I would imagine that most use cases for PDFs would just expect users to know what to do with PDFs and most computers have some app associated with them.
What about the report writer in Xojo – is that any good at all? Another functionality I haven’t explored yet. It’s been around since at least 2014 (the date of a YouTube tutorial on the topic), so one would THINK it’s pretty mature by now.
Agreed. I send Björn an email pointing to this thread.
Only if it had actually been well maintained and developed. But this is Xojo Inc we are talking about.
And I will never understand why Xojo didn’t buy Bob’s Shorts … that was soooo much better it isn’t even funny.
Not sure if pdfium can generate a PNG as I only converted the parts I needed to get the pixels into a Xojo picture.
Converting this code into a more friendly Xojo class wouldn’t be difficult but not something I’m interested in doing as I have no need for it (I use DynaPDF).
I’m also sure Björn could wrap this into a plugin very easily but that would mean a non-free solution.
uh… and why does this not fit?
Dim f As folderitem = SpecialFolder.Desktop.child("test..pdf")
HTMLViewer1.LoadPage(f)
seems to view if to me
Specifically what are the pain points?
What is Bob’s Shorts?
Besides what DaveS wrote, there is also this solution, where you would also use the HTMLViewer to display the PDF:
He sold it to a company that promised to “bring it to the next level” but I don’t think they have maintained it at all:
My guess is that the move to API2 made them reconsider if they actually want to update it …
Same here - but I am on an old Mac on macOS Mojave. According to posts I read on TOF that does not work too well on newer browsers / systems … but I have no first hand experience on that.
There is actually an open source project:
Just tried … doesn’t seem to do anything. I select a pdf file, it shows it loaded the file, and then complains that it can’t find the file … 
well I used Xojo 19r1.1, on Monterey with Safari 16.1
Doesn’t seem to do drag&drop … any ideas?
3 of them.
Not really what I had in mind. Although Hybrid kinda-sorta is.
I’m talking about a single project targeting desktop, mobile and web which is beyond what even Xojo claims to be capable of.
The Hybrid project type you mention means creating a web app and then rendering it on mobile and desktop in a browser control of some kind (it can also mean a mix of server and client models). I think that web UI on desktop and mobile is the approach Omnis takes also, internally. And probably others. I prefer a system that renders a full native app on desktop and mobile and does the best it can at reproducing that for web. The jury is out for me on whether each device must use native controls or favor pixel-precise identical visuals and possibly universal behaviors that are nonstandard on some devices. I lean toward the latter.
It is possible that it’s impossible to build a framework that provides that rather big “ask” without too many compromises (slow performance, platform-specific UI compromises from purist user perspective, the need to still support platform-specific features or feature implementations, etc). It has been tried, with varying levels of success, but I don’t think anyone has hit a home run yet. Maybe inherently they never can.
Only language I know is capable to do it: JavaFX. For Desktop, For Mobile (compiled to native code) and for Web with Jpro you can have one Code for all platforms and also one UI for all platforms (What I would not prefer for Desktop you may use Menus and for Web and Mobile not). I don’t know any language else with this capability.
There are others at least making pretense to it – Avalonia and Uno on the .NET side for example. I just see a big learning curve plus some people complaining about, e.g., performance issues or claiming that some bright shiny new more officially sanctioned direction such as .NET MAUI or Blazor rendered in browser controls on devices is the Way of the Future.
It seems to me like the bar to entry into this circus is too low anymore. Anyone here could start an impressive-looking open source project and get some participation and then just have it peter out after a year or two without reaching v 1.0 or getting crushed by some more commercial effort. I think people underestimate the difficulty of what I want, which is:
I haven’t figured out if this would have been decisively done by now by the likes of MSFT or Google if it were doable, or if it’s just better for their business model to keep it more balkanized / complicated with multiple toolchains and technology stacks.
It is a recurring theme over the decades I’ve been active – 4GLs were supposed to be the Second Coming of Development in the 90’s, for instance, but they died with client/server-oriented patterns and the rise of web apps.
Anyway I will have to break down and look at JavaFx if it has truly cracked this particular nut.
JavaFX for Desktop, with Package you get DMG, MSI and deb files for installing, generate them with GitHub actions you even don’t need all platforms, if you have questions ask me I am using it in thes wise.
JavaFX for Mobile: use GluonMobile, edit your Pom File and setup the mobile target without the Gluon Charme plugin you can use even the UI for your Desktop App but I would then workout the Desktop App looking like a mobile and Web app.
Jpro.One lets run your Application (JavaFX Application) as a WebApp.
Definitely works.
What IDE are you using? JetBrains? Something else?
IntellöiJIdea from Jetbrains, I am building mostly dxml Projects so I use also Scenebuilder from Gluon. JavaFX Design you can setup with CSS in your needs like you want to. Will work also with the IntelliJIdea Community Edition and for testing all of them are free (IntelliJ, Gluon and also Jpro.
I actually have the whole-enchilada license from JetBrains so I have the non-community edition of Intelli-J Idea, so that works for me.
(I had a fit of ambition last year and wanted Ryder for macOS and the all-in license wasn’t all that much more … I have actively used Ryder, am mostly happy with it especially given the immaturity of VS for macOS, but then I tried their Python IDE, got hung up getting it to find the Python interpreter on macOS, got lost on hold w/tech support, gave up. Never tried the Java IDE or anything else but probably will renew for at least 1 more year).