.

Even as a stop-gap feature request, I'd like to have jEdit *warn* me before *attempting* to open a file over x KB/MB/GB, based on a setting somewhere in Global Options that specifies x. (Turned off by by default?)

There have been times when I've initiated the Open File function only to realize that the file (commonly a log file) is a mammoth beast, and I get that sinking titantic feeling as jEdit submerges into the unreachable depths, foolishly reeling in a file manifold larger than itself. And sometimes I have to cut the rope and kill the jEdit process, along with any unsaved buffers.

This simple "warning" feature would be a one-up compared to other text editors... most that I know of at least. Should be easy to implement, right? User interface dialog reads "Warning, this is a large file and may take a long time to load. Do you wish to continue? Yes/No"


For the 2GB and larger files, which text editors can cope with these, and how so? vi, emacs? Any Windows flavors? Could jEdit borrow the same sort of logic for handling these? Read just part of the file into memory at a time?

-Paul

There is a Unix app for working with large files call 'less'. From 'man less':

------------------------
less - opposite of more

Less is a program similar to more (1), but which allows backward move-
ment in the file as well as forward movement. Also, less does not have
to read the entire input file before starting, so with large input
files it starts up faster than text editors like vi
---------------------

I've used less to work with large files and it is very responsive. Could JEdit implement a reduced set API mode for situations where it is not practical to pull the entire document into memory?

Eric Everman

Maybe because, result of grap is very large too. I'm still with this problem and my files is very little than yours. Other problema is why 50mb file use 140mb of heap in JEdit ?
I'm really don't understand.

Welcome to Unicode. When jEdit opens a file, it changes the coding from whatever was used to store the file, e.g. UTF-8, ISO Latin-1, etc., normally a single-byte character set to Unicode which is a double-byte character set. Then add the overhead for Java and jEdit (25 MB on my setup), plus plugins.

Is there a way to determine what encoding a current buffer or file is in? For example, I opened a file in euc-kr by selecting euc-kr instead of autodetect. Now in jedit is this in utf-8?? If I save the file will be it euc-kr or utf-8?? If I save as... to a new file will the new file be utf-8?

Hi,
Really, this make sense. You know any workaround for specify a native caracter code ?
PS: In global options->General have an option calling "Default caracter encoding", this one have any connection with the problem ?

The option only determines how the the file is translated to/from Unicode. The use of Unicode is part of Java, and I doubt that there is any easy way to get around it.