Граф коммитов

375 Коммитов

Автор SHA1 Сообщение Дата
darin%netscape.com 00a7e27ee6 Necko API changes, bug 74221. r=valeski, sr=rpotts. 2001-04-10 06:01:08 +00:00
disttsc%bart.nl 72cea9bb38 Clean up MODULE/REQUIRES, bug=73353, r=cls 2001-04-08 08:33:11 +00:00
locka%iol.ie 082e014189 Updated GetTarget so windows are created if the named target cannot be found and for _new & _blank targets. b=72491, sr=mscott 2001-04-03 21:56:02 +00:00
rpotts%netscape.com b714e61819 bug #73244 (r=darin, sr=mscott). The DocLoader cannot assume that an nsIRequest is necessarily an nsIChannel... 2001-03-29 00:12:49 +00:00
alecf%netscape.com fcaf601272 fix for bug 73640 - make uri loader support "true" weak references so we can implement them from JavaScript
sr=mscott
r=jag
2001-03-28 18:09:03 +00:00
valeski%netscape.com 169a5140b5 r=rpotts. comment changes only. 48726. adding status to idl files of api rev. ifaces. 2001-03-24 00:22:18 +00:00
alecf%netscape.com 01b6ed44fd no bug - just reworking nsIURIContentListener to be more easily implemented in JS - does not affect any C++ interface signatures
sr=mscott
2001-03-21 19:21:34 +00:00
timeless%mac.com e76ccfd647 fix Bug 70989 Clean up lots of "shadows" and other common warnings
r=peterv, jst, ducarroz, edburns. sr=jst, shaver
2001-03-21 05:32:22 +00:00
disttsc%bart.nl a6f2f5861a Mass REQUIRES update to synch up with string lib and xul changes in an attempt to fix senna bustage. r=jst, sr=cls 2001-02-22 09:35:51 +00:00
dougt%netscape.com 128f95aa9b Relanding Necko Changes.
Revising nsIChannel to allow for overlapped i/o. This consists of three parts:

1. Factoring nsIChannel into a protocol specific part, the nsIChannel, and a socket specific, the nsITransport.
2. Derive the nsIChannel from a nsIRequest.
2. Changes the notification system from necko and the URILoader to pass the nsIRequest interface instead of nsIChannel interface.

This goal stems from wanting to be able to have active AsyncRead and AsyncWrite operations on nsSocketTransport.
This is desired because it would greatly simplify the task of maintaining persistent/reusable socket connections
for FTP, HTTP, and Imap (and potentially other protocols). The problem with the existing nsIChannel interface is
that it does not allow one to selectively suspend just one of the read or write operations while keeping the other active.

r=darin@netscape.com
sr=rpotts@netscape.com
2001-02-21 20:38:08 +00:00
disttsc%bart.nl 3d2d80d536 Back out dougt's channel changes 2001-02-12 03:14:23 +00:00
dougt%netscape.com 1b9ca82439 Revising nsIChannel to allow for overlapped i/o. This consists of three parts:
1. Factoring nsIChannel into a protocol specific part, the nsIChannel, and a socket specific, the nsITransport.
2. Derive the nsIChannel from a nsIRequest.
2. Changes the notification system from necko and the URILoader to pass the nsIRequest interface instead of nsIChannel interface.

This goal stems from wanting to be able to have active AsyncRead and AsyncWrite operations on nsSocketTransport.
This is desired because it would greatly simplify the task of maintaining persistent/reusable socket connections
for FTP, HTTP, and Imap (and potentially other protocols).  The problem with the existing nsIChannel interface is
that it does not allow one to selectively suspend just one of the read or write operations while keeping the other active.

The full details of the change on written up in the netlib newsgroup.

r=darin@netscape.com
sr=rpotts@netscape.com
2001-02-10 00:16:26 +00:00
mscott%netscape.com 2019b92f07 Bug #67598 --> we no longer need to invoke the ucth dialog here.
sr=sspitzer
2001-02-07 02:44:09 +00:00
locka%iol.ie 06506207df Added code to create a new window for _blank, _new targets before URI loader tries to push content into it and fails. sr=mscott@netscape.com b=41241 2001-02-05 10:50:33 +00:00
valeski%netscape.com 10230f80ff r=ccarlen, sr=rpotts. 65925. making webprogress implementation store weak refs to listeners. 2001-01-31 21:02:08 +00:00
rpotts%netscape.com 3228f83073 bug #62955 - Dialog boxes are the wrong size. 2000-12-23 06:07:38 +00:00
mscott%netscape.com ee60df101f Bug #59827 --> pass the redirect flag up via a OnStatechange if we get a redirect. Also set the is document
flag if the original channel matches mDocumentChannel.
sr=rpotts
2000-12-19 22:07:58 +00:00
rpotts%netscape.com 359b761343 bug #46866. Added a read-only DOMWindow attribute to nsIWebProgress. 2000-12-14 23:26:35 +00:00
cls%seawood.org 66a18fcbbf Resurrect REQUIRES so that we have some sort of means to track intermodule dependencies. Bug #59454 r=blizzard@mozilla.org 2000-11-20 07:16:06 +00:00
mscott%netscape.com 94e52d3266 Bug #56938 --> add a method for retrieving the doc loader's
document channel.
r=sspitzer, sr=rpotts
2000-11-01 08:04:49 +00:00
warren%netscape.com cd56c0575b Bug 47207. Backing out logging/PRINTF changes until we can fix stopwatch.h, introduce double parens, etc. 2000-10-28 22:17:53 +00:00
warren%netscape.com 0735771512 Removed extra 0 arg 2000-10-27 23:21:30 +00:00
warren%netscape.com 9a6d92a433 Bug 47207. Changing printf to PRINTF to use new logging facility. r=valeski,sr=waterson 2000-10-27 22:43:51 +00:00
mscott%netscape.com f5614e7782 Bug #56934 --> properly determine the status for on end document load. We used to
use the status of the last request processed in loading the document to determine
success or failure. That's incorrect. Instead, test to see if the load group is being
canceled. If it is, use that as the status for the entire document. Otherwise, ignore
the status for the last request and instead use the status for the main document
(the default load channel).
sr=rpotts, r=sspitzer
2000-10-26 06:34:20 +00:00
pollmann%netscape.com 8911ee3487 Bug 13871: Frameset spoofing - cleanup unneeded DOM interface. r=jst, sr=vidur 2000-10-24 04:35:25 +00:00
pollmann%netscape.com 3da7c0c668 Bug 13871: Prevent frameset spoofing r=mstoltz, sr=mscott, a=rpotts 2000-10-19 10:25:49 +00:00
mscott%netscape.com 07637ef394 Bug #47636 --> add methods to convert back and forth between a nsIDocShellLoadTypeInfo
and a docshell load tye. Unify nsIURILoadCommand and nsIDocShellLoadType enums so they
can be treated as the same type. This allows the uriloader to pass the correct load info
from the docshell that originates the load over to the docshell that actually
ends up loading the url.
r=radha, sr=rpotts
2000-10-11 23:14:38 +00:00
mscott%netscape.com 7cb10d986d Bug #47203 --> force more code to use the helper app dialog.
sr=alecf,rpotts
2000-10-04 06:53:52 +00:00
dougt%netscape.com c2923c8476 + Added OnSecurityChange() to nsIWebProgressListener interface.
+ Added dummy function to all users of nsIWebProgressListener
+ Added new security event sink.
+ Hooked up new event sink to docloader and friends.
+ Fixed memory leaks and crashes in nsSecureBrowserImpl.
+ Added AlertPrompt to nsIPrompt Interface.
+ Enabling xpcom test on unix.

Fixes bug 46872.  r=valeski/rpotts
2000-09-14 22:56:56 +00:00
rayw%netscape.com 0257791053 Bug 37275, Changing value of all progids, and changing everywhere a progid
is mentioned to mention a contractid, including in identifiers.

r=warren
2000-09-13 23:57:52 +00:00
mscott%netscape.com 91c59f8553 Fix unix breakge by using .get() 2000-09-01 23:07:57 +00:00
mscott%netscape.com 14371bacb4 Bug #47365 --> set the retargeting URL flag when we are retargeting to another window
r=radha
2000-09-01 22:28:37 +00:00
mscott%netscape.com 0e2dcfbd89 Bug #50427 --> comment out our OnRedirect implementation. It was causing nasty side effects
because we aren't told if the main document url is getting redirected or if a part (like an
image) is getting redirected. This caused the urlbar to get incorrectly updated.
r=sspitzer
2000-09-01 22:26:32 +00:00
vidur%netscape.com e153192875 Fix for bug 46861. nsIDOMWindow is now split into two interfaces: a public interface (still called nsIDOMWindow) for embedders and an internal interface (nsIDOMWindowInternal) for internal use. The union of the two interfaces is still visible from JavaScript. r=jst@netscape.com 2000-09-01 01:54:35 +00:00
locka%iol.ie ecebe2ac21 API changes for embedding. b=46847, r=dougt, a=valeski 2000-08-25 18:39:46 +00:00
mscott%netscape.com d7a4c2790e Bug #27048 --> implement nsIHTTPEventSink
r=valeski
2000-08-25 01:32:40 +00:00
rpotts%netscape.com b569585070 bug #39381. The last channel in a document was not firing a request done notification - only a document done... 2000-08-24 00:08:15 +00:00
heikki%netscape.com 418d0168ea Fixed bug 41772, if XLink target was XML file it never opened it in new window. r=jst. 2000-08-10 20:18:49 +00:00
warren%netscape.com 2b01440bc0 Factored string bundles out of necko. Fixes bug 42107 - need to fix string bundle init workaround. Also bug 40506 - nsIWebProgressListener needs status text. Status messages now work in mozilla and viewer with internationalized and parameterized text! Added temporary error architecture until bug 13423 can be fixed. Extended nsIStringBundleService to provide method to format status message. r=valeski,jband,tao. verified=mstoltz 2000-07-25 05:45:56 +00:00
valeski%netscape.com 048c917d19 45723. r=rpotts. webshell-docshell consolodation changes. 2000-07-18 23:37:53 +00:00
mscott%netscape.com 2bdc7ded6d Bug #41707, #43652 --> JS urls aren't working. We need to check for the correct return value and don't propogate
the on start request in that case.
r=alecf
2000-07-07 02:23:19 +00:00
dougt%netscape.com 17461eaf11 Check for null. If a NS_EXTERNALHELPERAPPSERVICE_PROGID was not found,
crash would occur.
2000-06-27 21:43:17 +00:00
mscott%netscape.com 5ce6a320cc Bug #38374, 30157, 10802 --> turn on launching of helper applications to handle content types we don't know how to handle!!
turn on the ability to launch urls using OS default protocol handlers.
uri loader modifications to make open attachment possible.
r=sspitzer
2000-06-23 04:43:28 +00:00
rpotts%netscape.com dfe7dd1ea3 Fixed memofy leak of StdURLs introduced by the last checkin... 2000-06-20 05:14:42 +00:00
jdunn%netscape.com b86b3b5990 adding newline, required by HPUX 2000-06-19 13:03:52 +00:00
rpotts%netscape.com 44dfb4b825 Reworked the nsIWebProgress interfaces and implementation... 2000-06-19 05:54:37 +00:00
mscott%netscape.com 6e2ee7ff03 Bug #40116 --> check to see if the channel was canceled before
propogating calls to the content listener...
r=valeski
2000-06-16 21:21:04 +00:00
rpotts%netscape.com 9c4c702bdc bug #29178. Do not show a blank document when a HTTP server returns a status 204 - No Content 2000-05-25 05:12:58 +00:00
pollmann%netscape.com ae19609176 Bug 32898: Fix crash when targetting new window. r=rpotts 2000-05-11 01:33:25 +00:00
ruslan%netscape.com e7d80eb3ca Prevent infinite loop in OnStop in case when Cancel is issued 2000-05-10 02:05:54 +00:00
rpotts%netscape.com d6a55e5b45 Removed nsLoadCookie and made nsDocLoader implement nsIInterfaceRequestor directly... 2000-05-07 02:01:10 +00:00
mscott%netscape.com 2b7f09c1b0 add thread safey addref and release macros to docloader. 2000-05-04 21:34:50 +00:00
pavlov%netscape.com a3a300fdf2 string api changes r=scc 2000-04-25 05:40:45 +00:00
mscott%netscape.com 96bcbcb394 need to fake a onprogress change for the parent
docloader...Fix viewer crash by returning aChannel instead of the
doc channel.
2000-04-25 01:45:37 +00:00
tbogard%aol.net 7b4745b063 Added new flag for a timed out net connection. 2000-04-24 05:03:49 +00:00
mscott%netscape.com e80ecd5bda help support the doc loader using the web progress listener. 2000-04-22 23:17:11 +00:00
mscott%netscape.com a9b0a90d18 Bug #33156 --> implment onChildStatusChanged AND progress notifications for the doc loader
through the web progress interfaces. In order to implement progress, the doc loader now implements
nsIProgressEventSink and receives events directly from the channels.
2000-04-22 23:16:31 +00:00
warren%netscape.com eeb4df595e Bug 34217. Making js protocol not clear window when it returns no result. 2000-04-21 05:18:15 +00:00
mscott%netscape.com 9671270568 Bug #36221 --> throbber never stops when loading a frameset doc.
I was signaling the stop notification via a OnchildStatus in this scenario. We should always signal the start
and stop of documents as a status change even if a child is originating the change.
2000-04-18 21:39:09 +00:00
mscott%netscape.com c4a408e35b Bug #36218 --> fix ftp! We were unable to open the unknown content handler because we couldn't get
the dom window from the original docshell. I changed it to make this possible.
2000-04-18 20:41:38 +00:00
mscott%netscape.com 59ef60bcc9 Pass in the right channel into the web progress listener. 2000-04-14 03:28:23 +00:00
tbogard%aol.net 1e930b11ae Removed the viewSource member and the GetStringForCommand as it is no longer used. 2000-04-11 23:52:37 +00:00
tbogard%aol.net 3cbce7298c Changed the Status flags to be transition flags rather than flat current state. This makes it easier to implement a listener. Otherwise you are forced to cache the last status and then determine transitions based on the deltas. 2000-04-04 02:45:32 +00:00
mscott%netscape.com c075035b09 Bug #34321,34350 --> be sure to set the web progress listener on the doc loader
before every load.
r=travis
2000-04-03 22:41:26 +00:00
mscott%netscape.com ccd7510b99 Bug #33156 --> the doc loader now implements nsIWebProgress and makes basic notifications to a registered
nsIWebProgressListener.

Right now, the only methods that are hooked up are signaling when the doc loader is busy loading a document and
when it is done loading a document.
2000-04-03 02:26:05 +00:00
scc%netscape.com d8f4cef205 making string conversions explicit 2000-04-02 09:15:46 +00:00
valeski%netscape.com 45a4daefea 29858. special case a */* content type returned from the server. we want to treat it as unknown and ask the user what to do. 2000-03-30 22:24:07 +00:00
jim_nance%yahoo.com dc0e139673 Bug 33322 - Fix compiler warnings. Its been on n.p.m.builds for a few
days for module owners to comment on.  r=troy@netscape.com
2000-03-30 03:21:26 +00:00
tbogard%aol.net 98cb0f4583 Updated the documentation to clearly state the parameters of OnChildProgressChange. 2000-03-29 10:14:59 +00:00
warren%netscape.com 727f312b32 Necko API changes: primarily nsIChannel, changing initialization parameters to accessors. Got javascript: evaluation to happen at the right time (when AsyncRead is called) as well as on the right thread. 2000-03-29 03:58:50 +00:00
scc%netscape.com b9eb8da352 small changes to clients of string converting APIs 2000-03-26 08:05:46 +00:00
tbogard%aol.net 8797c75acb Removed the cookie from AddProgressListener and RemoveProgressListener because frankly that parameter was dumb. :) 2000-03-26 04:51:38 +00:00
tbogard%aol.net ef7f9961dc Added a new method onLocationChange that is called when the location of the watched window changes. 2000-03-25 01:00:04 +00:00
mscott%netscape.com f6d844d6cb Bug #28776,28569 --> add method for clearing out the parent doc loader
now that we don't leak the global doc loader, there are times when it gets destroyed b4 it's children during
the shutdown process.
r=sspitzer
2000-03-24 05:45:31 +00:00
tbogard%aol.net 549310ce03 Added a new method to nsIURIContentListener to allow the windowContext listener to get a crack at cancelling a new load that is about to occur. 2000-03-24 00:23:40 +00:00
ftang%netscape.com 8e1d13ceca change the interface and pass nsISupport* for document instead of PRInt32 so we won't drop
data on 64-bits machine. r=harishd
2000-03-23 00:53:53 +00:00
tbogard%aol.net e648223c3a HandleUnknownContentType is no longer part of the nsIDocumentLoaderObserver interface. 2000-03-11 01:24:29 +00:00
mscott%netscape.com 333edb2c69 Bug #30553 --> new windows with window targets were getting the target associated with the
chrome window instead of with the content window. And the window sizes were coming from the
	       hidden window instead of the parent window the url originated from.
r=travis
a=jevering
2000-03-09 08:17:40 +00:00
warren%netscape.com c0497e31be Bug 21556: Making linux be thread-safe. Making tons of classes implement threadsafe AddRef/Release. a=jar 2000-03-05 21:26:01 +00:00
tbogard%aol.net 8ca243538f Adding the new webProgress interfaces to the build. a=jevering. 2000-03-02 03:03:32 +00:00
tbogard%aol.net b601c2d4b5 First cut at new webProgress interfaces. Not yet connected to the build. 2000-03-02 02:24:57 +00:00
tbogard%aol.net 86df273966 URILoader now will invoke the unkown content handler when the DoContent call fails. This clears the way for the new docShell creation of the content viewer code to land. I'm actually checking this in for mcott. b=13374 r=me a=jevering 2000-02-29 21:33:23 +00:00
mscott%netscape.com 9a7c1b0e8e Bug #27419 --> if uri load had a window target name and a window
with that name didn't exist yet, we weren't creating a new window,
instead we were loading it in the current window. Now we properly
fall through to the create new content handler case.
r=travis
a=phil
2000-02-19 02:05:52 +00:00
mscott%netscape.com 1388ba58b5 Bug #17468 --> if you pass a window target of _new, we weren't creating a new window. We do now.
r=travis
2000-02-16 07:10:42 +00:00
mscott%netscape.com c8fd767935 Fixes Bug #27419 and various other uri re-targeting bugs. The uri loader now properly implements retargeting.
Allow the load cookie to support nsWeakReference.
r=travis,mscott (we co-authored it).
2000-02-15 04:04:34 +00:00
tbogard%aol.net 84174fe997 Added a Stop method to allow stopping a load that has been requested. r=hyatt 2000-02-14 10:12:55 +00:00
mscott%netscape.com 9a386e5c0a More prepation for docshell landing. Add a IsPreferred method to nsIURIContentListener and use the Ispreferred
method when the load type is user click to find the preferred registered window for handling the content

r=travis
2000-02-04 08:43:34 +00:00
neeti%netscape.com 81b45a50d4 Returning the correct error value instead of NS_OK, from OnStartRequest() and OnStopRequest().r=rpotts 2000-02-04 01:54:47 +00:00
mscott%netscape.com e707c020b3 fix solaris breakage..include nsIChannel.h when using a nsCOMPtr. =) 2000-01-29 08:32:03 +00:00
mscott%netscape.com bf0b2945cd build nsdocLoader 2000-01-29 06:03:01 +00:00
mscott%netscape.com 81ddbb480d Part of docshell/webshell landing prep work. The doc loader is being re-factored into the uri loader.
add get content listener parent and load cookie attributes to nsIURIContentListener
Bug #21173 --> set the redirected flag on the channel if we are going to redirect the url. Add load cookie
support.
build nsDocLoader in the uriloader.
r=travis
2000-01-29 06:02:36 +00:00
mscott%netscape.com ec5a0e3977 (NOT Part of the build yet!!)
Refacting of doc loader into the uri loader
r=travis
2000-01-29 04:41:04 +00:00
mscott%netscape.com d33b6baf54 Part of docshell/webshell landing prep work. The doc loader is being re-factored into the uri loader.
These files aren't part of the build yet.
r=travis
2000-01-29 04:26:29 +00:00
norris%netscape.com 8507a58ec3 Files:
caps/include/nsScriptSecurityManager.h
	caps/src/nsScriptSecurityManager.cpp
	modules/libpref/src/init/all.js
Fix
24565 nsScriptSecurityManager::GetSecurityLevel() is a performance
24567 re-write DOM glue security checks to avoid NS_WITH_SERVICE()
r=waterson

Files:
	dom/src/base/nsGlobalWindow.cpp
	layout/base/src/nsDocument.cpp
	layout/base/src/nsGenericElement.cpp
Fix assertion failure for 1-character property names.


Files:
	dom/src/jsurl/nsJSProtocolHandler.cpp
	webshell/src/nsDocLoader.cpp
Fix 18653 "javascript:" URLs cross windows problems (probably regressi
r=nisheeth

Files:
	layout/events/src/nsEventListenerManager.cpp
Fix
23834 document.onkeypress allows sniffing keystrokes
24152 document.onclick shows links from other window
r=joki
2000-01-23 04:23:14 +00:00
mscott%netscape.com ab00d9bb79 Bug #24174 --> when clicking on a mailto ur, we were bringing up the compose window and
the unknown content handler dialog. we should only be bringing up the compose window =)
r=sspitzer
2000-01-17 22:21:25 +00:00
mscott%netscape.com a770fdf7e5 loading a uri takes the channel instead of the uri. The caller
must create the channel before call openuri.
r=rpotts.
2000-01-17 19:56:31 +00:00
mscott%netscape.com 0d953fbac3 loading a uri takes the channel instead of the uri. This allows us to remove
several fields from OpenURI interface and allows us to remove the private
post data interface. Now the nsIURILoader interface doesn't need to take
all of the arguments necessary to create a channel.
r=rpotts.
2000-01-17 19:55:27 +00:00
rpotts%netscape.com be79b9fb76 bug #23055 (r=warren). Added LOAD_DOCUMENT_URI load attribute for the DocLoader to distinguish doc URIs 2000-01-15 06:12:54 +00:00
mscott%netscape.com 6bbb531aa6 Short term hack...allow the uri loading to proceed even if we don't have a load group. eventually we should try to
create a loadgroup for the caller.
2000-01-09 23:45:36 +00:00
rpotts%netscape.com 5dd5ac6a7e Changed URI loader to properly handle stream decoders (ie. mixed-replace) 2000-01-05 04:44:27 +00:00