As I understand, jEdit allows one Option Pane per plugin as evident in the dock position assignment screen.

It would be very cool to have multiple views for one plugin.

Example:
A CVS plugin could have these views
1. "CVS Repositories" - contains the list of repositories with checkout function
2. "CVS Resource History" - contains revision change details
3. "Synchronization" - list of files that are changed locally or remotely

Hello everybody,

I love the bookmarks. They are perfect when working with long files. But the setting of a bookmark changes the status of the buffer to CHANGED. I thing it's a little bit confusing. Is it necessary? I hope it isn't.

Petr

I am a new user, is there a feature that can automatically list the header file which includes the highlighted variable's definition? This can make reading C programs easier.

Hello,

I would like know your opinion about SWT.

JEdit will be more speedly if it use SWT. And more beautiful on Linux OS with GTK native look.

The "go to line" feature should not only go to the specified line, but also set the focus to the buffer's pane.

The way it currently is, if your focus is in a different pane, it stays there. So, for example:

Say you compile something, with the output going to the Console plug-in's pane. You scroll through that pane (thus setting the focus to it), looking for reports of syntax errors. It says there's an error on line 37. You hit CTRL-L (i.e. "go to line"), and tell it to go to line 37. The buffer's pane now shifts so that line 37 is displayed. You start typing the correction for the syntax error.

Unfortunately, the focus is still on the Console plug-in's pane, not the buffer's pane. So instead of fixing your syntax error, you're doing random things in the Console plug-in.

There should be a "Windows" menu, or something like it, that allows you to easily switch to a particular buffer with a couple keystrokes.

I know that you can do the "CTRL-PAGEUP" kind of stuff, but you're reliant upon the (seemingly random) order in which the buffers are stored. With a "Windows" menu, each open buffer would be assigned a number, and if you want to get to, say, buffer 3, you do an "ALT-W 3". See Ultraedit for an example.

Incidentally, why are the buffers stored in a seemingly random order, rather than in the order in which they were opened? In particular, I'm using the BufferTabs plugin, which may or may not have something to do with this.

When the file-choosing dialog box is open, you should be able to select a file (or files) and delete them using the "delete" key.

You can currently delete them by right-clicking and choosing the "delete" option, but the natural keyboard-driven equivalent is missing.

I know that there's a hex plugin. But it just doesn't cut it. All it seems to do is allow you to load a file so that its contents are displayed in hex.

You shouldn't have to load a file, as the plugin makes you do - you should be able to toggle the display of any buffer back and forth from normal mode to hex mode, with just a simple built-in menu option and/or hotkey.

You should be able to edit the file; the plugin doesn't seem to allow you to. You should be able to edit the file both via the left-side hex display and via the right-side ASCII display.

I really like jEdit's ability to narrow my view to certain level of indentation/folding. However I find this feature rather incomplete. When eg. designing a long form in HTML, you often have around 10 levels of indentation (imagine a table with few more tables in it [I know - tables for layout are evil :-)] and perhaps a javascript code in one of the table cells and you get the picture). This makes editing and navigating the code pretty difficult - especially if you use whitespace's plugin feature called "Remove trailing spaces". Every time you hit an empty line then, jEdit will jump to the begining of that line, leaving you with a whole screen of empty tabs and no code. You then have to grab your mouse and scroll the textarea using horizontal scrollbar. Now if the narrowing feature had an option to hide the leading tabs, it would improve the code readability dramatically.

I RTFM hard, but came to the conclusion, that this is currently not implemented in jEdit. In case I missed something, please tell me. Appart from this little detail, jEdit is really cool and I evangelise it among all my friends .-)

I've looked for this feature but haven't found it, but I would love it if I could mark pages as I'm working on updates for projects so I know what files I need to update on the live server. Would love to hear any suggestions too....

Hi,

I think moving the build system over to maven will make a lot of difference.

Im looking to see if there is a plugin, or feature that will tell me what Functions are in the file i have open as well as tell me which function im currently in.

Example text:

function Something()
{

# when typing in here it would tell me im in Something.


}

And then i would imagine there being some sort of drop down list that i can see the functions in the file, and be able to select one and be taken to the first line of that function.


Thanks for your help in advance!

I just found out about jEdit last night and I like it more than my previous editor; however, it's HTML syntax highlighting is lacking. So I endeavored to improve upon it. I got lists of all the elements and tags according to the HTML 4.01 specs and entered them as keywords, but problems occur distinguishing the two. For instance, "style" can be both a tag and an element.

I decided that the best approach would be to use regexp to highlight things according to their positions within a tag.

Element -- </({0,1}\w+?)
Attribute -- <\w+?(\s+?(\w+)=".+?")*>

[Note: I'm not sure that the question marks should be there, I just read in the Help Docs that "if a repeating operator is immediately followed by a ?, the repeating operator will stop at the smallest number of repetitions that can complete the rest of the match" so I hope I'm understanding that correctly.]

So the problem currently is that this doesn't really work for two reasons.

1) The SPAN_REGEXPs are set to highlight everything that matches them, but we need to specify stuff that needs to come before and after the attribute in order to be sure that we have an attribute. A solution could be to only highlight what is matched by the innermost parentheses when they exist. An option to toggle this behavior may also be needed in case someone needs to specify a repeating pattern but still highlight the entire regex.

2) Using regex under the main rules will include searching of things outside of the HTML tags. By making a ruleset to apply only to things inside HTML tags, you can get better results. But then, the 'greater than' and 'less than' signs cannot be used in the matching. So what I'm trying to suggest here is that the BEGIN and END sequences are included.

For a simple demonstration of what I'm trying to say, Consider this abbreivated mode file:

<SPAN TYPE="MARKUP" DELEGATE="TAGS">
    <BEGIN>&lt;</BEGIN>
    <END>&gt;</END>
</SPAN>

<RULES SET="TAGS" DEFAULT="MARKUP" ESCAPE="\">
    <KEYWORDS>
        <KEYWORD1>xxx</KEYWORD1>
    </KEYWORDS>
</RULES>

And this source code:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
</html>

Now if xxx were replaced with "<html" to ensure that only the HTML tag (and not the instance of "html" within the !DOCTYPE tag) is highlighted--it wouldn't work because in ignores the BEGIN and END marks. Of course, if this did work properly then the < sign would also be highlighted, which is not the enteded behavior. That's why this is just a simplified example of the behavior that I'm talking about and why using regex and highlighting the inner-most parenthesis would be a better way to handle this.

So with the way it currently behaves, you would have to replace xxx with "html" to see any type of result. In this case, the instance of "html" within the !DOCTYPE tag is also highlighted. This may not seem like terribly bad behavior, but it is when applied to "style" because you can have:

<STYLE type="text/css">
H1{ font-size: 10pt; }
</STYLE>

and:

<span style="font-size: 10pt">

Which clearly shows that "style" can be both a tag and an element, but it will have the same format applied regardless.



I hope I explained everything clearly here. I'll try to check back frequently to answer any questions.

When I type "proc" and then press C+b in PL-SQL editmode, it gets completed with "EDURE", forming the (correct) keyword "procEDURE". I know, I could type "PROC" and it would be "PROCEDURE" and if I wanted "procedure", I could change PL-SQL.xml accordingly. But why not check for the last letter typed, and add an uppercase version or an lowercase version of the substring.
Personally, I wouldn't replace the whole word with the text case used in the xml, because that would look like "parsing" (ask VB developers, they use this all the time) and then the case should also be changed for completely typed words.

Dirk

Hi, the comando file for intel fortran compiler ifort would be nice thing to have!!

It would be nice (in a code editor environment) to have the option that the complete command (C-b) does not consider the words inside comments (as defined for example in the mode XML file or by the highlight features).

I'd like to be able to configure the print footer such that it will display the total number of pages in the file (e.g. Page 1 of 100).

I would like to be able to open a file as readonly via a command line. I use the editor to load different reference files that I do not want to accidently change (e.g. phone listing). A command line option also enables me to define an icon for each reference file. An indicator of the readonly status in the status bar would be helpful.

I know I can create a script that will set the file readonly once I have it open, but I'd rather not have to remember to run the script each time I open the file.

Setting the properties for the file at the OS level is not an option.

As I was working with jEdit, I realized that the only purpose for another java editor I use was to batch-compile and run my programs.

So, I was wondering if anyone could make a plugin that would provide basic javac, javadoc, and java capabilities to projects... (sort of like JIndex, except actually compiling/running/javadocing the code)