ExtendScript does not include a JSON object with the associated methods for parsing, namely JSON.parse()
and JSON.stringify()
. Nor does it provide any other builtin feature for parsing JSON.
Solution:
Consider utilizing a polyfill to provide JSON functionality such as JSON-js created by Douglas Crockford.
What you'll need to do:
Download the JavaScript file named json2.js from the Github repo and save it in the same location/folder as your .jsx
file.
Note You can just copy and paste the raw version of json2.js from the same Github repo to create the json2.js file manually if you prefer.
Then at the top of your current .jsx
file you'll need to #include
the json2.js
file by adding the following line of code:
#include "json2.js";
This is analogous to how you might utilize the import
statement to include a module in modern day JavaScript (ES6).
A pathname to the location of the json2.js
can be provided if you decide to save the file in a different location/folder than your .jsx
file.
By including json2.js
in your .jsx
file you'll now have working JSON methods; JSON.parse()
and JSON.stringify()
.
Example:
The following ExtendScript (.jsx
) is a working example that generates JSON to indicate all the links associated with the current InDesign document (.indd
).
example.jsx
#include "json2.js";
$.level=0;
var doc = app.activeDocument;
/**
* Loads the AdobeXMPScript library.
* @returns {Boolean} True if the library loaded successfully, otherwise false.
*/
function loadXMPLibrary() {
if (!ExternalObject.AdobeXMPScript) {
try {
ExternalObject.AdobeXMPScript = new ExternalObject('lib:AdobeXMPScript');
} catch (e) {
alert('Failed loading AdobeXMPScript library\n' + e.message, 'Error', true);
return false;
}
}
return true;
}
/**
* Obtains the values f XMP properties for `DocumentID` and `instanceID` in
* each linked file associated with an InDesign document (.indd). A returns the
* information formatted as JSON,
* @param {Object} doc - A reference to the .indd to check.
* @returns {String} - The information formatted as JSON.
*/
function getLinksInfoAsJson(doc) {
var linksInfObj = {};
linksInfObj['indd-name'] = doc.name;
linksInfObj.location = doc.filePath.fsName;
linksInfObj.links = [];
for (var i = 0, len = doc.links.length; i < len; i++) {
var linkFilepath = File(doc.links[i].filePath).fsName;
var linkFileName = doc.links[i].name;
var xmpFile = new XMPFile(linkFilepath, XMPConst.FILE_INDESIGN, XMPConst.OPEN_FOR_READ);
var allXMP = xmpFile.getXMP();
// Retrieve values from external links XMP.
var documentID = allXMP.getProperty(XMPConst.NS_XMP_MM, 'DocumentID', XMPConst.STRING);
var instanceID = allXMP.getProperty(XMPConst.NS_XMP_MM, 'InstanceID', XMPConst.STRING);
// Ensure we produce valid JSON...
// - When `instanceID` or `documentID` values equal `undefined` change to `null`.
// - When `instanceID` or `documentID` exist ensure it's a String.
instanceID = instanceID ? String(instanceID) : null;
documentID = documentID ? String(documentID) : null;
linksInfObj.links.push({
'name': linkFileName,
'path': linkFilepath,
'docId': documentID,
'insId': instanceID
});
}
return JSON.stringify(linksInfObj, null, 2);
}
if (loadXMPLibrary()) {
var linksJson = getLinksInfoAsJson(doc);
$.writeln(linksJson);
}
Output:
Running the script above will log JSON formatted something like the following example to your console:
{
"indd-name": "foobar.indd",
"location": "/path/to/the/document",
"links":[
{
"name": "one.psd",
"path": "/path/to/the/document/links/one.psd",
"docId": "5E3AE91C0E2AD0A57A0318E078A125D6",
"insId": "xmp.iid:0480117407206811AFFD9EEDCD311C32"
},
{
"name": "two.jpg",
"path": "/path/to/the/document/links/two.jpg",
"docId": "EDC4CCF902ED087F654B6AB54C57A833",
"insId": "xmp.iid:FE7F117407206811A61394AAF02B0DD6"
},
{
"name": "three.png",
"path": "/path/to/the/document/links/three.png",
"docId": null,
"insId": null
}
]
}
Sidenote: Modelling your JSON:
You'll have noticed that the JSON output (above) is structured differently to how you were attempting to structure it in your given example. The main difference is that you were using link filenames as property/key names, such as the following example:
Example of a problematic JSON structure
{
"one.psd": {
"docId": "5E3AE91C0E2AD0A57A0318E078A125D6",
"insId": "xmp.iid:0480117407206811AFFD9EEDCD311C32"
},
"two.jpg": {
"docId": "EDC4CCF902ED087F654B6AB54C57A833",
"insId": "xmp.iid:FE7F117407206811A61394AAF02B0DD6"
}
...
}
Producing JSON like this example isn't ideal because if you were to have two links, both with the same name, you would only ever report one of them. You cannot have two properties/keys that have the same name within an Object.
Edit:
As a response to the OP's comment:
Hi RobC, other than using #include 'json2.js'
, is there any other way to include external js file in the JSX file?
There are a couple of alternative ways as follows:
You could utilize $.evalFile()
. For instance replace #include "json2.js";
with the following two lines:
var json2 = File($.fileName).path + "/" + "json2.js";
$.evalFile(json2);
Note: This example assumes json2.js
resides in the same folder as your .jsx
Alternatively, if you're wanting to avoid the existence of the additional json2.js
file completely. You could add a IIFE (Immediately Invoked Function Expression) at the top of your .jsx
file. Then copy and paste the content of the json2.js file into it. For instance:
(function () {
// <-- Paste the content of `json2.js` here.
})();
Note: If code size is a concern then consider minifying the content of json2.js
before pasting it into the IIFE.