To create a (navigable) Epub TOC you check the 'Make text anchor in source paragraph' checkbox. The links are created correctly when the TOC is created manually, but when the TOC is created by a script, the links are not created.

To confirm, please open the attached document and run this very simple script:

app.documents[0].createTOC (app.documents[0].tocStyles.item('Epub'), true);

The used TOC style ('Epub') has the 'Make text anchor in source paragraph' box checked.

Then export the document to reflowable Epub. In the Export dialog's General tab, select 'Multilevel (TOC Style)' at 'Navigation TOC' and at 'TOC Style, select the Epub style. Click OK.

Expected result: An error-free export and a valid Epub file.

Actual result: An error message saying "Found TOC hyperlinks which lack target text anchor information. Please update TOC with 'Make text anchor in source paragraph' option turned on."

To verify that this is a scripting problem, re-open the document and export it manually. The error does not occur, and the epub has a valid html TOC.

I'll attach a self-contained script which first creates in order to demonstrate problems with textStyleRange.convertToNote(). The script creates a small document containing "b"s. These are marked with a smaller pointSize so that they form individual textStyleRanges. The script then converts these textStyleRanges into notes. Unfortunately, the script terminates with a message: "Object is invalid; tsr.convertToNote()". This message happens because the scripting engine has become confused and is trying to convert a "b" within a note into another note. This shouldn't happen and is the first problem.

Open the story window. The beginning of the story has a clear pattern of notes between sentences. This should continue to the end of the document, but looking further down, two more problems can be seen:

(1) Overflow.jpg shows that the text at the frame's overflow boundary has become confused. It should say "The quick green fox ...", but instead it says something like "The The The quick The ...".

(2) Nearer to the end of the story, Confusion.jpg shows that larger phrases are converted into notes even though only "b" should be converted.

I can't attach a script, so here it is:

// Run this script in InDesign CC 2019, Roman Region.
//THIS IS THE PROBLEM PART THE REST IS JUST TO SET IT UP.

// Convert all "b" characters into notes
function hideNotes(story){

 // Find all the "b"s

 app.findTextPreferences = null;

 var findPrefs = app.findTextPreferences;

 findPrefs.findWhat = "b";

 var matchingItems = story.findText()


 // It shouldn't be necessary to copy the findText() results, but just in case ...

 var copy = []

 for (var i = 0; i < matchingItems.length; i++) { 

     copy[i] = (matchingItems[i])

 }

 // Loop through the "b"s and convert them to notes

 for (var i = 0; i < copy.length; i++) { 

     var tsr = copy[i].textStyleRanges.firstItem();

     tsr.convertToNote();

  }

}

// This is just setting up a document with reasonable defaults
function setup(){

// Set up application

var millimeterMeasurementUnits = 2053991795; // This means millimeters

app.viewPreferences.verticalMeasurementUnits = millimeterMeasurementUnits ;

app.viewPreferences.horizontalMeasurementUnits = millimeterMeasurementUnits ;


// Set up new document

var doc = app.documents.add();

doc.zeroPoint = [0, 0];

doc.viewPreferences.rulerOrigin = 1380143215 // This means PageOrigin;

doc.documentPreferences.pageWidth = 210

doc.documentPreferences.pageHeight= 297

doc.marginPreferences.left  = 20

doc.marginPreferences.right  = 20

doc.marginPreferences.top  = 20

doc.marginPreferences.bottom  = 20

// Set up new page

var page = doc.pages.firstItem();

page.marginPreferences.left  = 20

page.marginPreferences.right  = 20

page.marginPreferences.top  = 20

page.marginPreferences.bottom  = 20

// Set up new frame

var textFrame = doc.textFrames.add();

textFrame.geometricBounds = [20, 20, 277, 190];

// Add frame text

for(var i = 0; i < 250; i++){

    var insertionPoint = textFrame.insertionPoints.lastItem()

    insertionPoint.pointSize = 12

    insertionPoint.contents = "The quick green fox jumped over the lazy dog. "

    insertionPoint = textFrame.insertionPoints.lastItem()

    // This point size will set the "b" into its own textStyleRange

    insertionPoint.pointSize = 7

    insertionPoint.contents = "b"

}

}

//Entry Point
setup();
var stories = app.activeDocument.stories;
for (var i = 0; i < stories.length; i++) {

hideNotes(stories[i]);

}

Hi,
if I set value for snippetImportUsesOriginalLocation of documentPreferences to true, InDesign will place a snippet file with spread.place( snippetFile ) always at the current zero point [0,0].

Wheras I set the value to false the snippet will be placed at its original position.

It should be exactly the other way.

Currently value false will use the original position and value true will chose the current zero point as defined with app.documents[0].zeroPoint .

Tested with German InDesign versions:
CS6 8.1.0
CC 2014 10.2.
2020 v. 15.0.1.209
All on WIndows 10.

In the discussion linked to below Gerald Singelmann can see this also on a Mac OSX-version of InDesign:

https://www.hilfdirselbst.ch/gforum/gforum.cgi?post=573538#573538

Regards,
Uwe Laubender
( ACP )

Instructions:

1. Load Script4.txt into ExtendScript CC


2. Open ID CC 2019


3. Connect ExtendScript CC to ID CC 2019


4. Run the script


5. Observe that line 8 gives an "Object is invalid" error. That's because the nextItem() on line 6 didn't work.

To see this more clearly:

1. Load Script4.txt into ExtendScript CC


2. Open ID CC 2019


3. Connect ExtendScript CC to ID CC 2019


4. Comment line 6 and uncomment line 7.


5. Run the script


6. The script runs fine. It prints out "2", the name of the second page.

In other words, nextItem() is broken. The work around (in this case) is to use an index of "Page_263.index+1". I think, in other situations, the work around will require a loop.

Instructions:

1. Load Script.txt into ExtendScript CC


2. Open ID CC 2019


3. Connect ExtendScript CC to ID CC 2019


4. Run the script


5. Observe that the script gives an error on line 6 (Invalid object for this request.) This is because TextFrame281 and TextFrame284 are the same object. They should be different objects.

To see this more clearly:

1. Load script.txt into ExtendScript CC


2. Open ID CC 2019


3. Connect ExtendScript CC to ID CC 2019


4. Uncomment line 4 (var temp = TextFrame_281.id)


5. Run the script


6. The script runs fine. If you look in the document that has been created, the two text frames are in the upper right corner and they are linked to each other.

In other words, the value of TextFrame_281 changes, depending on whether it is accessed. My only guess is that this is some kind of lazy evaluation gone wrong.

Currently the only way to manipulate the undo queue is to run a script with doScript and even then it's very limiting. With my current event listeners, the user clicks around a few times and the undo queue fills up with 'Undo Set Scripting Tag' and they are unable to undo simple changes without holding down CTRL+Z for a few seconds. I am unable to use doScript because I need to return values to my interface. I would like to suggest an option to temporarily suspend undo tracking during certain script actions. Some of my actions do not change the document and only update the interface.