Closed
Bug 263845
Opened 20 years ago
Closed 20 years ago
Firefox (branch) unregisters Internet Explorer as default browser but shouldn't
Categories
(Firefox :: General, defect)
Tracking
()
VERIFIED
FIXED
Firefox1.0
People
(Reporter: ronny.perinke, Assigned: bugs)
Details
(Keywords: fixed-aviary1.0)
Attachments
(1 file)
1.26 KB,
patch
|
Details | Diff | Splinter Review |
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1)
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.3) Gecko/2004108 Firefox/0.10
Firefox (branch builds since 20041008
) seems to unregister Internet Explorer as default browser (and does not
register himself as the new one) although Firefox should not bee the default
browser.
So after closeing Firefox and starting IE again, IE says it is not the default
browser anymore (but only if the option to perform this check is active, if not
it won't tell you that).
This is really annoying because links from chat windows of my IM (trillian)
won't open correctly anymore - IE opens but shows a blank side with no url in
the adress bar (just like the blank startpage but without the 'about:blank' as
url in the adress bar).
Reproducible: Always
Steps to Reproduce:
1.) Start IE and make shure it is set as default browser (check out the option
in internetsettings -> programs that IE will check on startup if it's the
default browser or not). If not, than make it the default browser again.
2.) Close IE and start Firefox (branch build 20041008) and make shure, that the
option for making FF the default browser is not set.
3.) Close Firefox and open IE again. You will notice that IE is not the default
browser anymore although Firefox was not allowed to do so.
Actual Results:
- links (from instant messengers or URL files) only opens an empty IE window
- when opening IE and don't making it the default browser again (because the
option to check if IE is the default browser is not set) you can not open
internet settings from extras menu in IE
for any further details see http://forums.mozillazine.org/viewtopic.php?t=142560
Comment 1•20 years ago
|
||
->NEW
->MAJ
(few volunteers to try it out)
Severity: normal → major
Status: UNCONFIRMED → NEW
Ever confirmed: true
I can confirm that I see this exact behaviour as well. Using build:
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.3) Gecko/20041008
Firefox/0.10 (MOOX M3)
I think this should be nominated as a blocker as it can negatively affects the
user's experience with the browser by causing problems.
Comment 3•20 years ago
|
||
If this bug is indeed happening (I don't use nightlies, so I can't confirm), it
should definitely be a blocker. Releasing 1.0 with this bug would be as bad as
releasing 0.9 with the bug that caused FireFox to incorrectly register itself as
the default browser. That bug caused almost everyone I was trying to convert to
FireFox to have second thoughts.
I have to have IE set as my default browser at work because we use some IE
specific apps, I couldn't possibly use a copy of FireFox that caused IE to
become unregistered as the default browser. I'm sure there are many other users
in the same position.
Comment 4•20 years ago
|
||
Confirming. I am seeing this on Win2k.
Reposting problem details from the forum (per Henbe):
---
I'm also seeing this bug.
And from the little digging i'we done it's because firefox has started to delete
the following registry keys on exit:
HKCR\HTTP\shell\open\ddeexec\Application
HKCR\HTTP\shell\open\ddeexec\Topic
HKCR\HTTP\shell\open\ddeexec
HKCR\HTTPS\shell\open\ddeexec\Application
HKCR\HTTPS\shell\open\ddeexec\Topic
HKCR\HTTPS\shell\open\ddeexec
HKCR\FTP\shell\open\ddeexec\Application
HKCR\FTP\shell\open\ddeexec\ifExec
HKCR\FTP\shell\open\ddeexec\Topic
HKCR\FTP\shell\open\ddeexec
HKCR\GOPHER\shell\open\ddeexec\Application
HKCR\GOPHER\shell\open\ddeexec\Topic
HKCR\GOPHER\shell\open\ddeexec
---
The workaround/fix that works for me is after exiting FireFox manually launch
IE, then Tools->Internet Options->Programs. Then press "Reset Web Settings"
button. I usually untick the "Also reset my home page" and press Yes.
Comment 5•20 years ago
|
||
Ok. i messed around with commits during the day when regression came in.
It looks like a regression introduced by the commit for #262953
Looking at the attached patch and seeing code like:
-------------------------------------------
-NS_IMETHODIMP
+nsresult
nsWindowsShellService::UnregisterDDESupport()
{
DeleteRegKey(HKEY_CLASSES_ROOT, "HTTP\\shell\\open\\ddeexec");
DeleteRegKey(HKEY_CLASSES_ROOT, "HTTPS\\shell\\open\\ddeexec");
DeleteRegKey(HKEY_CLASSES_ROOT, "FTP\\shell\\open\\ddeexec");
DeleteRegKey(HKEY_CLASSES_ROOT, "CHROME\\shell\\open\\ddeexec");
DeleteRegKey(HKEY_CLASSES_ROOT, "GOPHER\\shell\\open\\ddeexec");
return NS_OK;
-------------------------------------------
Would definitely break IE, esp if the keys are deleted by mistake? [see my
previous comment about the keys deleted on exit]
Assignee | ||
Updated•20 years ago
|
Flags: blocking-aviary1.0? → blocking-aviary1.0-
Comment 7•20 years ago
|
||
So wait a minute, this isn't going to be fixed for 1.0? I know you guys have a
lot of important bugs to work on, but I think this one is really important. This
bug actually screws with the settings of another program and there's no
workaround. I think there's a lot of users who are going to want to keep IE as
the default while they try out FireFox and aren't going to like that FireFox is
unregistering IE, most of these same users probably won't know how to set IE
back to default and therefore won't be very happy. Some users might even see
this as "hijacking" their settings.
Reporter | ||
Comment 8•20 years ago
|
||
(In reply to comment #5)
> Ok. i messed around with commits during the day when regression came in.
>
> It looks like a regression introduced by the commit for #262953
>
> Looking at the attached patch and seeing code like:
> -------------------------------------------
> -NS_IMETHODIMP
> +nsresult
> nsWindowsShellService::UnregisterDDESupport()
> {
> DeleteRegKey(HKEY_CLASSES_ROOT, "HTTP\\shell\\open\\ddeexec");
> DeleteRegKey(HKEY_CLASSES_ROOT, "HTTPS\\shell\\open\\ddeexec");
> DeleteRegKey(HKEY_CLASSES_ROOT, "FTP\\shell\\open\\ddeexec");
> DeleteRegKey(HKEY_CLASSES_ROOT, "CHROME\\shell\\open\\ddeexec");
> DeleteRegKey(HKEY_CLASSES_ROOT, "GOPHER\\shell\\open\\ddeexec");
> return NS_OK;
> -------------------------------------------
> Would definitely break IE, esp if the keys are deleted by mistake? [see my
> previous comment about the keys deleted on exit]
>
>
>
I already postet the following at mozillazine-forum ...
I think the reason is the function 'Observe' in nsWindowsShellService.cpp which
calls the function 'UnregisterDDESupport' everytime firefox is closed.
Flags: blocking-aviary1.0- → blocking-aviary1.0+
Comment 9•20 years ago
|
||
You are only allowed to ?1.0
resetting
Flags: blocking-aviary1.0+ → blocking-aviary1.0?
Reporter | ||
Comment 10•20 years ago
|
||
(In reply to comment #9)
> You are only allowed to ?1.0
> resetting
Sorry, I didn't know.
Updated•20 years ago
|
Assignee: firefox → bugs
Comment 11•20 years ago
|
||
lets try and get this fixed
Flags: blocking-aviary1.0? → blocking-aviary1.0+
Assignee | ||
Comment 12•20 years ago
|
||
don't unregister DDEexec keys if we're not the default browser.
Assignee | ||
Comment 13•20 years ago
|
||
landed.
Status: NEW → RESOLVED
Closed: 20 years ago
Resolution: --- → FIXED
Reporter | ||
Comment 14•20 years ago
|
||
the patch did not work o.0
Comment 15•20 years ago
|
||
So, if the patch didn't work, should it be reopened? If it turns out it did
work, what about fixed-Aviary keyword?
Updated•20 years ago
|
Keywords: fixed-aviary1.0
Target Milestone: --- → Firefox1.0
Reporter | ||
Comment 16•20 years ago
|
||
(In reply to comment #14)
> the patch did not work o.0
>
I'm sorry, I was a little hastily. The build I made yesterday in the evening
with this patch deleted the reg-keys, but todays build not.
The patch works :)
Comment 17•20 years ago
|
||
Just out of curiosity, what is the purpose of deleting these keys on exit
anyway? I would expect they should remain until someone/something else becomes
default - or not?
Updated•20 years ago
|
Status: RESOLVED → VERIFIED
You need to log in
before you can comment on or make changes to this bug.
Description
•