130820 - Small latin letter n-acute instead of o-acute

Status: VERIFIED WORKSFORME

(Core :: DOM: Editor, defect)

Description

[Martin Broz]

From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.9) Gecko/20020310
BuildID:    2002031008 (0.9.9 and previous 0.9.8, 0.9.7)

For Messanger, Composer, ... when writing in ISO8859-2 letters.

Small latin letter n-with-acute (0xf1 - unicode 0x0144) is displayed instead of
Small latin letter o-with-acute (0xf3 - unicode 0x00f3).

Reproducible: Always
Steps to Reproduce:
1.Character coding of X-server and Mozilla set to ISO8859-2
2.Write a o-with-acute letter (combination of keys dead-acute + o)
3.

Actual Results:  Displays n-with-acute letter

Expected Results:  Should display o-with-acute letter

Other applications and libraries are OK (GIMP, GTK+, Netscape, xterm,...). This
behaviour appeares only in Mozilla

Comment 1

[kinmoz]

--> ftang

Frank can you guys look into this one?

Comment 2

[Frank Tang]

give to katakai

Comment 3

[Masaki Katakai]

Hi Martin,

Thank you for filing bug report.

I have some questions,

1. try 'xev' to find what the exact keysym generates at dead-acute + o?
   keysym 0xf3, oacute is generated?

2. What is the locale when you start Mozilla?
   try 'locale' command on terminal

Comment 4

[Martin Broz]

There is my response for events on a xev window and locales.
I hope it can help you.
Thanks
Martin

==================================================================
KeyPress event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 4756341, (92,-11), root:(1056,10),
   state 0x2000, keycode 21 (keysym 0xfe51, dead_acute), same_screen YES,
   XLookupString gives 1 characters:  "´"

KeyRelease event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 4756470, (92,-11), root:(1056,10),
   state 0x2000, keycode 21 (keysym 0xfe51, dead_acute), same_screen YES,
   XLookupString gives 1 characters:  "´"

KeyPress event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 4756967, (92,-11), root:(1056,10),
   state 0x2000, keycode 32 (keysym 0x6f, o), same_screen YES,
   XLookupString gives 1 characters:  "o"

KeyRelease event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 4757107, (92,-11), root:(1056,10),
   state 0x2000, keycode 32 (keysym 0x6f, o), same_screen YES,
   XLookupString gives 1 characters:  "o"

===================================================================

LANG=cs_CZ
LC_CTYPE="cs_CZ"
LC_NUMERIC="cs_CZ"
LC_TIME="cs_CZ"
LC_COLLATE="cs_CZ"
LC_MONETARY="cs_CZ"
LC_MESSAGES="cs_CZ"
LC_PAPER="cs_CZ"
LC_NAME="cs_CZ"
LC_ADDRESS="cs_CZ"
LC_TELEPHONE="cs_CZ"
LC_MEASUREMENT="cs_CZ"
LC_IDENTIFICATION="cs_CZ"
LC_ALL=cs_CZ

===================================================================

Comment 5

[Martin Broz]

Another experiment, maybe it can halp you more: First four events were generated
by "dead-acute + o", the second four were generated by "dead-acute + n".
Martin

================================================================

KeyPress event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 6928293, (90,-10), root:(1054,11),
   state 0x2000, keycode 21 (keysym 0xfe51, dead_acute), same_screen YES,
   XLookupString gives 1 characters:  "´"

KeyRelease event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 6928433, (90,-10), root:(1054,11),
   state 0x2000, keycode 21 (keysym 0xfe51, dead_acute), same_screen YES,
   XLookupString gives 1 characters:  "´"

KeyPress event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 6929137, (90,-10), root:(1054,11),
   state 0x2000, keycode 32 (keysym 0x6f, o), same_screen YES,
   XLookupString gives 1 characters:  "o"

KeyRelease event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 6929287, (90,-10), root:(1054,11),
   state 0x2000, keycode 32 (keysym 0x6f, o), same_screen YES,
   XLookupString gives 1 characters:  "o"
........................................................................................................................................................
KeyPress event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 6932295, (90,-10), root:(1054,11),
   state 0x2000, keycode 21 (keysym 0xfe51, dead_acute), same_screen YES,
   XLookupString gives 1 characters:  "´"

KeyRelease event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 6932445, (90,-10), root:(1054,11),
   state 0x2000, keycode 21 (keysym 0xfe51, dead_acute), same_screen YES,
   XLookupString gives 1 characters:  "´"

KeyPress event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 6933180, (90,-10), root:(1054,11),
   state 0x2000, keycode 57 (keysym 0x6e, n), same_screen YES,
   XLookupString gives 1 characters:  "n"

KeyRelease event, serial 21, synthetic NO, window 0x3000001,
   root 0x26, subw 0x0, time 6933309, (90,-10), root:(1054,11),
   state 0x2000, keycode 57 (keysym 0x6e, n), same_screen YES,
   XLookupString gives 1 characters:  "n"

================================================================

Comment 6

[Roland Mainz]

Is there any way that we can use XmbLookupString() or XwcLookupString() instead
of using our own, hardcoded keysym-->UCS2 table ?

Comment 7

[Kai Lahmann (is there, where MNG is)]

Martin Broz: can you still reproduce this with a recent Mozilla version (like
1.0RC3)? If not, mark this bug "worksforme"

Comment 8

[Martin Broz]

I am sorry, but it still persist - build 2002060107.
Martin

Comment 9

[Kai Lahmann (is there, where MNG is)]

and with 1.1beta? (no, we aren't waiting, until you give up, but I wait until
somebody finds this interesting enough to set it to "NEW" ;)

Comment 10

[Roland Mainz]

Kai Lahmann wrote:
> I wait until somebody finds this interesting enough to set it to "NEW" ;)

I can't really confirm it but the reporter has posted enougth details that this
bug can get - at least - the "NEW" status.

Comment 11

[Martin Broz]

Still not fixed - build 2002081104.
Martin

Comment 12

[Michal Kubecek]

This bug disappeared for me after upgrading to SuSE 8.1 (XFree86 4.2.1), both
for Mozilla 1.0.1 and 1.2beta. I don't think, however, that the bug is in
XFree86 since Mozilla was the only application behaving this way. Output of
'xev' seems to be the same as before:

-----------------------------------------------------------------------------
KeyPress event, serial 27, synthetic NO, window 0x2800001,
    root 0x3e, subw 0x0, time 3964050, (90,101), root:(560,415),
    state 0x2010, keycode 21 (keysym 0xfe51, dead_acute), same_screen YES,
    XLookupString gives 1 characters:  "´"

KeyRelease event, serial 27, synthetic NO, window 0x2800001,
    root 0x3e, subw 0x0, time 3964214, (90,101), root:(560,415),
    state 0x2010, keycode 21 (keysym 0xfe51, dead_acute), same_screen YES,
    XLookupString gives 1 characters:  "´"

KeyPress event, serial 27, synthetic NO, window 0x2800001,
    root 0x3e, subw 0x0, time 3964667, (90,101), root:(560,415),
    state 0x2010, keycode 32 (keysym 0x6f, o), same_screen YES,
    XLookupString gives 1 characters:  "o"

KeyRelease event, serial 27, synthetic NO, window 0x2800001,
    root 0x3e, subw 0x0, time 3964831, (90,101), root:(560,415),
    state 0x2010, keycode 32 (keysym 0x6f, o), same_screen YES,
    XLookupString gives 1 characters:  "o"
-----------------------------------------------------------------------------

Comment 13

[Masaki Katakai]

Thanks for the info Michal.

How about you, Martin?

Is it possible to try on other linux distribution?

Comment 14

[Martin Broz]

I have tried new XFree86 (version 4.1 instead of 3.3.6) and it works OK.
Distribution is Debian 3.0.
Martin

Comment 15

[Masaki Katakai]

Thanks, Martin.

Marked this as WORKSFORME on XFree86 4.x.

Comment 16

[sujay]

verified.