Eclipse Finally Gets Code Folding

binarysearch writes "Code folding is finally in the Eclipse project! After more than two years open, Eclipse's Bug 9355 has finally been marked FIXED. Code Folding was the most-voted for bug in Eclipse, with support for J2SE 1.5 features in a close second. Check out the I20040504 Integration build for folding in the Compilation Unit Editor (Class File Editor support is in HEAD). For those who dislike the implementation, it is requested that you create a new bug, rather than reopening 9355."

Comments

Code folding is:

[Elwood P Dowd]

Code folding is where you can collapse a block of code between a set of squirrely braces so it reads"

{ ... }

or

// ...

or

/* ... */

Or whatever. Yes, this is like VS.NET and a million other IDEs. I use it all the time, but it never would have occured to me that it was a big deal. I figgered y'all FS/OSS types would use incremental search and brace matching to go anywhere in your code, rather than the scrolly-wheel. Isn't that what "real" programmers would do?

Re:Code folding is:

[Yokaze]

> Isn't that what "real" programmers would do?

Or what about keeping the code blocks small and concise? See Linux kernel coding style Chapter 4.
If you have to scroll some pages to find the corresponding closing tag, something has gone wrong.

But I've to admit, that I use code folding myself, which shows, I'm not a real programmer. But I also prefer Emacs to vi, which only confirms the fact I'm a quiche eater.

Re:Code folding is:

[abrotman]

iirc .. VS.NET, Dreamweaver, Homesite,bluefish,anjuta,eclipse(now), probaby kate, and god knows how many more have this .. its a pretty common feature ive seen in most "major" code editors(dunno about vi(use %) or emacs).

Re:Code folding is:

[Elwood P Dowd]

I'm sorry. What on Earth lead you to believe that I have some kind of a problem with FS/OSS development methodologies? I feel that, for good programmers, brace matching and incremental search provide all the "ease of use" that's necessary for living without code folding.

What sort of a response would I have been trolling for?

Re:Code folding is:

[dedazo]

vim does support folding, though it's a little quirky to get just right. Most editors based on Scintilla (which is very popular with the Python crowd) do as well.

Re:Code folding is:

[BRSloth]

Vi (as 6.2.x, at least) have code folding:

Add in your .vimrc:
set foldmethod = marker

And, on your code, do comments like this: /* add function {{{ */
void stuff(void)
{
weird_stuff();
} /* }}} */

use 'zc' to close the fold and 'zo' to open it. More info? :help fold

Re:Code folding is:

[FAT_VIRGIN]

That's VIM, not vi.

code folding?

[jpu8086]

this is a misnomer. i was lost for a while. i was under the assumption that code folding is akin to code wrapping (which eclipse has had since day 1)

i would rather term it code collapse...and which is what they call it in the main bug report. however, lots of people call it folding in the followup comments.

i guess i better get used to it being called code folding too.

so, i am happy that i can collapse 100s of lines of code into just the relevant 10-15 lines of code easily. however, i think this can create a problem of introducing silly error pass through (because you dont have the whole perspective on things). i think they have a feature to collapse all code that doesnt involve a variable "x". Anyone tested this yet?

Availability

[jtheory]

I'm guessing this is only available in the nightly builds for now, since I just hopped over to the site and there's no sign of a new milestone release (would be M9).

So... don't get excited -- the feature is *coded*, but you can't use it yet. That is, I haven't tried an Eclipse nightly before, but in general it's a bad idea if you're depending on the tool.

It looks like the latest integration build (a step up from a nightly) is still failing its tests.

In my experience, even some of the milestone builds have been a tad flaky (I put up with it because I want the features).

Anyone involved in the project know anything about when the next milestone release is planned?

Re:Availability

[Dr_Java]

The M9 Build is scheduled for Friday May 21, 2004 - its hidden away in the documentation. Full release of 3.0 is scheduled for July 2004.

Hint for Emacs Users

[AT]

I know this article is about eclispse, but I'd like to point out this feature exists in emacs, too.

Turn on hs-minor-mode (M-x hs-minor-mode) and code blocks can be folded and unfolded with shift-middle-click (or C-c @ C-c).

Feature requests, not bugs?

[Guspaz]

Wouldn't support for code-folding and J2SE 1.5 features be feature requests, and not bugs?

Re:Feature requests, not bugs?

[Phil John]

Yes and no. Typically a lot of oss (and non oss) have one central place to track bugs and feature requests, it just makes it simpler for devs to check one source for what to do next.

Re:Feature requests, not bugs?

[plumpy]

They've been planning the 3.0 release for over a year, so they're not going to hold it up to get Java 1.5 support in.

That said, 1.5 support is coming along and development versions can be installed as a plug-in (that link will also show you the current status).

Once it's complete, I'm sure it will be included in an Eclipse point release. 3.0.2 or 3.1 or whatever.

questionable features

[f00zbll]

although the feature is useful and many programs have them. For example several HTML editors have this feature, since many html pages have tons of tables. Having folding is great for documents that are long like HTML. For code I feel it tends to lead to bad habits.

For example, most of the .NET programmers I work with are x-VB guys and they use this alot. what ends up happening is they put all the classes in one file. Which actually leads to a tendency of coding with lots of inner classes. One of the biggest strengths of Java in my mind is it encourages programmers write smaller chunks of code that is more modular and re-usable. If you write your code in a well organized manner, you won't really need folding. Especially since the right pane in Eclipse shows you the methods and allows you to double click and jump to that point quickly.

I read the bug/feature requests and many of the arguments in my mind are personal bias. they aren't really valid. Several people complained Eclipse wasn't usable because it didn't have folding. I question that way of thinking and suggest the user is inflexible and unwilling to adapt. If it's that important, then write it yourself and submit it to eclipse. Needless whining serves no one any good. The time spent whining could be spent coding.

Re:Eclipse and C++

[Gill Bates]

Can anyone comment on how far the CDT has come in the last 12 months? Oh and also - is there any support for refactoring C++ yet :)

Not very far, AFAIKT. I used it a couple of months ago for some JNI/C++ code I was working on, and found that although it looks a lot like the Java editor, it behaves vastly different.

No code completion, no code formatting, no refactoring.

Just a basic code editor with color syntax highlighting.

Nope. Just dealing with certain realities.

[Prior Restraint]

Why would an effective java programmer generate methods with 100's of lines of code that need summarizing? Such methods are known to be error-prone and hard to read and understand.

That's what I keep saying to my coworkers. Nevertheless, we have methods over a thousand lines long, and our largest class is hovering around 25kloc (and I assure you, there's little reason for most of that code to be in the same class).

If nothing else, code folding is an effective tool for when you're forced to maintain source code written by idiots.