54241 - chrome://necko/... should be fixed

Status: RESOLVED WORKSFORME

(Core :: Networking, defect, P3)

Description

[Warren Harris]

Right now we have a necko package in chrome (referred to via chrome://necko/... 
urls). This gets used in the following places:

/netwerk/protocol/http/src/nsHTTPChannel.h, line 52 -- #define 
LOOPING_REDIRECT_ERROR_URI "chrome://necko/content/redirect_loop.xul"

/extensions/cookie/tests/TestCookie.cpp, line 95 -- char* propertyURL = 
"chrome://necko/locale/necko.properties";
/embedding/base/nsEmbedAPI.cpp, line 155 -- char* propertyURL = 
"chrome://necko/locale/necko.properties";
/netwerk/base/public/netCore.h, line 65 -- #define NECKO_MSGS_URL 
"chrome://necko/locale/necko.properties"
/netwerk/resources/content/redirect_loop.xul, line 3 -- <!DOCTYPE window SYSTEM 
"chrome://necko/locale/redirect_loop.dtd">

We need to make either a contents.rdf file for the necko package so that we 
avoid this message on startup:

*** Chrome Registration of locale: Checking for contents.rdf at 
jar:resource:/chrome/en-US.jar!/locale/en-US/necko/
WARNING: Trying old-style manifest.rdf. Please update to contents.rdf., file 
d:\seamonkey922\mozilla\rdf\chrome\src\nsChromeRegistry.cpp, line 1769

or we should change from using necko to something more generic 
(chrome://communicator/...?). 

Dave: What do you recommend? We want necko to be a stand-alone module someday.

Comment 1

[John Morrison]

Warren: would this also be the required work for redirect_loop.xul to work
again? 

(It currently points to a location that it cannot load, which leaves the 
browser content area with no document, and the throbber spinning forever.
It would be good to fix this for rtm)

Comment 2

[Simon Lucy]

I'm getting a gazillion warnings of
WARNING: chrome: failed to get base url for chrome://necko/locale/necko.properti
es -- using wacky default, file m:\tip\mozilla\rdf\chrome\src\nsChromeRegistry.c
pp, line 511

From this morning's build 16/11/2000

Comment 3

[Peter "jag" Annema]

I attached a patch to bug 60055 which adds the contents.rdf... However, if it's
deemed better to remove the neck stuff and add it directly to say communicator
or global, then that WFM too.

Comment 4

[Peter "jag" Annema]

*** Bug 60055 has been marked as a duplicate of this bug. ***

Comment 5

[David Hyatt]

The file shouldn't even exist.  It should just be removed.

Comment 6

[Warren Harris]

It should exist. It's all the error messages for necko's protocols.

Comment 7

[Peter "jag" Annema]

Checked in the patch attached to 60055, r=cls, a=alecf... Added makefile.win, no
worries ;-)

We can't get rid of these necko files easily, and I doubt we should. We could
perhaps move them to toolkit/global, I don't care.

Comment 8

[James Mastros (theorb)]

This seems to be fixed for me today, as does the Unknown error in status bar...

Anybody else seeing this?  I've made some local changes, but nothing more then
adding NS_ERRORs and NS_WARNINGs.  The Unknown Error messages have also ceased.

Comment 9

[benc]

mass move, v2.
qa to me.

Comment 10

[Doug Turner (:dougt)]

to default for triaging.

Comment 11

[Doug Turner (:dougt)]

moving neeti's futured bugs for triaging.

Comment 12

[Phil Ringnalda (:philor)]

If it hadn't already been fixed, this would have been fixed by bug 271324 in 2004.