Has anybody had any luck installing the TextAutocomplete plug-in lately? I cannot get the Plug-in Manager to download it (it gives a download error). When I download it manually and try to install it, it doesn't really work within jEdit. I just wanted to see if anybody has had any luck with this plug-in.

Thank you,
CJ

Does jEdit support Right to left typing ? I could not find any support for it, and no plugin for this.

Kevin

Is there an Elixir edit mode? I see there is an Erlang mode, which is similar but not this same and is very primitive.

How do I set the default encoding for an edit mode?

For example: I want to create a new document in the current edit mode (CS-n) Erlang. The encoding for Erlang documents should automatically be UTF-8. This should be separate from any global default encoding.

However, I can find no way to set a default encoding for an edit mode separate from the global default encoding. This is so fundamentally obvious that I'm still looking for something I've missed.

I've been using this plug-in for remotely editing files and it works fine. The only trouble I have is that after I closed jEdit, I have to reopen each file I was working with.
Does anyone know a way to remember the opened files remotely as it happens with local files? Tks.

Hi folks,
I want to know if it is possible to remove or hide the scroll bars as it is possible with gutter bars.
I could not find any command for this.

There is *not* a lot of activity here.

When I search for font aliasing support under Linux, I find very diverse topics and messages everywhere and most of them are really old.

I will just say now that on my Ubuntu (Kubuntu) 16.04 system, in KDE, the actual configuration dialog looks good. But the actual text, as well as the interface.... even when using identical files (fonts) as other (KDE) programs, looks readily different and very ugly.

In general such things happen more in Linux (KDE) particularly when using some old GTK app (I think). Moreoever, some Linux distributions appear to have no good fonts at all; it seems and is sometimes said that only Ubuntu has good font support; my latest adventure into OpenSUSE was abysmal in that regard and full of problems.

For me a very complicated topic and I cannot see the forest for the trees.

Fonts that look AMAZING in e.g. "Kate" look abysmal in jEdit, and have very different font width, font spacing, etc.

I will post a screenshot later when I get down to it.

The difference is just outstanding and the minute anti-aliasing that I can detect in jEdit screenshots just doesn't cut it.

It is so bad that I even refuse to use jEdit even though it has been my favourite java editor since like forever. I now edit in console-mode Vim and only use jEdit for copy & paste operations, in general.

It is just horrid. This has been going on for such a long time in Linux, and when I see the instructions for jEdit of the past, I get sick in the stomach to begin with.

So I wonder if anyone can tell me in advance, before going through all that, if any of those instructions are still current?


I have attempted "http://jedit.9.x6.nabble.com/jEdit-fonts-on-Ubuntu-Linux-td1823304.html" but none of it did anything.

Meaning, this line:

exec java -Dawt.useSystemAAFontSettings=on -Dswing.aatext=true -jar /usr/share/jedit/jedit.jar org.gjt.sp.jedit.jEdit -reuseview "$@"

doesn't change anything.

So I don't know what to do. I haven't tested anything else like NetBeans.

I would like to use the jEdit XML plugin to validate XHTML5 using the RELAX NG schema from the Nu Html Checker (v.Nu):

https://github.com/validator/validator

I have downloaded the RELAX NG schema files from the v.Nu GitHub repo, and set the schema (Plugins > XML > Set Schema) for a sample XHTML5 document to refer to the XHTML5 RELAX NG schema.

As I expected, jEdit reports the following error:

> datatype library "http://whattf.org/datatype-draft" not recognized

due to the following line in the schema files:

datatypes w = "http://whattf.org/datatype-draft"

I know that I need to make available to jEdit - in its class path - a .jar that contains the datatype library, as described in:

http://www.thaiopensource.com/relaxng/pluggable-datatypes.html

but so far, I've been unsuccessful. I'm a newbie Java developer, so I am very likely missing something obvious, but I thought this would be relatively straightforward, because both jEdit and v.Nu use Jing, so I thought it would simply be a matter of copying the relevant files from v.Nu to jEdit.

I've taken the latest-release vnu.jar and stripped everything out of it except:

\META-INF\services\
org.relaxng.datatype.DatatypeLibraryFactory

that contains the lines:

com.thaiopensource.datatype.xsd.DatatypeLibraryFactoryImpl
nu.validator.datatype.Html5DatatypeLibraryFactory

and

\nu\validator\datatype\
Html5DatatypeException.class
Html5DatatypeLibrary.class
Html5DatatypeLibraryFactory.class

but I don't know how to add that .jar to the jEdit class path (I'm using jEdit 5.2.0 with XML plugin 2.8.7 - I know, not quite the latest - on Windows 7). I don't know how to tell jedit.exe to add the location of this .jar to its class path. Whatever I try - adding a classpath option to the jEdit.l4j.ini file, or defining a CLASSPATH environment variable - the jEdit activity.log reports:

Log: java.class.path=C:\Program Files\jEdit\jedit.jar

I even tried copying the contents of my "cut down" vnu.jar into jedit.jar, to no effect.

I'm stumped. How do I add the custom datatypes from v.Nu to jEdit?

Hi,
I'm using ver 5.1 on Ubuntu 14.10 using MATE ver 1.8.2 on a Dell Latitude 3450 laptop. I've noticed two problems that were not in Jedit 4.* but may be due to my setup.

1. A short-cut defined for CTRL+. (ie, control plus period) does not work. Other shortcuts with alphanumeric works. (Just noticed that "ALT+." works, so it must be me...???
2. An abbreviation with containing a non-alphanumeric does not work. The match algorithm stops at the non-alphanumeric. So, an abbreviation like "\bf" (w/o the quotes) does not work, but "bf" does get expanded.

If this is an issue with Jedit, I'd like to know how/if the above can be fixed. Obviously a work around is possible. If it is a problem with my laptop, any help from a guru would be appreciated.

Jim

Hi. How do I get rid of the vertical lines a that appear to mark a block of indented code? Thanks.

KP>

Haven't used jEdit in a couple months. This weekend I went to work on a project and whenever I'd try to use the console plugin, jEdit would hang.

I suspect it's due to something that changed with OS X 10.10.4. Any ideas?

Hi,
I use ubuntu 14.04 lts, jEdit version 5.1
$ java -version
java version "1.7.0_79"
OpenJDK Runtime Environment (IcedTea 2.5.5) (7u79-2.5.5-0ubuntu0.14.04.2)
OpenJDK Server VM (build 24.79-b02, mixed mode)

I have strange behavior of caret moves near the top and bottom of visible lines. Specifically, "go to first visible line" goes to the fourth from the top, and "go to last visible line" goes to the 4th from bottom. Further, "go to next line down", when within the top 4 visible lines jumps to the 4th line. It works correctly when the caret is positioned w/in the middleof the text.

Any one else see this?

I have used jEdit to create web code file in HTML for several years. I met this problem recently and do not know why. When I copy a paragraph and try to paste in jEdit, the whole paragraph become one long line. I paste the same thing to other text editor like Microsoft word or open office writer, it do not have this problem. I do not have this problem in jEdit before, I even can not remember when this problem started.


Please help.

When writing code for the Atari 2600 I use the compiler dasm.

With dasm, the opcodes are case-insensitive so LDA can be written as LDA, lda, Lda, etc. As such, I've set the mode file to use <RULES IGNORE_CASE="TRUE">.

I've configured the mode file so the 2600's registers such as GRP0, which holds sprite0's shape for the current scanline^*, get syntax highlighted. However, dasm is case-sensitive when it comes to variable names. As such, if I type grp0 into my source code the compile will fail.

As such, I'd like to be able to make some of the keywords case-sensitive even though most of them are not. Can this be done with jEdit? My current mode file, assembly-6502, can be found here. I'm planning to do an update soon and will be renaming it Atari 2600 Assembly as the mode file's not really for generic 6502 assembly language.

^* Yep, just the current scanline. When the Atari 2600 was developed memory was extremely expensive, so much so that the 2600 only contains 128 bytes of RAM. So TIA, the video chip, was designed to be scanline based. This means the registers in TIA only contain enough information to draw a single scanline, so your code must update TIA's registers on each and every scanline in order to draw the screen the player sees.

I've been using jEdit for Atari 2600 homebrew development since 2006. Spurred on by a fellow homebrewer, I recent delved into the plugins to figure out how to compile and run code from jEdit. I got it mostly working, though I ran into a problem with the environment variables. To figure out what was going on, I replaced my Custom Command for the Interpteter with an echo command and did some tests:

echo $f
/Users/darrellspice/Projects/Atari/Draconian/draconian.asm

echo $d
/Users/darrellspice/Projects/Atari/Draconian

echo $n
draconian.asm

echo $c
draconian

I know I can use multiple environment variables per command because my custom compile command works. The following compiles my source code (draconian.asm) and creates three output files(draconian.sym, draconian.lst and draconian.bin):
dasm $n -f3 -v0 -s$c.sym -l$c.lst -o$c.bin

For the interpreter step I needed to use the full path to the bin file, which is the source file with the extension .asm replaced with .bin. I thought that could be accomplished by $d/$c.bin but it fails to work.

echo $d/$c.bin
/Users/darrellspice/Projects/Atari/Draconian/.bin

The results I was expecting were:
/Users/darrellspice/Projects/Atari/Draconian/draconian.bin

So for some reason the second environment variable is being ignored. Any ideas? About jEdit shows:
jEdit 5.2pre1 server-background mode, using Oracle Corporation Java 1.7.0_67

In case you're wondering, Draconian is a port of Bosconian for the 2600. You can read about it, and even download a WIP ROM to run in Stella, over at AtariAge:
http://atariage.com/forums/topic/225044-draconian-2600/
http://stella.sourceforge.net

Thanks!

The default prompt (current working directory) is sometimes long and leaves no room on the command line. I would prefer to change the prompt to something minimal (like '$' on Linux or '>' on my Windows system). Seems there would be an easy way to do this, but I can't find it. Can anyone help?

Hello,

is it possible to save file in ISO-10646-UCS-2 (alias csUnicode) encoding?

Thanks

Hello.
I am currently using jEdit with my own defined mode file (mymode.xml). I'd like to customize the help menu. I've found the initial help page to be the welcome.html located under the doc directory.
But for the completeness of my IDE (based on jEdit), I would like to modify the contents structure of the help menu to include different html pages and a custom Contents. But I have not found where these files are located. Could you please help?
Best regards,
Luiz Razera

Howdy,

v5.1.0, OSX 10.9.2

Is there a way to switch between views via the keyboard? NOT buffers, views.

Don't see that in the menu anywhere, otherwise it could be bound to a shortcut key combo.

Using multiple monitors it seems the only way to go to the other screen is with the mouse.

I often have each view showing on a different monitor, and each view may show a different active buffer. When tabbing back into jEdit from another program, focus always switches to the primary monitor, even if secondary was active when the program last had focus. This means involving the mouse.

Any tips? Thanks!

Hi mode writers,

I am trying to implement a jedit mode for a tcl like language and am missing a good way to deal with line continuation.

The command syntax is roughly:

command -option1 arg1 -option2 arg2...

E.g.

create_variable -name Vg -dataset Data -function "<v:Data>"  
create_curve -name IVref($P) -dataset Data -axisX "Vg" -axisY "Id"  

I'd like to provide a bit context sensitivity in the mode, meaning each command should have its on set of valid options. So in above example

• create_variable should have the options -name, -dataset and -function

but

• create_curve should have the options -name, -dataset, -axisX and -axisY

I can acomplish this with

<EOL_SPAN_REGEXP TYPE="KEYWORD1" DELEGATE="CREATE_VARIABLE">create_variable(?=.*)</EOL_SPAN_REGEXP>

and dealing with the options in separate rules.

So far so good, but as the commands typically get pretty long, users typically use line continuation backslash just as common in tcl.

create_variable -name Vg -dataset Data \ 
	-function "<v:Data>"  
create_curve -name IVref($P) -dataset Data \ 
	-axisX "Vg" -axisY "Id"  

This of course immediately kills EOL_REGEXP. I tried various other approaches with SEQ... However, it is very hard to keep the right context for the options, without breaking anything else.

I have the feeling that a RULE option LINE_CONTINUATION_CHAR="\" that if present as last char on the line, concatenates the current line with the next line, before further parsing, would be the only way to go here.

Any hints, ideas appreciated

Best regards
Gergoe