Note: This is a public test instance of Red Hat Bugzilla. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback at bugzilla.redhat.com.

Bug 57302

Summary: Entering Norwegian characters in tcsh
Product: [Retired] Red Hat Linux Reporter: Thomas Skogestad <tskogest>
Component: tcshAssignee: Miloslav Trmač <mitr>
Status: CLOSED DUPLICATE QA Contact: David Lawrence <dkl>
Severity: medium Docs Contact:
Priority: medium    
Version: 7.2CC: pekkas, pto, teg, tobias
Target Milestone: ---   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2006-02-21 18:48:17 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Thomas Skogestad 2001-12-09 11:11:58 UTC
From Bugzilla Helper:
User-Agent: Mozilla/4.77 [en] (X11; U; Linux 2.4.16 i686)

Description of problem:
Entering Norwegian charcaters (fxe) can be difficult, especially f and e. 
x seems to work. "ff" gives only one f, or it might be a space and a f or
some other odd combination. f follow by x works (first I get a space when I
press f, but pressing x replaces the space with a f).

LANG and LC_TYPE are set to no_NO, but I haven't found any settings that
will give normal behavior.

Some other charcters also seem to be affected: 4 and (

tcsh works normally in RH 6.2 and 7.0, and also if I compile the most
recent release (6.11.00) from http://www.tcsh.org/page?MostRecentRelease


Version-Release number of selected component (if applicable):
6.10-6 (and also release 4 from RH 7.1)

How reproducible:
Always

Steps to Reproduce:
1. Use tcsh as shell
2. Make sure you have Norwegian language support
(I have the variables LANG and LC_TYPE are set to no_NO)
3. Press f, x and e in various orders, also try to combine with other keys.

Actual Results:  
f followed by o: <space>o
just f: <space>
f followed by f: 
<space>f, then the space is replaced by f

The results may vary according to the order the letters are entered.



Expected Results:  

The letters should have been entered normally.
f should give f and so on.

Additional info:

Comment 1 Thomas Skogestad 2001-12-09 17:30:56 UTC
There seems to have some kind of error in submitting the special Norwegian characters. fxe are 
represented as fxe. I don't know how this will look when I submit this comment, but I do _not_ mean 
"eff" "eks" "ee", but "ae" "oe" and "ae".

Comment 2 Trond Eivind Glomsrxd 2001-12-09 18:19:41 UTC
The characters can be produces with compose - "/o", "aa", "ae".

Comment 3 Peter Toft 2001-12-09 22:39:54 UTC
It seems that Danish people are also hit with this bug.
In Denmark we have three national characters fxe (Typical translation is
ae, oe, aa). 

When I type on of these national characters I get a space
Then I type the second national character I get both characters.
This pattern now repeats....

"f"  -> " " <-----!!
"ff" -> "ff"
"ffx" -> "ffx"
"ffxe" -> "ffx " <-----!!
"ffxee" -> "ffxee"


I use RH7.2 with tcsh-6.10-6.i386.rpm

The error can be cured by downgrading to tcsh 6.10-1 ->
ftp://rpmfind.net/linux/redhat/updates/7.0/en/os/i386/tcsh-6.10-1.i386.rpm



Comment 4 Tobias Ringstrom 2001-12-10 01:20:56 UTC
The problem is the patch tcsh-6.10.00-dspmbyte.patch which does the following:

   -	if (eq(pcp, dspmt[i].n)) {
   +	if (strncasecmp (pcp, dspmt[i].n, strlen (dspmt[i].n))) {

which is so wrong that it's embarrasing. The strings pcp (current locale) and
dspmt[i].n (locale to look for) are 16-bit strings in this case, and:

1. strncasecmp does not handle 16-bit strings
2. strlen does not handle 16-bit strings
3. The result code is zero if equal, i.e. the test is reversed, and multibyte
kanji input is enabled for everyone with a locale not beginning with 'j' or 'k'.
 Yuck!

Please remove this sad excuse for a patch, and release a new tcsh RPM via
up2date.  The bug is serious, since it prevents entering local characters for
many european countries, including Sweden, Norway and Denmark.

Comment 5 Tobias Ringstrom 2001-12-10 01:25:15 UTC
I forgot to mention that the workaround is to "unset dspmbyte".

Comment 6 Peter Toft 2001-12-10 05:40:45 UTC
I also support that a new tcsh-package is released. The bug is very annoying.
I am also 99% sure that the bug is also found in Red Hat 7.1 as well as 7.2.

Comment 7 diego.santacruz 2002-01-07 17:49:40 UTC
This also happens when trying to type any accented characters into a terminal
window running tcsh under the "normal" en_US locale. It is extremely annoying to
not to be able to rename a file that has an accent in its filename (I have to
switch to bash or use a GUI thing!). Please do fix this thing!

BTW, thanks to tobias.nu for the workaround, althought that is
not a permanent solution.

Comment 8 Peter Toft 2002-05-23 20:35:40 UTC
This is embarrasing!

Red Hat 7.3 was shipped with the same tcsh package which 
still is having the problems with european national letters like
fxe.

Please do something with this bug.

Comment 9 Tobias Ringstrom 2002-09-13 16:33:00 UTC
This bug is a duplicate of bug 41991

Comment 10 Tobias Ringstrom 2002-09-13 16:34:19 UTC
...which in turn is a duplicate of bug 41991

Comment 11 Pekka Savola 2002-09-23 07:57:12 UTC
I got hit by this nastie too, and created new RPMs for our local use.

Comment 12 Miloslav Trmač 2004-08-17 13:26:30 UTC

*** This bug has been marked as a duplicate of 41991 ***

Comment 13 Red Hat Bugzilla 2006-02-21 18:48:17 UTC
Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.