I need a tool to programmatically convert epub files to a series of images. The output should look like screenshots taken on a canonical device (for this application, an iPad). I haven't been able to find any tools that do something like this.
So what I'd really like (1) is a tool that does that. But assuming that I'm correct that no such tool exists, is there (2) a library (preferably a Perl module, but I'm not that picky) that will read and render ePub?
Obviously, rolling my own I could combine tools for unzipping, reading html, reading xml, putting everything in the right order, and rendering html within certain constraints. Though I'd rather not do that, and if that's the only option I'll have to go on to look for a tool that will do the last part of that or I'll have to create that too.
Any leads on (1), or failing that (2)?
Apologies if what I'm about to type is just crazy-talk on my part--in fact, I'm pretty sure it is--but perhaps something like this might work and I'm kind of interested in knowing how well it might work for you:
Use Frank (https://github.com/moredip/Frank) to control the iOS Simulator on a Mac. Program it to open up the EPUB docs you need.
All you need then is something to automate the taking of the screen shots. Obviously, these will look like the EPUBs are being rendered in an iPad (or an iPhone if you wish--the iOS Simulator does both, of course).
Automating the screenshots can probably be done with AppleScript, although the hard part might be getting it to talk to Frank. Worst case, you can tell Frank to pause for 5 seconds after it loads each page and tell AppleScript to take a screen shot every five seconds. That sucks, but if you're desperate, it will get it done. It's also possible Frank can somehow make the screenshots happen--I haven't used it enough to know.
Pandoc can convert from EPub to LaTeX (and therefore to PDF) or to any number of other formats. Conceptually this should be a type (1) solution.
depends on your definition of "look like" - do you want the user-chrome or just the epub rendering for a given screen size.
I would check out the various epub readers for your platform of choice, size the window to your preferred dimensions, and then just "print" the epub to a virtual printer that outputs to image files - on windoze I use imageprint.
You could easily make a "frame" from an iPad product shot and place your screenshots within that - only thing missing would be as I said the user chrome.
Related
I have a Default.png which includes a version number on it. Every time I update my app, I have to change it both in the lite and full version's default.png and default#2x.png. Hassle, no?
I'm pretty sure I've been going about this the wrong way. What should I do instead? (I would like to show a version number on launch, not just nix it altogether.)
Compile-Time Image Compositing
If your logo doesn't need to change other than the version number, then you can use your graphics library of choice at compile-time to refactor the png. Pseudo code below:
Pseudo-Code:
UpdateLogo(String logoName, String version)
{
WidgetImage MyLogo(logoName + ".png");
MyLogo.DrawText(800, 650, version);
MyLogo.Write(logoName + "Final.png");
}
UpdateLogo("Logo.png", "Version 1.0.0");
Compile that program and keep it around as a custom build tool. Then whenever you need to build your application you can compile Logo.png into LogoFinal.png. If you need help using XCode or other tools to generate image files I suggest you search for image manipulation tools separately from "dynamic versioning".
Ideally your version string will use constants defined in an easily-editable table or controlled by your build system. At the very least it will save you from opening up Photoshop every time you need to build your app.
For Display in a Running Application
You should be using a font to draw the version number on top of the logo. Then you can just include a resource file that is text-based and can be easily updated by automated tools for each build.
Sources
Can you create custom build rules for XCode based on file type?
Apple's Human Interface Guidelines say that the Default.png shouldn't be used as a splash screen; it should represent all of the UI controls the application will show, but without any localizable text or content. (Think of how the the built-in apps like iPod and Contacts behave.)
If you're doing it for a client and they demand it, you can always use the "But the app store might reject it for violating their terms!" argument.
Of course, this doesn't apply if you're not submitting to the Store or if you just don't care. :)
A technical add-on for the people posting above: make sure that any png compositing you're adding to the build process runs before pngcrush executes, so that you're not replacing an optimized image with a script-generated (and likely unoptimized) one. You may also run into weird issues if you try doing it after pngcrush runs (it not displaying), anyway.
I have been looking for a way to present an interactive pdf file (created by in-design) on
the iPhone. I read a bunch of questions here but none says how to do it. The pdf file contain the text and in the middle it contains a 3d module, but when I present it on the iPhone it shows only the text and an empty white box where the module should appear.
Is it even possible to do it?
I'll be glad for any assistant on this subject or even where to look.
Thanks in advance,
Shahar.
Apple's PDF parser does not support 3D stuff. You're better of implementing the 3D part yourself and just adding that as a UIView on top of the PDF. There are several PDF frameworks that help with that (see https://stackoverflow.com/questions/3801358/pdf-parsing-library-for-ios)
Another alternative might be licensing Adobe's iOS rendering engine. But I doubt that they already added 3D support (or that they will be). Also, from what my sources tell me, pricing is rather high and apparently the framework not very developer friendly. (But I haven't used it myself)
In my iPhone / iPad application, I show a person's medical reports in the form of a PDF. I have saved the reports in the documents directory and am reading them from there.
I want the user to be able to add or edit comments on these PDFs, as well as be able to highlight certain sections in the PDF. After editing, the application should be able to save the PDF back into the documents directory.
Is this possible within an iOS application? If so, how? Is this a task for Core Graphics?
Editing PDF directly on iPad/iPhone is a rather big job because the standard API only supports showing it (and only a bit more.) If you want to do anything more, you need to invest a huge amount of time to implement generic pdf handling code.
There is an open-source library handling these, e.g. this one. I don't know if it fits your needs, though.
A better idea, in my opinion, is to create a native UI showing the data contained in the PDF file using the standard Cocoa-Touch UIKit and create the PDF once the user is done with it so that the user can export it back. That way, you don't have to write a complicated PDF handling code.
In any case, it's not a good idea to show generic PDF on iPhone, because the screen size is so small (iPad is a different question, especially if you expect the user to be familiar with the particular format of your pdf.). A dedicated UI would be much better.
I was recently asked if I could code a simple IPhone/IPad app that:
Checks an FTP server for changes to a PDF file (easy)
Downloads the lastest PDF (easy)
Adds or replace the file on iBooks (hmm...)
I have tried to find any code that deals with inserting/adding/updating files inside iBooks, but sadly there is very little on the subject. Is this even possible except manually through iTunes? My gut feeling says that it's just a matter of writing the file to a folder..
It makes sense if it were such an API. There are many companies that would like to keep their product portfolios up to date, or research manuals that (in theory at least) could benefit greatly from being updated automatically through an app.
Any help or comments are welcome.
My gut feeling says that it's just a
matter of writing the file to a
folder..
Unfortunately, this probably isn't the case. I'd imagine iBooks using some sort of database to keep track of books, rather than simple files.
There is a URL schema for iBooks, but it's undocumented, and nobody has found any methods that would update/replace a book (I don't know if any even exist): How do I launch iBooks e-reader programmatically on iPad?
If you feel this is something that would be beneficial to iBooks, you should consider filing a feature request on the Apple dev site.
UIDocumentInteractionController is your friend in this case
Alright, what I need is a command-line application that allows you to take a screenshot of a file's audio stream.
For example it should be run like this:
app.exe "C:/artist-title.mp3" "C:/mp3Stream.jpg"
app.exe "C:/artist-title.wav" "C:/wavStream.jpg"
It only has to be able to capture mp3 streams, other streams are a bonus.
Preferably all audio channels are listed in the image, but if all channels are combined into one mono stream it would work just as good for me.
So, is there such a application out there? So that I don't re-invent the wheel.
If not does anyone have tips on how I should go about writing such a application myself? Preferably in Java. I can handle programming pretty well but I'm not exactly an expert on the MP3/WAV formats.
Why do I need it...? Well, it's more fun to link to a file online with some sort of preview image besides the link. It gives you a hint of the audio character before you listen to it (is it loud? does it look like "bit music"? does it have any parts that are more quiet than others? etc).
Never mind, I wrote my own little application in Java.
It was a piece of cake once I found this excellent guide:
http://codeidol.com/java/swing/Audio/Build-an-Audio-Waveform-Display/
Although you can't download the source from that page (as far as I can tell, though he makes it apparent that you should be able to) he does provide some very useful key lines of code that makes it easy to puzzle together the application.
Adding a little bit of help (easy stuff). You can get a graphics object from doing so:
BufferedImage img = new BufferedImage(500, 100, BufferedImage.TYPE_INT_RGB);
Graphics2D gfx = (Graphics2D) img.getGraphics();
And once you have drawn everything you need on the gfx you can save it to disk just by one line:
ImageIO.write(img, "jpg", new File("waveform.png"));
It's hard to get it to look very good though. Doesn't look as nice as for example Audacity. Guess they have spent more time on it than a few hours though.
The biggest pain about this is however that Java don't support MP3 import. They really should get around to that.
So to get the waveform of MP3s I first convert them into WAV using "javazoom.jl.decoder.Decoder.java", it's on their website. Very easy to use, just give the input path and the output path and it's done.
javazoom dot net (couldn't post more than one "hyperlink" on this website)
The big downside of this is of course that a huuge WAV file has to be created, and woe be unto thee if the MP3 happens to be 15 minutes or so... The WAV will be over 100 MiB (maybe even 200 MiB, haven't found out since I got a Java-out-of-memory-error, even though I gave the VM 512mb).
MP3 support in Java today please. Guess the reason they don't have it is because of copyright issues. Copyright really is slowing man down.
Also take a look at http://www.jsresources.org/
It provides a pretty good FAQ section about everything Audio in Java, and some example applications.