on the new computer: Javascript Error! Error Number: 104705 Error String: Unable to load Adobe InDesign CS5 Type Library (Version 1.0). The script code line that raised this error was the following: gUserName = app.doScript(getUserCodeStringVB, ScriptLanguage.visualBasic); The solution was to start InDesign CS5 as Administrator (right click the shortcut to InDesign in the start menu and select "Run as administrator"). Obviously that made the keys go to the right places in the registry. Now it works when the user runs InDesign as himself as well. There are similar errors with Photoshop installations, and I had read that starting Photoshop as Administrator (once) might help. That's why I asked the user to try this. And it worked! Perhaps this can be helpful for other people with the same error. // Andreas ... View more

Thanks for the file and the explanation, now "Adobe Photoshop CC 2015 Object Library" is present in the object model viewer dropdown list. Great! On windows, you can write like this to accommodate for any variants on the "Program files" folder name: %ProgramFiles%\Common Files\Adobe\Scripting Dictionaries CC\Photoshop 2015 ... View more

I did the following (again): Uninstalled Photoshop CC2015 Uninstalled Photoshop CS4 Installed Photoshop CC2015 Now the registry seems to be more in order. Different versions of Photoshop obviously do not go well together. Still no Photoshop CC2015 in the object model viewer though. For more information about the COM exceptions I mentioned above see my last input in: COMException and Photoshop classes missing in Object Model Viewer after installing Photoshop CC 2015 (The main problem seems to be that a call to "Save As" sometimes opens up a Save As dialogue box in Photoshop CC2015.) ... View more

It's no longer working on my computer either. Now it says: Retrieving the COM class factory for component with CLSID {6568E194-96D8-4035-9087-99E4F75E9ABC} failed due to the following error: 80080005 Server execution failed (Exception from HRESULT: 0x80080005 (CO_E_SERVER_EXEC_FAILURE)). ... when trying to get a reference to Photoshop from C# The entry in the OMV for CC2014 is still there. I get the error above when having Photoshop CC 2015 already open. If it's closed then Photoshop CS4 opens up, and runs the code up until it's preparing for a "save for web" call using the following code: dynamic options = Activator.CreateInstance(Type.GetTypeFromProgID("Photoshop.ExportOptionsSaveForWeb")); The error I get from the line above (CS4) is : A first chance exception of type 'System.Runtime.InteropServices.COMException' occurred in mscorlib.dll Additional information: Retrieving the COM class factory for component with CLSID {C19EF106-2B05-41ED-A28D-EB9BBD7C936B} failed due to the following error: 80080005 Server execution failed (Exception from HRESULT: 0x80080005 (CO_E_SERVER_EXEC_FAILURE)). Perhaps I need to reinstall Photoshop CC2015 again? The customer has still problems. They have not been able to run our image conversion and registration program since they upgraded. Since the COM exceptions above are not strictly scripting errors, I put the question up here as well: COMException and Photoshop classes missing in Object Model Viewer after installing Photoshop CC 2015 I've got some people indicating that they've got the same problem, but no responses. Please go there and vote for it if you experience similar errors in CC2015. But I guess that the missing entries in the OMV indicate that something vital is missing now... and that neither the new version of Photoshop is working nor the old ones. What should I do? ... View more

The user in question is admin on her computer, and the IT department on her company has tested on other computers as well. They get the same error anyway. ... View more

This solution (uninstall / install) did not work for the users of the aforementioned photoshop automation C# program. ... View more

Uninstalled PhotoShop 2015. Tried to run my Photoshop "automation" program with an older version of Photoshop still on the computer, similar errors: Like this (perhaps this one was from 2015, not quite certain). Error processing 102178.eps: System.Runtime.InteropServices.COMException (0x80040154): Retrieving the COM class factory for component with CLSID {14300B0E-D530-4964-885F-5C1C4C760EBE} failed due to the following error: 80040154 Class not registered (Exception from HRESULT: 0x80040154 (REGDB_E_CLASSNOTREG)).    at System.RuntimeTypeHandle.CreateInstance(RuntimeType type, Boolean publicOnly, Boolean noCheck, Boolean& canBeCached, RuntimeMethodHandleInternal& ctor, Boolean& bNeedSecurityCheck) or this (with Photoshop CS4, having removed Photoshop CC2015): The registered components seem to become a mess after installing Photoshop CC2015, and it also seems to destroy (unregister?) older versions of PhotoShop components, since the components could not be created even after Photoshop 2015 was uninstalled. Solution (at least it seems to work now) I reinstalled Photoshop 2015. My automation program works with Photoshop CC2015. The Object Model Viewer even has a Photoshop entry (but it says "CC2014"... why not 2015? InDesign 2015 is in the list). ... View more

I did that now. I chose to remove preferences as well (a question you get when uninstalling). The effect was that more entries in the list disappeared. Compare the scrollbar in my initial question, with the one here, after reinstall: ... View more

Photoshop object model disappeared from the Object Model Viewer ( OMV) for me, when installing Photoshop 2015. Is that the case for other scripters? Please open ExtendScript ToolKit CC and check for Photoshop in the list. Also see: Photoshop missing in Object Model Viewer (ExtendScript ToolKit CC) after installing 2015 CC for a screen shot. ... View more

Great! I managed to recompile the old c++ dll with a simplified version of your example (added below). To make it compatible with the "simple save" function that I already had, I changed the name and parameters (names and types) to the same as the old function had. The only thing left is that fdk.lib now references icuuc32.dll, icuin32.dll and netlib.dll, which exist in the FrameMaker folder (in program files / Program files(x86) or similar). I guess I will have to write some code to try and add the framemaker folder to the PATH environment variable, for the clients that are going to run this. int __stdcall FMSaveDoc(int nDocID, LPCSTR lpszSaveAsName, BOOL bInteractive) {     F_PropValsT saveScript, *returnp = NULL;     UIntT sn;     saveScript = F_ApiGetSaveDefaultParams();     sn = F_ApiGetPropIndex(&saveScript, FS_AlertUserAboutFailure);     if (bInteractive)         saveScript.val[sn].propVal.u.ival = False;     else         saveScript.val[sn].propVal.u.ival = True;     sn = F_ApiGetPropIndex(&saveScript, FS_AutoBackupOnSave);     saveScript.val[sn].propVal.u.ival = FV_SaveNoAutoBackup;     // File type  - Binary format of current InDesign version.     sn = F_ApiGetPropIndex(&saveScript, FS_FileType);     saveScript.val[sn].propVal.u.ival = FV_SaveFmtBinary;         sn = F_ApiGetPropIndex(&saveScript, FS_RetainNameStripe);     saveScript.val[sn].propVal.u.ival = False;         sn = F_ApiGetPropIndex(&saveScript, FS_SaveMode);     saveScript.val[sn].propVal.u.ival = FV_ModeSaveAs;     sn = F_ApiGetPropIndex(&saveScript, FS_SaveAsModeName);     saveScript.val[sn].propVal.u.ival = FV_SaveAsNameProvided;         // Save the document      nDocID = F_ApiSave(nDocID, (StringT)lpszSaveAsName, &saveScript, &returnp);     // We are done with these structures, so deallocate      F_ApiDeallocatePropVals(&saveScript);     F_ApiDeallocatePropVals(returnp);     // If something went wrong, ensure the return value is zero      if (!nDocID) nDocID = 0;     return nDocID; } ... View more

Thank you Russ! I just discovered that the .fm files produced by F_ApiSimpleSave() when running FM12 are in fact MIF files, similar to the original MIF files generated by our application. It seems F_ApiSimpleSave() does not save the file as binary fm any longer. Perhaps that is causing the "View only" as well(?). F_ApiSimpleSave() did work fine for at least 13 years. I believe it's just called upon to save the mif-files as fm... Could it be a bug, that Adobe would want to fix? Otherwise we will have to look into F_ApiSave. ... View more

Okay, thanks. I've repeated the question in the FrameMaker scripting forum: Breaking change in F_ApiSimpleSave saving documents in FrameMaker12 ... View more

Hi Jeff (and anyone else who might have an idea of what to do)! I'm trying to help Monica with this question as well. To begin with, this question concerns FrameMaker12. You were correct that the files of the book were locked (View only). Some more background: The document files are originally created as MIF, from an external application. That application then makes an F_ApiSimpleSave call to FrameMaker (http://www.adobe.com/content/dam/Adobe/en/devnet/framemaker/fdk/10/fdkprogrammerguide.pdf‌), thus producing the .fm files for the book. In version 12 of FrameMaker the F_ApiSimpleSave call obviously results in "View only" files, which in our case is not good at all. The call to produce fm files from MIF has been unchanged for many years and versions of FrameMaker. Why does the F_ApiSimpleSave call to FrameMaker produce these read only (locked or View only) files, and more important: what can be done about it? Thanks, Andreas J ... View more

Since I had a similar question I'll add some words here. I had no padlock icon and there are no toolbars at all when I open a locked dockument. If I close the document I can make toolbars visible in the View menu (View / Toolbars / Quick Access Bar). This will remain open when I again open the document. The help file says: View Only 12.0 Produces a book or document that you can open but cannot edit. If you accidentally save a file in View Only format, you can make it editable by pressing Esc Shift+F l (lowercase L) k. Then you can save the document or book in Document or Book format. This works, but be sure to do it exactly as it says. In other words: Press the Esc key (release the Esc key) Press the F key (UPPERCASE F, and release it) Press the l key (that is a lower case L. Release the l key.) Press the k key (release the k key) Also, make sure to click in the document before unlocking. It seems unlock has no effect when the wrong parts of FrameMaker has focus. Now the menus are "full", and the document can be edited. ... View more

What operating system? Win64, Win32 or Mac? I've not seen this yet, using Windows 32 bit version and Adobe Extendscript Toolkit CC 4.0.0.1. ... View more

We've got the same problem as you, most customers using CS4 to CS6. I hope to be able to use the same main code for both user interfaces. I've kept the ScriptUI inside the jsx file, while adding the new CEP/HTML in a separate file, and adapted functions in the jsx to manage requests from the new HTML and JS files. I'm not done yet, but I hope to be able to start the old interface with the jsx only, but to begin with I have just focused on getting the HTML version to work. There are a few bugs in ScriptUI for CC, so it's strange that Adobe doesn't fix them. Ever from the treeview bug was detected and opened in June 2013 until a year later they "escalated" the bug and told me that "hopefully we will get a fix soon", and "Again I’m very sorry for this, a bug should not be open so long time with us". But last summer (2014) the responsible people  began to suggest that I built a new user interface, and a couple of months ago I gave up on waiting for a bug, since there are a couple of customers who want to use CC. It's hard to make an interface that looks quite as integrated as with ScriptUI when using HTML. Below the kind but somewhat unsatisfactory answer from Adobe, June 2014: "[...] I’m afraid that there’s not a fix for this yet, however I can with you some more detail on the scope of the problem. The change in behaviour was a side effect of some of the changes to the UI architecture for the CC release. It appears that AfterEffects also suffers from this issue and the engineering teams are reviewing if this can be fixed. I’m sorry that we can’t commit any timelines for a fix, however, in the meantime there is the alternative of using Extension Builder to create the UIs for your scripts. [...]" It would still be interesting to find out what the review from the "engineering teams" found out, and their conclusion. ... View more

After waiting a long time for Adobe to fix this, asking them about progress on this bug every other month or so, I was eventually recommended to rewrite the whole user interface as HTML/CEP. ... View more

It was in the HTML user interface that I wanted access to the layer name, and its only communication with InDesign is via function calls (asynchronous) and events. But I did make a call from the HTML/js interface to get the active layername from InDesign, and with that I could go on making another call to a database that needed the layer name. So it's working ok. ... View more

Thanks, I will probably not use events for this when there is no actual layer changed event. At least I'll try another way first – it would just have been convenient to always have access to the correct layer name. And by the way, ignore the bold lines in my message above, there are lots of classes with events presented on them, that do not have a class XxxxEvent. ... View more