Is it possible to split a string within a Rich Text Label based on Line Count?

KiwiCaveGremlin

Hey all!

As the title says, I've run into an issue in my game and I'm not quite sure how to proceed, or if there is an easy solution adept godot users are aware of.

The problem is thus, I have multiple RichTextLabels that make up a book, and one relatively large string that makes up the contents of the pages.

What I would like to do, is dynamically split the string so that once RichTextLabel box #1 has reached it's maximum capacity (given here as 13 lines) we split the string there, and then push the rest to RichTextLabel box#2, and continue that process for as long as there is text.

Safe to say my plan, thus far, has NOT worked out so well!

The number of characters that can fit into a Word_Smart Autowrap text box can vary, the RichTextLabel function "get_visible_line_count" doesn't seem super reliable, and even when it does work I end up with an int which is what it says on the tin, but I don't know if there's any way to interact with said line count, such as a "split string at x lines" even if I could get it to work reliably.

My plan since then has been a "simple" approach as I can see it, set a constant for the max number of characters, read the string from that point and walk backwards until I reach a space character (" ") and split there. Progress is slow, on general account of myself not being adept, but if any of ye wandering souls new of a better approach to this or something Godot already has that does this better, I'd love to hear it!

xyz

KiwiCaveGremlin

put the text into the label
await for it to render
iterate from 0 to get_total_character_count()
for each character check get_character_line()
when it reaches get_visible_line_count() break iteration and split the text into two strings at that point: visible and remainder
use RegEx to check if there are any opened tags left in visible. If yes, append closing of those tags to the end of visible and also append opening of those tags (in reverse order) to the start of remainder
save visible as a page in your book
repeat all above steps for remainder until you exhaust all of the text

KiwiCaveGremlin

xyz This seems like a clean version of what I am attempting! Thanks for the pseudo code!

I'll do my best to get it running, and post my implementation here should it all work out! Thanks for taking the time out yer day, have a good one!

KiwiCaveGremlin

The method I ended up using was a mostly inferior version, on account of me wanting this text sorted prior to rendering

public void WholeChunks(string str, int chunkSize)
{

    int stringLength = str.Length;

    // Calculate chunkSize here, take the maxCharacters, first reduced by x (where x is the number of line break counts within the string capped to maxCharacters)
    // then reduced by a variable number equal to the distance to the closest " " character or ".", to ensure no words are cut in half.

    for (int b = 0; b < stringLength; b += chunkSize)
    {
        //Will need a variable chunk size rather than one passed in
        if (b + chunkSize > stringLength) chunkSize = stringLength - b;

        pageText.Add(str.Substring(b, chunkSize));

    }



}

xyz

KiwiCaveGremlin I was really hoping you'd implement a full generalized solution following my little "algorithm" there. It could even be turned into an add on. Of course if this minimal solution is good enough for your use case - that's perfectly fine.

KiwiCaveGremlin

xyz

I very well come back to this though at a later point in my project, as my current solution "works" but following your logic and psuedocode would result in a much better result I think for the long term

Tomcat

xyz I was really hoping you'd implement a full generalized solution following my little "algorithm" there. It could even be turned into an add on.

It reminds me of a layout program for printed books. It would also be good to include an algorithm to avoid “orphan” and “hanging” lines (when one line from a paragraph on another page appears on the page). This is achieved by tracking (changing the distance between letters).

6monkey9

I this currently the only solution? When puting my 5mb text file into the label it absolutly breaks the game lol

DaveTheCoder

6monkey9 the only solution

What's wrong with that solution?

6monkey9

DaveTheCoder when putting large amounts of text inside a label godot explodes

xyz

6monkey9 Break the text into smaller parts. In fact in each iteration you only need to use the amount of text that's guaranteed to overflow the label.

6monkey9

you mean manually? it's literally the entire bible, 5mb of a text file.

xyz

6monkey9 No I don't mean manually. Iterate through whole text stored in memory or in a file and process piece by piece. The label just serves as a measuring cup. No need whatsoever to put the whole text into it.

6monkey9

xyz ooh I see, will try that. ty

6monkey9

6monkey9 just realized this is not a viable method at all, the await makes it extremly slow... will have to look for another solution, will try to post it here if I found something

xyz

6monkey9 You don't have to do it at runtime. You can pre-process the text once and store its structure for the actual realtime use.

6monkey9

xyz okay i'm dumb, will try this. with due respect, still a shitty method bc epub readers can dynamically change in miliseconds how text fits the pages acording to different parameters like font, line space etc... but it works for now I guess

xyz

6monkey9 Godot is a game engine, not an ebook reader. It's not optimized for rendering large amounts of text, especially not rich text.

Also, the Bible consists of many separate books and chapters that can be processed separately. Working with a single big string that stores all of it is completely unnecessary and very impractical for myriad of reasons.

And note that this thread was about splitting the text into multiple boxes that are displayed at the same time. If you only have one box (as ebook readers do) - you don't need any of this.

6monkey9

xyz Well there is a difference from "not optimized for rendering large amounts of text" to exploding when trying to put large amounts of text.

xyz

6monkey9 Instead of vaguely complaining in circles, why not describe what exactly you're trying to do. There's probably a viable approach other than bruteforce pasting megabaytes of text into a label. But unless we know the exact use case, we can't suggest optimal solutions.

6monkey9

xyz Yes you are right. Basically just trying to display a large book in a 2 page book format. So player can turn pages and read the book. Your implementation works for me, so just leaving this if someone want to make a more advanced implementation... maybe it's helpful.

What I tried:

Entire text into the labels and using the line range settings to split in two pages. Works for small text, but when pasting an entire text book in a label it makes the engine explode.
Using xyz's function to presplit the whole book string into an array of page strings based on some of the label parameters the text will be displayed at. It works perfectly for me, but for bigger implementations it's not viable bc it takes a lot for the function to do the spliting. Not to mention for dynamic font size, etc...

Again, your pseudocode was usful, so thanks a ton. But just having this itch that for bigger applications godot is flawed here. (Just trying to see where the current label node could be improved)