Unicode Autoconversion: wrong editor initialization

White-Gandalph · Feb 23, 2006#12006-02-23T08:12+00:00

Hallo,

i'm at the moment testing uedit 11.20b and have found the following annoying behaviour:

UTF8 coded HTML-Files are autoconverted to Unicode-16 in memory.
Cursor Movement by left/right arrow keys is then wrong, going 2 characters at once.

As it shows in the threads in this forum, the conversion is a wanted behaviour and not the point of problem.
The problem is the fact, that the rest of the environment gets somehow initialized by the file information that is gathered based on the original file format, and is not updated once the file format got changed in memory.

May be, the simpliest solution would be to delay the initialization of the editing environment until the unicode-autoconversion (which, as it looks, is hardcoded) is done.
Once the behaviour is corrected at this point, i bet the other problems with wrongly counted characters (as in the macro forum) go away, too.

Thanks for your interest.
Harry Boeck

Mofi · Feb 23, 2006#22006-02-23T13:45+00:00

White-Gandalph wrote:Cursor Movement by left/right arrow keys is then wrong, going 2 characters at once.

Yes, I also noticed this strange behaviour, reported it to IDM and the developers of IDM could already reproduce it. This bug will be fixed in next version. This bug exists only for Win9x and only for v11.20b. On Win2k and WinXP there is no problem with cursor left/right with Unicode files and UltraEdit versions pre-v11.20b (for example v11.20a) do not have this problem.

White-Gandalph wrote:The problem is the fact, that the rest of the environment gets somehow initialized by the file information that is gathered based on the original file format, and is not updated once the file format got changed in memory.

I don't really understand what you mean here. How the file is handled in memory should never have an influence on editing or saved format after editing. There is this 1 bug described above at v11.20b of UltraEdit on Win9x, but this is not a general problem.

White-Gandalph wrote:Once the behaviour is corrected at this point, i bet the other problems with wrongly counted characters (as in the macro forum) go away, too.

Yes, the number of selected bytes (not characters!) and the file size (in bytes, not characters) in the status bar is filled with the data from the temporary file format. But to handle Unicode files easily, the various Unicode formats must be converted for editing to a format which can handle all Unicode formats (UE uses UTF-16 LE).

It would be really an extrem performance loss to convert the temporary Unicode format to original format on every update of the status bar just for showing the correct byte values there. A simply division by 2 for UTF-8 is not correct, because if a character must be saved with 2 bytes in UTF-8, this produces again a wrong number (file size or bytes selected) in the status bar.

White-Gandalph · Feb 23, 2006#32006-02-23T17:23+00:00

Thanks for this hints.

I'm currently looking for a replacement for this editor version.
My old one from the 90th (a 7.x version) is very outdated,
but the current one is simply unusable on Win98.

Unfortunlly, I'm not seeing any archive for ultra edit on the IDM site.
When i google for ultra edit versions, i get nothing than cracker sites,
which i don't trust. (That's really not good for the reputation of IDM to let people go into those networks when they simply want to get an older version of the editor!)

If you could post a link to an archive of IDM, i would greatly appreciate it.

Edited 2006-02-23 13:10 GMT:

Well, a downgrade seems not to be a solution.
My browser (firefox) seem to have problems with those files, too.
When i convert such a file into ISO-8859-1 with Ultra Edit, despite the fact that the browser should use this encoding for display with autodetection turned off (as it is now in Ultra Edit to fix the cursor movement problem), it insists on wrongly displaying it in UTF8 as it does correctly with the original file version.

I have no clue about how to fix this otherways than using pure english files as in the old days of dos!

Thanks for the help.
Harry Boeck

Mofi · Feb 23, 2006#42006-02-23T21:13+00:00

I have asked IDM support by email for a hotfix or for v11.20a. A hotfix is currently not available, but I got the English setup zip-file for v11.20a for my Win98 by email. I don't use the German version!

You can also request the v11.20a setup zip-file from IDM support or I can sent English version by email to you.

Why do you use UTF-8 for HTML? I hope not because of the German umlauts and the ß. There are HTML entities for these characters which can be inserted automatically when you press ä or Ä, ... Using the HTML entities for the German umlauts solves the problem with UTF-8 because all characters are then ASCII characters.

See my macro posted at Auto-completing brackets and quotations.