I suspect OP meant to write: var selectionArray = []; … selectionArray.push(item); … docRef.selection = selectionArray; But your way is simpler.   OP may also want to allow for slight differences in object size, as widths and heights may not be exactly identical depending on how the objects were created:   var margin = 5; // the amount of variation to allow, in pts if ( item.width >= objWidth - margin && item.width <= objWidth + margin && item.height >= objHeight - margin && item.height <= objHeight + margin )   ... View more

I’m guessing “BASE FILE” is a template file containing two existing rectangles, from which all of the generated files are derived. For each line of the CSV, the rectangles should be resized (remaining centered on their original position) and the document saved under a new file name. OP can confirm/correct this as appropriate. ... View more

It is certainly doable via scripting. It will help if you can post an example of the CSV data you’re using and a mockup of the artwork you wish generated from it. That will allow folks here to gauge the requirements and complexity of the job, request any additional details, and offer code or quotes as appropriate. ... View more

Alternatively, for the `source art` parameter you can pass only the page items that lie on a specific artboard, leaving page items on other artboards untouched. Alas, AI doesn’t allow you to write `page items of artboard "NAME"`, which would be ideal, so you will have to figure out which page items lie on a particular artboard for yourself.   The easiest solution is to put each artboard’s content into a different layer while building the artwork, allowing you to write `rasterize source art page items of layer "NAME" …`. Otherwise, you’ll need get the bounding boxes of all path items in the document, then sort those into separate lists according to which artboard they lie on, although that’s more work to implement and will run slower.   I can’t really provide more specific recommendations without knowing more about your documents and why you want to rasterize them. ... View more

By default, `rasterize` uses the bounds of artboard 1. You can specify a different area to rasterize via the `inside` parameter:   tell application "Adobe Illustrator" tell document 1 set boundRect to {0, 1024, 2048, 0} -- {left,top,right,bottom} in points rasterize source art page items with options {resolution:72} inside boundRect end tell end tell ... View more

Having ensured the text frame is area text, set the height and width of the path that defines its border: tf.textPath.height = rects[i].height; tf.textPath.width = rects[i].width;   ... View more

Sure. Looking at the characters used, I’d say that is Base64 encoded data. You can find tools online for decoding that.   As for what it decodes to, that may not be what you’re looking for. I imagine anything from a script that’s been run through the ExtendScript obfuscator to protect its author’s IP, thru some evil malware executable, to some other data encoding (e.g. ZIP compression) that needs further processing to unfold.   Honestly, we can’t even tell if it’s anything to do with Adobe Illustrator scripting. Because you’ve not told us anything about it.   It will really help in future if:   1. You state exactly where it found it, what [you think] it contains, and why you need it decoded. You don’t want anyone mistaking your post for nefarious activity, lest they sic Adobe’s moderators on you.   2. Not paste a huge lump of compressed data, of unknown provenance into the forum here, but rather took small screenshots of its start and end (which is what provides the clues that it is Base64), and attach the original file to your post using this control: Or, don’t even provide the full file until somebody asks you for it first. As I say, ettiquete, moderators, etc.   Sorry I don’t have time further, but perhaps if you can fill in the missing details then others can.   HTH ... View more

Yeah, I did. Teaches me not to read closer. Damn zombie posts with tiny dates. Really wish Adobe forums’d indicate old posts brought back to life after X months/years. It can be useful to do so, but I’m not the first to be suckered by that. ... View more

It wasn’t that. It was the fact OP’s code obviously intended to iterate over the text frames manipulating each one in turn, but it forgot to ask for `app.activeDocument.textFrames[i].contents` so was actually attempting to manipulate `app.activeDocument.textFrames.contents` multiple times instead. i.e. A bug in OP’s code which might well explain the [admittedly cryptic and unhelpful] error message. ... View more

I suspect you meant this:   function AddTextToFrame(num){ var idoc = activeDocument; for(var i=0; i<idoc.textFrames.length; i++){ var textFrame = idoc.textFrames[i]; if(textFrame.contents == "99999"){ textFrame.contents = num; } } } ... View more

That lets you create a new object. However, JXA can’t tell AI where in the stacking order to insert it.   AppleScript: tell application "Adobe Illustrator" tell document 1 set theEllipse to make new ellipse with properties {bounds:{100, 100, 500, 500}} -- create new object at front move theEllipse to after page item -1 of layer 1 -- move it behind the backmost object end tell end tell   JSX: var doc = app.activeDocument var pathItems = doc.layers[0].pathItems var newEllipse = pathItems.ellipse(100, 100, 400, 400) newEllipse.move(pathItems[pathItems.length-1], ElementPlacement.PLACEAFTER) // this works   JXA: var aiApp = Application('Adobe Illustrator') var doc = aiApp.documents[0] var newEllipse = aiApp.make({ new: 'ellipse', at: doc.layers[0], withProperties: {bounds: [100, -100, 500, 100]} }) newEllipse.move({ to: doc.layers[0].pageItems[-1].after }) // this fails   That’s a JSX defect: it doesn’t support beginning/end/before/after specifiers. Contrast appscript, which gets it right:   #!/usr/bin/env python3 from appscript import * ai = app('Adobe Illustrator') doc = ai.documents[1] el = ai.make(new=k.ellipse, at=doc.layers[1], with_properties={k.bounds: [100, -100, 500, 100]} ) el.move(to=doc.layers[1].page_items[-1].after) # this works   And there are lots of other bugs and omissions as well. Many of which I reported while JXA was still being developed; most of which they ignored. I’m not kidding when I say JXA is not fit for purpose: I wrote appscript and several other production-apple event (“AppleScript”) bridges over the last 20 years, and Python 3 + appscript still drives my own high-end AI automations, so I know this stuff way better than JXA’s authors ever did.   The real deal-killer though is that none of JXA’s bugs and omissions will ever be fixed, because Apple has already abandoned it and scrapped the department that wrote it. So adopting JXA for your own workflows now is a fool’s errand.   If you want to use JavaScript then use Adobe’s JSX (ES3), and let Adobe know (politely) that you’re keen for them to bring UXP (Node) to AI as quickly as possible.   (I am currently investigating TypeScript as a modern, forwards-compatible option for future AI automations, as TS runs on everything from ES3 to the latest Node. But I do wish Adobe would hurry up with UXP as that will greatly simplify life all round.)   If you need to integrate with other Mac applications then use AppleScript—and pray that Apple doesn’t let it rot too quickly, as the likelihood of them ever creating a competent modern successor to AppleScript is roughly nil.   ... View more

Hi Eddy, I’ve sent you a private message with my contact details. Best regards, Hamish ... View more

Honestly, you’re wasting your time trying to use JXA (Apple’s “JavaScript for Automation”). It has never worked properly and is long abandoned by Apple. (JXA failed so hard, Apple killed off its entire Mac Automation department.) It works up to a point, albeit confusingly, whereupon something breaks and there’s nothing you can do about it. For instance, its support for “class” and “constant” types is all screwed up, and it’s impossible to specify “beginning”/“end”/“before”/“after” when making new elements or moving/duplicating existing ones. It also tends to break more often on older apps that predate Mac OS X, such as Adobe CS and Microsoft Office, due to the greater variation and unique quirks in how those apps’ “AppleScript” interfaces are designed.   If you want to use JavaScript, use JSX (Adobe’s “ExtendScript”). It’s ancient ES3 and not without its own bugs (e.g. the notorious “out of space” PARM/MRAP error), but at least those bugs are understood and can be worked around (e.g. don’t define any global names and wrap all your code in an anonymous function call). You can write JSX code in any plain text/code editor, although if you want integrated debugger support you’ll have to use Microsoft’s VSCode with Adobe’s JSX plugin. If you need to control multiple Adobe apps, use JSX + Bridge (see Adobe’s BridgeTalk class documentation). And Adobe still actively support JavaScript and will hopefully start rolling out modern UXP for AI and ID in the next couple years; JSX is just dead product walking.   Otherwise, stick to AppleScript for Mac desktop automation. It’s a crappy, moribund language, but at least speaks Apple events correctly and has semi-decent documentation and community support. ... View more

Copy and paste the code into a plain text file and save it with a .jsx extension.   In AI, select the objects from which you want to remove rotation info.   To run the script, click File > Scripts > Other Script and choose the JSX file. The script should only take a moment to run.   If you need to run it frequently, copy the script to AI’s “Presets/Scripts” folder and relaunch AI. The script will then appear in the File > Scripts menu. ... View more

This should be a bit more thorough:   function removeRotationTag(item) { if (item.typename === "GroupItem") { removeAllRotationTags(item.pageItems) } else { try { item.tags.getByName("BBAccumRotation").remove() } catch (_) {} if (item.typename === "CompoundPathItem") { removeAllRotationTags(item.pathItems) } } } function removeAllRotationTags(items) { for (var i = 0; i < items.length; i++) { removeRotationTag(items[i]) } } removeAllRotationTags(app.activeDocument.selection)   ... View more

I am not sure what you’re asking. Are you talking about moving the numbers slightly left or right to visually equalize the negative space on each side?   I don’t think there’s anything in AI that would do that for you automatically (e.g. optical kerning only works inbetween characters; overhanging punctuation only works for punctuation). Aesthetic choices generally require human decisions. Once you’ve decided (e.g.) “1” needs to move 0.3pt right and “6” needs to move -0.2pt, you could encode those choices so the next time the same numbers come up the script moves them automatically. That’d be worth doing if the same numbers appear frequently. Although that is more than a two-minute job, if you’re looking for someone to develop it for you. ... View more

Amusingly, I just thought of a much simpler solution: just link your artworks’ text frames so that you paste the text directly into the first frame and it flows automatically from one to the next. As long as each text frame is the height of a single line of text and none of the lines overflow, you’re sorted.   (One advantage of a scripted approach is it can check for overflowing lines and adjust automatically, e.g. by tightening tracking and/or reducing horizontal scaling until each line fits the available width. That would require additional code to detect overflows and adjust character attributes according to your design guidelines, of course.) ... View more

For that you’ll need a script. Something like:   (function() { function copyLinesToFrames(frameName) { var textFrames = app.activeDocument.textFrames var values = textFrames.getByName(frameName).contents.replace(/^\s+|\s$/g, "").split(/[\r\n]/) for (var i = 0; i < values.length; i++) { textFrames.getByName(frameName+"_"+(i+1)).contents = values[i] } } copyLinesToFrames("names") copyLinesToFrames("ages") })()   This assumes two existing text frames, named “names” and “ages”, into which you’ll paste the source text.   The corresponding frames on the artworks must be named “names_1”, “names_2”, …, “names_7”, and “ages_1”, “ages_2”, …, “ages_7”.   Use the Layers palette to name your text frames. (If the Layers palette currently displays only layer names, click the menu at its top-right and change the Panel Options preferences to show all objects.)   The script assumes the number of lines of text wll always match the number of text frames on the artworks. I’ve not bothered with error checking; you can add that yourself if you need it. ... View more

You don’t say where your source text is coming from, but assuming a text file (e.g. Excel-exported CSV) this is the sort of work for which you’d typically use AI’s built-in Variable Data features. Here are some (free/paid) tutorials:   https://www.youtube.com/watch?v=mpwXj2E7YNA   https://www.lynda.com/Illustrator-tutorials/Welcome/450909/489328-4.html   https://helpx.adobe.com/illustrator/using/data-driven-graphics-templates-variables.html   https://www.christianda.com/blog-variable-data.html   ... View more

One other thought: I think Script Editor keeps a cache of previously-read terminology, and I wouldn’t be surprised if SD does as well. I think those caches may get flushed on restarting macOS, though don’t quote me on that. If not, quit SE and look in ~/Library/Caches, you’ll probably find a “com.apple.ScriptEditor2” which you can toss. (Of course, it’s possible there are caches elsewhere, possibly kept by macOS rather than individual apps. I’m not going to spelunk it right now.) If that still doesn’t work, I’d be having another look at AI to see if the problem lies there, but honestly this stuff is such a pain to troubleshoot. You might just want to revert to AI 25.0 for now and hope someone else figures it out before you next have to upgrade. ... View more

The reponse I received informed me that there is an issue with Illustrator's Info.plist file, specifically that it is missing its NSAppleScriptEnabled entry.   Pretty sure† Shane’s incorrect. The (confusingly named) NSAppleScriptEnabled flag enables macOS’s Cocoa Scripting framework to handle incoming Apple events (“AppleScript commands”). However, Illustrator already implements its own Apple event handling via the older Carbon/CoreServices Apple Event Manager APIs. Turning on Cocoa Scripting would very likely interfere with that, causing behavioral and/or terminology bugs.   If the application stores its “AppleScript terminology” in SDEF format then what should appear in application’s Info.plist is an OSAScriptingDefinition entry identifying the app’s .sdef file. If that entry is missing, macOS won’t be able to read the app’s SDEF and the app will appear “unscriptable” as far as Script Editor/Script Debugger are concerned.   However, that only applies to apps that have an .sdef file. Apps that date from MacOS9 days may still use the original AETE terminology format, which is stored in a .rsrc file. That’s what previous versions of AI use, up to and including 25.0. If Adobe have migrated from AETE to SDEF in AI 25.2 then I’m already grimacing as it’ll almost certain that Stuff Will Break as a result. (Perhaps explaining the woes in 25.1?)   -- † Actually 99.9% certain that Shane is completely—and dangerously—wrong. However, it’s been a few years since I’ve had to deep-dive the neglected shambles‡ that is Apple’s AppleScript/Apple event/OSA system. Badly documented APIs and unexpectedly entangled behaviors are par for the course, and it’s always possible some inexperienced engineer somewhere has subsequently done something to make the NSAppleScriptEnabled flag mandatory for all Apple event-aware apps, in which case unexpected fireworks will likely follow.   ‡ Apple disbanded its Mac Automation team in 2016 and fired the manager responsible after years of failures, so until/unless there are strong public indications that they’re investing in and rebuilding that platform following a decade of decay I think it sensible to approach it as abandonware and plan projects accordingly. I’ve already switched to JSX for new AI work: although Adobe’s JavaScript automation is also ancient and awful in lots of ways, at least Adobe is actively working to modernize it and eventually migrate everyone to UXP. ... View more

As others have said, your questions are really inappropriate for this (user-to-user) forum. Adobe CC is a professional product for professional working users and is priced accordingly.   As a high-school(?) student who has presumably just landed some freelance artworking to help pay your way (congratulations on that, btw), ask yourself if you really need Adobe Illustrator to do it, or if you could do the work using a free application such as Inkscape or a low-cost single-payment one such as Affinity Designer. If it’s for website artwork, the answer is very likely “yes”. If it’s for laser printing or plain old CMYK digital print, any decent graphics app should be able to export a PDF. (If it’s for anything more complex than that, then honestly you’re out of your depth.) Check with your customer (and their printer, if they have one) what file formats they can accept, and go ask on those apps’ forums if those apps are suitable for the job.   One more thing: I’m sure there must be artist-to-artist forums on the web where young amateurs wanting to break into the trade can find help and advice. Perhaps some of the experienced artworkers here can direct you to one of those, as it certainly sounds like you could use some help learning the business side. Good luck. ... View more

macOS natively uses POSIX file paths. Windows-style file paths will not work.   Read the JavaScript Tools Guide:   javascript-tools-guide.readthedocs.io/file-system-access/index.html   That explains how to construct paths that will work cross-platform. (The syntax is similar to that used in URLs.) ... View more

Not in ECMAScript 3 (JSX), which predates the dinosaurs. (Technically not impossible, but you’d have to write everything from scratch and it’s quicker just to go mad.)   There are several XLSX parsing libraries for Node.js if you don’t mind using Node/CEP (or UXP, once that ships). Just search on NPM.   However, expect to run into various gotchas when reading cells that use text styles, automatic number and date formatting, and other Excel features. What Excel displays on screen is not necessarily how that data is stored in the file. Not so much an issue when using Excel files in Excel as mathematical spreadsheets, as they were meant to be used. But a source of excitingly unanticipated bugs when reading XLSX files from customers who like to use them as ersatz databases. (e.g. Barcode numbers that start with “0”, which Excel will happily discard on cells that aren’t explicitly set to TEXT format.)   Otherwise, use Excel to export your spreadsheet file to .csv or .txt plaintext format, which bring their own sets of issues but are at least comparatively simple (relative to .xlsx) to parse and troubleshoot. ... View more

Of course it’s possible. Heck, you can run an HTTP server in a CEP panel if you really want (but if you don’t appreciate how dangerous this could be then you absolutely shouldn’t be doing it):       medium.com/adobetech/how-to-build-a-node-js-server-in-a-panel-ba1d63ea67e2   The practical questions are 1. is it cost-effective, and 2. does it violate Adobe’s licensing terms?   Regarding licensing restrictions, start here:       community.adobe.com/t5/illustrator/how-to-use-illustrator-as-a-server-in-automation-workflow/m-p/8703954#M30486   As for cost-effective, speaking from my own past experience building an intranet-hosted AI workflow server†: 1. probably not, and 2. see licensing restrictions.   As far as an implementation goes, you can either whip up something quick-n-dirty in a week to run on an old desktop machine on a nearby desk, or spend months developing a robust, fault-tolerant, load-balancing system to host on a shiny rack in your server room. The first (which is doable with modest web and AI scripting skills) will regularly fall over, requiring regular manual nursing (hence the nearby desk). The second (which demands experienced web and automation developers) can provide its users with a reliable web service, once you’ve identified all the different problems that AI will throw up (modal dialogs, scripting bugs, internal AI errors that occur with prolonged use) and written code to detect and recover from those failures automatically. Both, IME, are a pain.   So unless you’re developing it for in-house use in a large graphics studio where 1. there’s a high-volume of work, and 2. the automated work is too time-consuming for operators just to run scripts on their own desktops (which is the simple and obvious solution), it really isn’t an effective use of anyone’s time. (The automation system I built as a followup to that intranet-based workflow server ran as a standalone desktop app on individual users’ machines, driving their copy of AI directly. It was a much more satisfactory solution.)   That said, there is big money to be made selling and/or running customer-facing artwork automation servers (independent of whether or not they actually meet users’ needs). Kallik and Chili Publish are typical B2B products being sold to clients that wish to automate their own artwork production. (Kallik uses its own PDF composition engine; Chili runs atop InDesign Server.) And Moonpig is a classic example of variable artwork production (greetings cards, etc) being sold as a B2C product. (I don’t what Moonpig use but assume it’s custom-built PDF-based system.) But you’re not going to break into that field running atop AI because the standard licensing terms do not allow it; not unless you’re big and rich enough to negotiate a one-off licensing deal with Adobe enterprise. (And if you are that big and rich you’ll just license ID Server or, more likely, roll your own so you own your IP outright and aren’t dependent on a third-party vendor or their twitchy desktop apps.)   So, again, if you want specific advice then please describe what type of artwork you’re looking to produce and for what audience. Otherwise, my general advice is: don’t bother. It might sound good in theory but it’s a waste of time in practice; and whatever it is you’re trying to do there are better ways to achieve it.   --   † Linux VM running an HTTP-based job-queue service farming out work to several OS X VMs running AI. As a technology it was quite impressive. As an ROI, not so much. ... View more

For what purpose? Are you wanting to produce business cards over your company intranet, or are you trying to be the next Moonpig? If you want detailed advice you’ll have to tell us more about the work itself. Otherwise, some general pointers:   As Vasily says, 1. the AI license does not allow exposing AI as a public web service, and 2. running a desktop app like AI on a headless server is a pig to make reliable. Adobe will advise you to use InDesign Server which is designed for this sort of use, but whether that fits your needs and budget is for you to determine.   Alternatively, if your layout needs are very simple and limited to a small number of fixed templates, you might build a custom solution on top of an existing free/paid PDF or SVG generator library. That’s a common (if inflexible) approach for generating standard documents such as PDF invoices. Or, if you require more complex layouts but aren’t tied to AI specifically, you might build a custom solution on top of a Linux-friendly automatable app such as Scribus or Inkscape. ... View more

There are lots of free/paid barcode reader apps already available for Mac/Windows/Android/iPhone. I suggest throwing your barcode images at one of those—it’ll be a lot quicker and cheaper than writing your own. ... View more