Adobe Community
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Exit
About Inventsable
since
Dec 18, 2014
Apr 16, 2024
Inventsable
Enthusiast
375
Posts
294
Upvotes
47
Solutions
Mar 27, 2022
11:33 PM
2 Upvotes
for (var i = 0; i < myArray.length; i++) {
alert(typeof myArray[i]); // "string"
}
try {
selection[0].position = ["44", "55"]; // This is what you're doing:
} catch (err) {
// Which results in an error because you can't apply strings to numeric values:
alert(err); // "Error: Point value expected"
} As Charu points out you'd have to convert them to typeof Numbers before being able to apply. Just as an extra trick, you can do this in all sorts of different ways in native JS: ~~"3.14" // 3
+"3.14" // 3.14 > This is what I normally always use, but it can hurt readability
"3.14"*1 // 3.14 > This can be the fastest, most performant way to convert from String > Number
Number("3.14") // 3.14 > This is the most reliable for largest integers
Math.round("3.14") // 3 > Math.floor, ceil, and round accept either string or number
... View more
Mar 26, 2022
02:19 AM
1 Upvote
I don't think my manual parser can handle 300k+ characters. Going forward, can you please put strings like that into a pastebin? Felt like I was click-dragging for a solid 3 minutes just trying to copy that data here on the forums. Try this instead. I've already tested the file you've given me and successfully written an image, though I haven't put it through any PNG validation to ensure it's been encoded correctly -- it writes to my disk and appears correct, though.
... View more
Mar 26, 2022
01:27 AM
2 Upvotes
No worries. Just an FYI, I'd added a few functions that I thought you might find useful, like reverse PNG > base64 conversion so you can handle the data both ways. Technically you can programmatically control this from a script or JSX, but it becomes a bit more complicated. You could use what's called an "Invisible Extension" (read here) to have a CEP panel that is headless or contains no UI and is automatically run when the application starts. If this panel contains code such as: // Must include CSInterface.js as a preloaded script in your HTML:
// https://github.com/Adobe-CEP/CEP-Resources/blob/master/CEP_11.x/CSInterface.js
// The below must be run after CSInterface.js is loaded, like in the custom logic / <body> tag
let CSI = new CSInterface();
CSI.addEventListener("myScriptMessage", (msg) => {
if (msg.data && msg.data !== "undefined") alert(msg.data);
}); Then we can actually have a script send it a message directly and invoke the above listener function at any point. If this msg.data were base64 text or a JSON.stringified object containing the base64 rawdata and a filepath to write to, etc, you could trigger the function fully remotely. In JSX/scripting you'd do this in order to trigger the above in the CEP panel: function JSXEvent(payload, eventType) {
try {
var xLib = new ExternalObject("lib:PlugPlugExternalObject");
} catch (e) {}
if (xLib) {
var eventObj = new CSXSEvent();
eventObj.type = eventType;
eventObj.data = payload;
eventObj.dispatch();
}
return;
}
// And then to send some data to our CEP panel:
JSXEvent("Hello world from JSX", "myScriptMessage"); You could make this much more robust by adding a JSON polyfill in Adobe scripting, then sending all the data you'd need in the first argument for JSXEvent stringified, then parse it in CEP and potentially have your base64 rawdata and filepath to write to all in a single call without needing any UI from CEP at all. Unfortunately I'm not clever enough to know how to convert from base64 > PNG in pure JSX, but you could still trigger it from JSX if needed, then evalScript a callback function in JSX to indicate you're done. I don't know the scope of what you're doing but presumably if you originally wanted to do it in scripting it could still be helpful to know.
... View more
Mar 25, 2022
10:17 PM
1 Upvote
Been a while since I'd done this and I realized my main AHK script doesn't directly call COM this way so I wanted to reconfirm before leaving any bad information online. I've successfully ran this AHK script: #NoEnv
SendMode Input
SetWorkingDir %A_ScriptDir%
com := ComObjCreate("Illustrator.Application")
F1::
com.DoJavaScriptFile("C:\Users\TRSch\OneDrive\Documents\Adobe Scripts\testAlert.jsx") ; Couldn't get this to work without absolute path for some reason
return For some reason, I wasn't able to use local paths despite having SetWorkingDir -- may be rusty in AHK now that I primarily code in other languages though.
... View more
Mar 25, 2022
09:48 PM
1 Upvote
I use Logitech software to bind my gaming mouse to F13 - F24, then AutoHotKey to bind F13 - F24 to different keybindings (anything I want) including running scripts which trigger actions which have no hotkeys associated with them. This is specifically a Windows solution, but I've heard you can do similar things with other macro programs like Alfred, SharpKeys, etc (believe these aren't free, though). Scripting named actions is rather easy: app.doScript("Revert", "Default", false);
// (actionName, actionSet, showDialog?) And you can execute JSX scripts in Illustrator via AHK arbitrarily through COM objects: ; At some point above the hotkey, like the Auto-Execute/header section of script
com := ComObjCreate("Illustrator.Application")
; Then coding something that triggers on an F23 keypress:
F23::
com.DoJavaScriptFile("action1.jsx") ; Filepath is relative to AHK script locale
return ; Return null / stop execution of code after this line for this hotkey
F24::
com.DoJavaScriptFile("action2.jsx") ; etc
return You can just write one-line JSX scripts referring to the named Action like above, and connect them to any hotkey you'd want via the AHK COM objects above. That allows you to run any Action from any key or combination of any amount of keys regardless of the limitations you'd have with the keybindings Adobe gives you. There's probably some equivalent in other macro programs but I'm personally unfamiliar.
... View more
Mar 25, 2022
02:16 PM
1 Upvote
Sorry, you can't use import/export syntax without using a compiler like Webpack which is typically how Node is used. If you're wanting to do it in purely inline HTML: <!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8" />
<title>base64</title>
<!-- MUST INCLUDE THIS IN PRELOAD TO HAVE ACCESS TO REQUIRE() -->
<script>
window.require = require || cep_node.require;
</script>
<!-- THEN LOAD THE INLINE SCRIPT VERSION: -->
<script src="./convertBase64Legacy.js"></script>
</head>
<body style="margin: 0; border: 0; overflow:hidden;">
<button id="button">TEST BUTTON</button>
<!-- AND ADD CUSTOM LOGIC: -->
<script>
// Get our HTML element:
let button = document.getElementById("button");
// Add an event handler to trigger the function:
button.addEventListener("click", (evt) => {
convertBase64ToPNG(
"data:image/png;base64,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",
"test.png",
"C:/Users/TRSch/OneDrive/Desktop"
).then((result) => {
if (result.okay) alert("SUCCESSFUL");
else alert(result.error);
});
});
</script>
</body>
</html> The legacy/inline version of the script can be found here. Notice that by loading this inline file through a script tag, I have access to all the functions inside of it without needing to import or require them explicitly. Also notice that the Adobe forums auto-replaces colon characters in the base64 data, so for a non-modified version of the above, you'd look at this page.
... View more
Mar 24, 2022
07:42 PM
2 Upvotes
You could simulate the Pathfinder > Unite command exactly as the native UI behaves by loading .aia rawtext as an Action, executing it, then unloading the Action so it disappears from the Actions panel: /**
* Creates and runs a temporary action from given .aia rawtext
* {String} actionName Exact name of action
* {String} actionSet Exact name of set containing action
* {String} contents Raw .aia string of action file
*/
function runActionFromString(actionName, actionSet, contents) {
// Rather oddly AI always wants you to use a file instead of a string,
// so we'll just create a temp file every time we need to:
var tmp = File(Folder.desktop + "/tmp.aia");
tmp.open("w");
tmp.write(contents);
tmp.close();
// Then load, execute, and unload the action:
app.loadAction(tmp);
app.doScript(actionName, actionSet, false);
app.unloadAction(actionSet, "");
// And remove the file:
tmp.remove();
}
var actionStr =
"/version 3 /name [ 12 756e697465436f6d6d616e64 ] /isOpen 1 /actionCount 1 /action-1 { /name [ 5 556e697465 ] /keyIndex 0 /colorIndex 0 /isOpen 1 /eventCount 1 /event-1 { /useRulersIn1stQuadrant 0 /internalName (ai_plugin_pathfinder) /localizedName [ 10 5061746866696e646572 ] /isOpen 0 /isOn 1 /hasDialog 0 /parameterCount 1 /parameter-1 { /key 1851878757 /showInPalette -1 /type (enumerated) /name [ 3 416464 ] /value 0 } } }";
runActionFromString("Unite", "uniteCommand", actionStr);
... View more
Mar 24, 2022
03:53 PM
1 Upvote
I still don't know what this means, because "nesting" has a few definitions in the context of programming and art. Can you be more descriptive, or actually provide screenshots/files of before and after results?
... View more
Mar 24, 2022
02:30 PM
1 Upvote
It's not at all clear what you're asking. Can you describe it or provide a less blurry video, or screenshot examples of the input compared to the output? Are you asking for a script that condenses your selection to the smallest surface area, i.e. bin packing?
... View more
Mar 24, 2022
02:04 PM
1 Upvote
You can't do link.click() or automated downloads in CEP as an isolated sandbox, iirc. Also I live in a different timezone as you so please don't respond multiple times asking "Can you please reply" within a 2 hour span, I'm probably sleeping or busy/working, not intentionally ignoring you. You were posting at 2AM in my timezone. I made a standalone CEP extension that does this as a demonstration. It has a specific file containing utility functions that handle the entire process and a basic UI that allows changes to this main function's 3 parameters. It does require you to use NodeJS but it's framework agnostic and purely vanilla Javascript, this function should work regardless of whatever setup you're using.
... View more
Mar 23, 2022
06:03 PM
2 Upvotes
You'd use the Canvas API. Say we have a canvas element somewhere in our HTML: <canvas id="canvas" /> Setting it's data as base64 is relatively simple: // Grab canvas element, then utility variables:
let canvas = document.getElementById("canvas"),
context = canvas.getContext("2d"),
image = new Image();
// Manually write in base64 as it's source attribute:
image.src="data:image/png;base64,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";
// Then during it's loaded lifecycle
image.onload = function () {
// Force a redraw of the image to ensure it displays correctly:
context.drawImage(image, 0, 0);
}; If at any point we want this as PNG after having done the above: let myPNGimageData = canvas.toDataURL("image/png"); If you have issues, just search for the steps: "HTML canvas base64" "HTML canvas download png"
... View more
Mar 16, 2022
11:56 AM
3 Upvotes
This seemed like good practice so I've written one this morning: /**
* SetPath.jsx
*
* When the user selects a single anchorpoint of a single shape, this script will mimick a "Set First Vertex"
* menu action in After Effects along with adding "Reverse Path Direction" support either via prompt or config.
*/
var config = {
reverse: {
showPrompt: false, // If true, asks whether you want to reverse
value: false, // If true, will reverse paths. This is overridden whenever showPrompt above is set to true
title: "Reverse?", // Title of the window
message: "Choose yes if wanting to reverse path direction:", // Description of window
},
showAlert: true, // If true, displays an alert on completion
message: "Done!", // Contents of the alert on completion
//
// Various debugging utilities for catching errors and edgecases:
debug: {
alertSelectedAnchor: false, // if true, alert with original selected point index
reportNewAnchors: false, // if true, alert string of indices to rewrite as
endBeforeRunning: false, // if true, ends function before any modifications
},
};
//
var shouldReverse = false; // Needed as a global variable
Array.prototype.filter = function (callback) {
var filtered = [];
for (var i = 0; i < this.length; i++)
if (callback(this[i], i, this)) filtered.push(this[i]);
return filtered;
};
Array.prototype.setNewOrigin = function (index) {
if (index > this.length || index < 0) return this;
return [].concat(this.slice(index), this.slice(0, index));
};
Array.prototype.findIndex = function (callback) {
for (var i = 0; i < this.length; i++)
if (callback(this[i], i, this)) return i;
return -1;
};
Array.prototype.forEach = function (callback) {
for (var i = 0; i < this.length; i++) callback(this[i], i, this);
};
Array.prototype.map = function (callback) {
var mappedParam = [];
for (var i = 0; i < this.length; i++)
mappedParam.push(callback(this[i], i, this));
return mappedParam;
};
function get(type, parent) {
if (arguments.length == 1 || !parent) parent = app.activeDocument;
var result = [];
if (!parent[type]) return [];
for (var i = 0; i < parent[type].length; i++) result.push(parent[type][i]);
return result;
}
// Utility function that checks to ensure only one vertex is active and returns it if so
function hasSinglePointSelection(item) {
function findActiveVertices(item) {
return get("pathPoints", item)
.map(function (point, index) {
return {
index: index,
point: point,
};
})
.filter(function (item) {
return item.point.selected == PathPointSelection.ANCHORPOINT;
});
}
var selectedPoints = findActiveVertices(item);
if (!selectedPoints.length) return false;
else return selectedPoints.length == 1 ? selectedPoints[0] : false;
}
function main() {
// First grab selection
var selection = get("selection");
// But ensure only one object is selected:
if (!selection.length || selection.length > 1)
return alert("Must have a selection and only one path selected");
// Isolate our target shape and ensure it's the correct type:
var target = selection[0];
if (!/path/i.test(target.typename) || /compound/i.test(target.typename))
return alert("Selection must be a valid PathItem and not a compound path");
// Ensure only one anchor has been selected so we know our intended origin:
var chosenVertex = hasSinglePointSelection(target);
if (!chosenVertex) return alert("Must have only one vertex selected");
var newOriginIndex = chosenVertex.index;
if (config.debug.alertSelectedAnchor) alert(newOriginIndex);
// Since PathPoint is a ref and mutable, we need to shallow copy it to rearrange the entire set:
var points = get("pathPoints", target)
.map(function (point, index) {
return {
index: index,
data: shallowCopy(point),
};
})
// And use a shorthand utility for setting the new index as the first vertex:
.setNewOrigin(newOriginIndex);
// Very basic UI to flip path direction via confirm dialog:
shouldReverse = config.reverse.value;
if (config.reverse.showPrompt)
shouldReverse = confirm(
config.reverse.message,
!config.reverse.value,
config.reverse.title
);
if (shouldReverse) {
// We can't simply reverse an array like [0,1,2,3] if 0 is our selection because then it becomes [3,2,1,0]:
points = points.reverse();
// We have to set the new origin of the reversed array back to the new origin, like [0,3,2,1]
points = points.setNewOrigin(
points.findIndex(function (point) {
return point.index == newOriginIndex; // So we use an ES6-like array method of findIndex
})
);
}
// This just reports a list of anchors like [2,3,0,1] in case anything weird is happening:
if (config.debug.reportNewAnchors) {
var str = "";
str += points
.map(function (point) {
return point.index;
})
.join(",");
alert(str);
}
if (config.debug.endBeforeRunning) return null;
// So for each point, we rewrite it with shallow copy values:
get("pathPoints", target).forEach(function (point, index) {
rewritePoint(point, points[index].data, index);
});
// Then we have to manually clear the selection, because for some reason AI deselects the entire object if you toggle point selections too much:
get("pathPoints", target).forEach(function (point, index) {
point.selected = PathPointSelection.NOSELECTION;
});
// And manually reselect the first vertex, which should correspond to the original selected vertex we'd wanted to set as First Vertex before running the script
target.pathPoints[0].selected = PathPointSelection.ANCHORPOINT;
if (config.showAlert) alert(config.message);
}
function shallowCopy(ref) {
var temp = {};
for (var key in ref) temp[key] = ref[key];
return temp;
}
function rewritePoint(a, b, index) {
var keys = ["anchor", "leftDirection", "rightDirection"];
keys.forEach(function (key) {
var tempKey = key + "";
// If we reverse directions, then we have to set leftDirection > rightDirection and vice versa otherwise we produce artifacts:
if (/direction/i.test(key) && shouldReverse && b[key])
tempKey =
key.replace("Direction", "") == "right"
? "leftDirection"
: "rightDirection";
a[tempKey] = b[key];
});
}
main(); Unlike the one shown above, this works based on user selection. If you select an anchor of a shape then run this script it will mimick a Set First Vertex action of After Effects without creating new annotation text or requiring you to know the point index in order to type it into a prompt, reversing the path is optional and can be set to default (or set to prompt you) in the config variable at the top:
... View more
Mar 12, 2022
01:55 AM
3 Upvotes
Here's my attempt. Recursive lookup of ancestors and sublayers, supports any amount of find/replace options, also has options for enabling whether to affect locked or hidden layers: var config = {
// Support for any amount of find/replace parameters here
targets: [
{
find: "foo",
replace: "bar",
},
{
find: /(locked|hidden)/i, // including support for Regex patterns
replace: "Success",
},
],
affects: {
// names: false, // Whether it changes names, not implemented but possible
content: true, // Must be true at this basic state of the script
},
searchIn: {
hidden: false, // Whether it changes hidden layer contents
locked: false, // Whether it changes locked layer contents
},
};
//
// Don't alter the below unless you know what you're doing.
//
// Very simple ES6 polyfills for Array methods to make code more concise:
Array.prototype.filter = function (callback) {
var filtered = [];
for (var i = 0; i < this.length; i++)
if (callback(this[i], i, this)) filtered.push(this[i]);
return filtered;
};
Array.prototype.forEach = function (callback) {
for (var i = 0; i < this.length; i++) callback(this[i], i, this);
};
// This is a bit too verbose for what we need here, but as-is can recurse to find sublayers no matter the depth if we use:
// var list = get("layers"); <-- This is *all* layers and sublayers, not just the root-level / depth 0 ones from doc.layers.
function get(type, parent, deep) {
if (arguments.length == 1 || !parent) {
parent = app.activeDocument;
deep = true;
}
var result = [];
if (!parent[type]) return [];
for (var i = 0; i < parent[type].length; i++) {
result.push(parent[type][i]);
if (parent[type][i][type] && deep)
result = [].concat(result, get(type, parent[type][i], deep));
}
return result;
}
// The main obstacle is getting a recursive parent chain, then determining if any part of this chain doesn't meet requirements:
function checkAncestry(frame) {
// So we'll split the logic into just constructing a recursive Array containing data about parents when applicable:
function getAncestry(frame) {
// And have a dedicated recursive function to look into the parent then recurse and concat an Array when needed:
function getParentChain(item, data) {
if (item.parent && /layer/i.test(item.parent.typename))
data = getParentChain(
item.parent,
[].concat(
{ hidden: !item.parent.visible, locked: item.parent.locked },
data
)
);
return data;
}
// For ease, we start the chain with the textFrame's hidden/locked data:
return getParentChain(frame, [
{ hidden: frame.hidden, locked: frame.locked },
]);
}
// Now that we have a list like [ { locked: true, hidden: false }, {locked: false, hidden: false }, ... ]
// We can filter it to see if there are any hidden/lockeds present:
var chain = getAncestry(frame),
hiddenStatus =
chain.filter(function (i) {
return i.hidden;
}).length > 0,
lockedStatus =
chain.filter(function (i) {
return i.locked;
}).length > 0;
// AI doesn't handle short circuiting well, so if no locked/hiddens exist, we'll explicitly return:
if (!lockedStatus && !hiddenStatus) return true;
// Otherwise we need to ensure the chain contents either match the queries or returned empty:
else
return (
(hiddenStatus == config.searchIn.hidden || !hiddenStatus) &&
(lockedStatus == config.searchIn.locked || !lockedStatus)
);
}
// Our main function can now be very concise, we get all text frames:
get("textFrames")
// But remove any that don't match our search queries for hidden/locked
.filter(function (i) {
return checkAncestry(i);
// Then for every remaining textFrame:
})
.forEach(function (frame) {
// And for every find/replace option:
config.targets.forEach(function (target) {
// Replace the text content of the frame:
frame.contents = frame.contents.replace(target.find, target.replace);
});
});
... View more
Mar 04, 2022
03:53 PM
Fair enough. You have the scripting tag on the thread fyi, so it looks like you're asking for scripts.
... View more
Mar 04, 2022
03:26 PM
Very easy to solve with a basic script: var active = app.activeDocument.activeLayer,
parent = getParent(active);
function getParent(layer) {
if (layer.parent && /(layer|group)/i.test(layer.parent.typename))
return getParent(layer.parent);
return layer;
}
active.move(app.activeDocument, ElementPlacement.INSIDE);
active.move(parent, ElementPlacement.PLACEBEFORE); // PLACEAFTER if wanting it to be below parent That way you could use Insert Menu Item within an Action and assign to a hotkey if needed, though looking at the document it'd be far better to parse through all layers and sublayers then separate any with a "\d*\.\s" prefix and automatically sort/order them at the root of the document instead of ever needing to do any manual dragging.
... View more
Mar 02, 2022
12:52 PM
1 Upvote
Hi Craig, we could make this a bit more dynamic and less reliant on the hard-coded values for selMask.fillColor assignment and name checking. Consider this: // Store all our colors in a global object with key values matching mask names:
var masks = {
Bar: createRGB(255, 0, 0),
Round: createRGB(0, 255, 0),
Heart: createRGB(0, 0, 255)
}
// Create a simple utility function to allow us to more easily make RGB colors:
function createRGB(red, green, blue) {
var temp = new RGBColor();
temp.red = red;
temp.green = green;
temp.blue = blue;
return temp;
}
// Now we dynamically retreive all our key mask names:
var keys = [];
for (var key in masks) keys.push(key);
// And create a RegEx to check them:
var maskRX = new RegExp("^(" + keys.join("|") + ")$");
// Equivalent to /^(Bar|Round|Heart)$/
function setCustomCrop() {
var aDoc = app.activeDocument;
var selMask = selection[0];
customCrop = selMask.name;
var CropGroup = aDoc.groupItems['CropGroup'];
var myItem = CropGroup.pageItems;
var myMask = myItem[customCrop];
myMask.locked = false;
// If the name matches our RegEx above:
if (maskRX.test(selMask.name))
selMask.fillColor = masks[selMask.name]
// ^ We use the name to match the color from our masks variable
/**
* No need for long ternary chains and makes our code more dynamic
* because adding a new mask is as simple as adding:
*
* SmallHeart: createRGB(150, 50, 50)
*
* In the original "masks" variable at the top.
*/
} In ES6 this would be much easier: let maskRX = new RegExp(`^(${Object.keys(masks).join("|")})$`); I mention this mostly because the more we rely on static or hard-coded values, the more opportunities we have to get errors on typos and misspellings, and the more modifications you'd have to make to the script in multiple locations if you were to change one of these variable names or add/remove them. Doing this in a dynamic way has a bit of an overhead cost but afterwards can save a lot of time and energy maintaining it later on, because I can now have any number of masks all neatly organized in the "masks" variable.
... View more
Feb 21, 2022
08:46 AM
Math.round: var foo = UnitValue(100, "pt").as("mm");
alert(foo); // 35.27777777778
alert(Math.round(foo)); // 35
... View more
Feb 21, 2022
02:45 AM
1 Upvote
The syntax for try/catch is: try {
// ...
} catch (error) {
// ...
} Whereas in the above you seem to have: // ...
} catch (error) {
// ... Be sure you have matching braces for the above } which needs to be preceeded with a "try {" prior and { which needs a closing brace "}" below.
... View more
Feb 20, 2022
08:06 AM
2 Upvotes
In this case René's script was based on a GroupItem and I could see that in your screenshot, your selection would have been an array of individual PathItems rather than an array with a single GroupItem containing a pathItems property so I assumed the error was likely "Property 'pathItems' does not exist" but that's mostly from experience. Most of us learned from trial and error and combing through these forums for references if not asking questions and others coming to explain, but if I were to do it all over again I'd just go off and take basic Javascript classes that aren't specifically es6+ versions of Javascript (arrow functions, let/const, extended Array prototype functions like filter, forEach, map, etc) to get a bearing on how Javascript works while experimenting and using what I've learned in JSX files for Illustrator scripting. If you're ever unclear on any thing in particular and the ESTK debugger (which I don't personally use) doesn't flag it with a useful error to know what exactly is wrong, you can always wrap the code in question in a try/catch block: try {
// JavaScript Document for Illustrator
app.coordinateSystem = CoordinateSystem.ARTBOARDCOORDINATESYSTEM;
var doc = app.activeDocument;
var gr = selection[0]; // if GroupItem
var str = "";
for (var h = 0; h < gr.pathItems.length; h++) {
str += h + "\t" + gr.pathItems[h].length + "\r";
}
str = str.substring(0, str.length - 1);
var pointText = doc.textFrames.pointText([20, -20]);
pointText.contents = str;
} catch (error) {
alert(error);
} You would have gotten an alert box saying "Property 'pathItems' does not exist on type 'PathItem'" or something similar, which is a common debugging technique and something you'd be taught in basic Javascript courses (though not specifically with an alert box and usually with a console object which we don't have access to). I personally use VSCode to write code and there are several Adobe plugins for it to run the code, if you want a sample of what that feels like try writing "app.activeDocument." (no quotes) and expanding the info icon in the popup in this web app I made a while back that has Intellisense and shows the DOM as you type. There are also links to some online documentation like AEenhancers that I find far better than Adobe's own PDF documentation. Basically: take it slow, remember this is normal (albeit an older version of) Javascript and there's no shortage of answers/classes online about JS, Google the exact text of any errors you have, ask questions in this forum whenever you get stuck, search for people who have asked similar questions to what you're currently trying on this forum and take the time to read them in full, be consistent and don't give up. It's frustrating and hard in the beginning but that's because you're learning a new language, you're learning how to speak to a computer and it turns out computers can be very dense and kind of dumb in their own way so you have to be extremely precise when you give them instructions. You wouldn't be fluent in Spanish or any other second language in just a week or even a few months either and no one is, it's a practiced ability that comes in "do" time.
... View more
Feb 20, 2022
06:07 AM
Try pressing control + g to group your selection before running the script.
... View more
Jan 16, 2022
10:44 AM
Lorem Ipsum text?
... View more
Jan 10, 2022
02:39 PM
5 Upvotes
A lot of great answers here with reasons why you should but just to cover all aspects, why you shouldn't: Recursive functions only implicitly stop execution which can cause unexpected behavior if using sync and async code simultaneously or when using async recursion (admittedly not relevant for JSX scripting) Recursion is exponentially harder to debug and understand the full boundaries of while reading either for someone else or your future self a few months from now Recursion is exponentially more error and bug-prone Recursion is exponentially slower because it adds to the callstack per lookup and descendant, resulting in several redundant function calls and loops The only time recursion should be used (at least for me in my own code) is when necessary: we're working with nestable objects or dynamic user-generated content and we cannot account for the complexity of depth/nesting, we're not using a rigid structure. This could be pageItem contents or folder/files contents as noted or it could be an HTML DOM where recursion is necessary to ensure accurate non-shallow results, but it could also be completely gratuitous and a way to shoot yourself in the foot multiple times if you try implementing recursion in cases when unneeded. For practice, there's no harm in playing with recursion liberally no matter what the case is just to refine your own skills and understanding. However for production since recursion can be such a double-edged sword, you have to consider the fact that adding it can be the source of several near-future patches and bugs (from unexpected parameters fed into your recursive function args) or that using it in a document with thousands of entries might take 10x longer to do the same thing as a shallow lookup.
... View more
Jan 03, 2022
05:57 AM
1 Upvote
Sorry, had a typo and a faulty condition. Serves me right for so confidently offering recursion thinking I had cracked it so easily ha, I've verified the below on your file: var targetLayers = ["Test"];
function getAllPathItems(parent) {
var list = [];
for (var i = 0; i < parent.pageItems.length; i++) {
var item = parent.pageItems[i];
if (item.pageItems && item.pageItems.length)
list = [].concat(list, getAllPathItems(item));
else if (/path/i.test(item.typename) && !/compound/i.test(item.typename))
list.push(item);
}
return list;
}
function getLayer(layerName) {
try {
return app.activeDocument.layers.getByName(layerName);
} catch (err) {
return null;
}
}
function restyleAllPathItemsWithin(layers) {
app.selection = null;
for (var i = 0; i < layers.length; i++) {
var layer = getLayer(layers[i]);
if (!layer) continue;
else {
var list = getAllPathItems(layer);
for (var index = 0; index < list.length; index++) {
var pathItem = list[index];
pathItem.stroked = true; // open stroke
pathItem.strokeCap = StrokeCap.BUTTENDCAP; //ROUNDENDCAP, PROJECTINGENDCAP
pathItem.strokeJoin = StrokeJoin.ROUNDENDJOIN; //BEVELENDJOIN, MITERENDJOIN
}
}
}
}
restyleAllPathItemsWithin(targetLayers); I'm not sure if you're wanting to target only specific layers by name or if you're just trying to change all document strokes -- you could do that though. If the goal is to change all strokes we could just feed a list of all the layers into the main function, and it might be easier to choose which layers (or names of annotation like "LWPOLYLINE") to exclude rather than include. Or target all "LINE" group children, etc.
... View more
Jan 03, 2022
05:32 AM
🤔 Hmmm, I thought that'd work. It's pretty hard to do this blindly, can you upload the file and link here?
... View more
Jan 03, 2022
05:16 AM
1 Upvote
Here's recursion in case you have deep nesting: var targetLayers = ["Test"];
function getAllPathItems(parent) {
var list = [];
for (var i = 0; i < parent.pageItems.length; i++) {
var item = parent.pageItems[i];
if (item.pageItems && item.pageItems.length)
// We call this same function to rerun within it's own loop:
list = [].concat(list, getAllPathItems(item));
else if (/path/i.test(item.typename) && !/compound/i.test(item.typename))
list.push(item);
}
return list;
}
function getLayer(layerName) {
try {
return document.layers.getByName(layerName);
} catch (err) {
return null;
}
}
function restyleAllPathItemsWithin(layers) {
app.selection = null;
for (var i = 0; i < layers.length; i++) {
var layer = getLayer(layers[i]);
if (!layer || !layer.pathItems.length) continue;
else {
var list = getAllPathItems(layer);
for (var index = 0; index < list.length; index++) {
var pathItem = list[index];
pathItem.stroked = true; // open stroke
pathItem.strokeCap = StrokeCap.BUTTENDCAP; //ROUNDENDCAP, PROJECTINGENDCAP
pathItem.strokeJoin = StrokeJoin.ROUNDENDJOIN; //BEVELENDJOIN, MITERENDJOIN
}
}
}
}
restyleAllPathItemsWithin(targetLayers);
... View more
Jan 03, 2022
05:06 AM
Sorry, I don't see the file yet. If you happened to try running the above snippet directly though, you'd be missing the value of "testLayers" as the param in case that was the issue.
... View more
Jan 03, 2022
04:29 AM
1 Upvote
No worries. That's true, it has to do with how AI contains items. For instance: layer.pathItems - This is a collection of root-level paths on the layer. layer.pageItems - This is a collection of any type of root-level object on the layer. Layer.pathItems is not every path inside of the layer, only every path whose parent is the layer. The issue would be that a group counts as a pageItem, so a path inside of a group would be something more like: layer.pageItems[0].pathItems[2] <= Because the pathitem parent is the group, not the layer. Pathitems may exist only in this group Layer.pathItems only contains paths that aren't in a group or a compound path, if the parent is explicitly the layer. We can fix this for shallow groups pretty easily: function getAllPathItemsInLayer(layer) {
var list = [];
for (var i = 0; i < layer.pageItems.length; i++) {
var item = layer.pageItems[i];
// If this is a group or compound path, we need to add it's pathItems to our list:
if (
/group|compound/i.test(item.typename) &&
item.pathItems &&
item.pathItems.length
) {
for (var e = 0; e < item.pathItems.length; e++)
list.push(item.pathItems[e]);
} else if (/path/i.test(item.typename)) {
// Otherwise if this is a pathItem, we can add it to the list directly:
list.push(item);
}
}
return list;
}
function getLayer(layerName) {
try {
return document.layers.getByName(layerName);
} catch (err) {
return null;
}
}
function restyleAllPathItemsWithin(layers) {
app.selection = null;
for (var i = 0; i < layers.length; i++) {
var layer = getLayer(layers[i]);
if (!layer || !layer.pathItems.length) continue;
else {
// We need to gather any children pathItems in case their parent is a Group/Compound Path
var list = getAllPathItemsInLayer(layer);
for (var index = 0; index < list.length; index++) {
var pathItem = list[index];
pathItem.stroked = true; // open stroke
pathItem.strokeCap = StrokeCap.BUTTENDCAP; //ROUNDENDCAP, PROJECTINGENDCAP
pathItem.strokeJoin = StrokeJoin.ROUNDENDJOIN; //BEVELENDJOIN, MITERENDJOIN
}
}
}
}
restyleAllPathItemsWithin(testLayers); But for complicated groups (pathItems within groups within groups or any dynamic nesting) then we'd have to do some recursion and it's likely going to look either very intimidating and confusing or very simple and confusing.
... View more
Jan 02, 2022
11:47 PM
2 Upvotes
Hello, consider this: var document = app.activeDocument;
var allLayers = app.activeDocument.layers;
var testLayers = ["Test"];
// Why do this? Because if we query a layer name that doesn't exist, it causes silent failure.
function getLayer(layerName) {
try {
return document.layers.getByName(layerName);
} catch(err) {
return null;
}
}
function restyleAllPathItemsWithin(layers) {
// An alternative to deselect all artwork, though it shouldn't actually matter here
// because we can change the strokes regardless of whether other art is selected
app.selection = null;
// Iterate through our layer list:
for (var i = 0; i < layers.length; i++) {
// Get our target layer:
var layer = getLayer(layers[i]);
// If a layer wasn't found or there are no items to restyle, continue looping:
if (!layer || !layer.pathItems.length) continue;
else {
// Layer.pathItems is a collection of objects, not a single one.
// To restyle any given object, we have to iterate over them individually:
for (var index = 0; index < layer.pathItems.length; index++) {
var pathItem = layer.pathItems[index];
pathItem.stroked = true; // open stroke
pathItem.strokeCap = StrokeCap.BUTTENDCAP;//ROUNDENDCAP, PROJECTINGENDCAP
pathItem.strokeJoin = StrokeJoin.ROUNDENDJOIN;//BEVELENDJOIN, MITERENDJOIN
}
}
}
}
restyleAllPathItemsWithin(testLayers); Essentially you have everything correct here, but the issue is that you're trying to assign properties to an Array-like collection (pathItems) rather than one entry of that collection (i.e., pathItems[0]).
... View more
Dec 28, 2021
09:49 AM
Might help to get a technical answer alongside an abstract one: If Javascript were music, then JSON would be something like mp4 -- just a way to store that music in a way that computers can understand and make use of in the correct context.
... View more
Dec 28, 2021
09:35 AM
1 Upvote
1) If run "directly" by selecting the script as a script menu item, the script opens up the Color Picker and allows manual input of a color code number. In my case, I run it from the script menu using Keyboard Maestro as such: If I were doing this through AutoHotKey (which is very similar to Keyboard Maestro) I'd probably run a sequence of key presses starting with Alt + F (selects File menu), R (selects File > Scripts), then double up keys to run the last entry in the Scripts menu. It's not ideal but it could be an alternative that also works given the problem you note with #2. 2) If I try to run the script inside of an Illustrator action -- run by pressing F12 -- when I press that hotkey the Color Picker opens with focus on the hex input window, ready for me to type a number. I manually input my selected color number. However, when I hit "Okay," Illustrator freezes (spinning beachball). So, although the latest script runs properly when selecting it manually as a script menu item or from KM as a script menu item, it freezes Illustrator when called from an Illustrator action. Is that what you would expect? I'd have to look into this. I've seen mentions on other threads about dealing with Actions that running one within another (like we're doing here, using an Action to run a script which itself creates and runs an Action) causes freezing. I wasn't sure if this was OS or app-version specific but it may just be a general rule-of-thumb. If this were the case (and it's very likely) then it does make sense here because the "inside" Action isn't opening the color picker but setting the color from our color value; the moment the picker closes would be the exact time freezing would occur. Also, am I supposed to be entering the pre-determined color code someplace before running the script? Or does the script simply "take" you to the point of entering the hex number in the proper Color Picker window (which is what it seems to be doing)? The script is supposed to take you there. Prompting the picker with a color value just sets the value to that given color, so if you were to have a green active fill then the picker should open on that green instead of coming up as black each time and so on. Thanks again for the very interesting solution to this and for providing the first baby steps in my JSON education. Do you have a recommendation for the best book(s) to learn about JSON? Or would you say that language is on the way out and one’s time would be better spent learning a different language? Well, JSON isn't formally a part of Extendscript but otherwise it's very much standard practice and arguably more relevant in modern coding than ever before. JSON is Javascript Object Notation, which is (and this is somewhat a simplification) a way of writing a given Javascript value to a string for use in a text file or as something like a message between two endpoints (like how Slack in your browser can prompt or open the desktop app version of Slack through WebSockets). So for instance, if I have a Javascript variable like this: var data = {
foo: "bar",
list: [1, 2, 3]
};
alert(data.foo); // "bar" Let's say we need to write the content of the variable data and everything inside no matter the count, how deep or complex then how would we do this? If we try writing it to a text file: var tempFile = File(Folder.desktop + "/tempFile.txt");
tempFile.open("w");
tempFile.write(data);
tempFile.close(); It would come out looking something like this: {
foo: "bar",
list: [1, 2, 3]
} That doesn't look bad and seems like it might be fine, but how do we get it out? How do we make it so that it's an object again? var tempFile = File(Folder.desktop + "/tempFile.txt");
tempFile.open("r");
var data = tempFile.read();
tempFile.close();
alert(data); // "\{\r\n\tfoo:\s\"bar\",\r\n\tlist:\s[1,\s2,\s3]\r\n}"
alert(typeof data); // "String" It looks a lot different when we try to retreive it and it's a string. How would we fix that? Well you could theoretically use eval() on it to try and evaluate it's value though eval is very dangerous (and deprecated specifically due to this) because you're trying to evaluate something you might not understand the value of or that value could be rewritten/reassigned just before evaluation in certain methods for hacking. The way we solve for this entire problem is by using JSON: var tempFile = File(Folder.desktop + "/tempFile.json");
tempFile.open("w");
tempFile.write(JSON.stringify(data)); // <== Note the difference here
tempFile.close();
// You cannot do this in scripting natively, you have to add JSON2.jsx or a similar polyfill. This means our new file looks like this: {
"foo": "bar",
"list": [1, 2, 3]
} Your first reaction might be "That looks identical to the first time" and, well, yeah 😏 It's very close. The real difference is that I can re-integrate this value from an external source like a file very easily: var tempFile = File(Folder.desktop + "/tempFile.json");
tempFile.open("r");
var data = tempFile.read();
tempFile.close();
alert(data); // "\{\r\n\tfoo:\s\"bar\",\r\n\tlist:\s[1,\s2,\s3]\r\n}"
alert(typeof data); // "String"
data = JSON.parse(data); // Because we can parse a JSON, it can become identical to it's original value.
alert(data); // "[Object Object]"
alert(data.foo); // "bar" And the difference looks very minor here only because I'm using such a simple example. JSON can't handle every kind of Javascript value though; it can't store functions or RegExp primitives mostly for security reasons; but you can store any amount of basic data like objects, arrays, strings, numbers, evaluated expressions, and so on regardless of the depth or complexity of that data.
... View more
Copyright © 2024 Adobe. All rights reserved.
AdChoices