Johnny Stenback
c262d55f79
Fixing bug 474022. Reflow if we're about to instantiate a plugin whose frame has not been reflown yet. This fixes a problem where the WMP plugin doesn't work right if its initial size is 0x0. r+sr=bzbarsky@mit.edu
2009-02-02 17:26:28 -08:00
Dave Camp
e27a8b62d1
Bug 473938 Stylesheets don't load without a safe browsing service r+sr=sicking
2009-01-16 21:54:39 +00:00
Dave Camp
c12a3b0506
Bug 441359: Check script and css loads against the classifier. r+sr=jonas
2009-01-13 23:13:48 -08:00
Dave Camp
b18d454609
Backed out changeset 44890ee1d15f for leaks
2009-01-13 02:22:59 -08:00
Dave Camp
ca7e07e266
Bug 441359: Check script and css loads against the classifier. r+sr=jonas
2009-01-12 21:52:00 -08:00
Benjamin Smedberg
24e7cb1f78
Bug 396185 - Make nsIFrame derivatives and helper abstract classes use a different dynamic-cast system than nsISupports:
...
* we know all types frames may be cast to at compile time, so instead of extensible GUID IIDs, use a big enum (see nsQueryFrame::FrameIID)
* eliminate all vestiges of refcounting, since frames aren't refcounted
Some frames (SVG frames in particular) still implement nsISupports-derived interfaces, for example nsISVGValue. There is a FrameIID for nsISVGValue that lets you go from a frame to the XPCOM interface, but you can't query back.
r+sr=roc
nsITextControlFrame didn't have an IID the first time around, but this wasn't a compile error because nsITextControlFrame::kFrameIID inherited from nsIFormControlFrame::kFrameIID. I've added a static analysis pass to verify the correct behavior, since I can't figure out a way to make the compiler do it.
--HG--
extra : rebase_source : 4894a2ca0278e2ab92f27459db77165f8348cf41
2009-01-12 14:20:59 -05:00
Benjamin Smedberg
50621e4f9f
Backed out changeset 4c4df6ed1b41 - Bug 396185 - Make nsIFrame not inherit from nsISupports due to mochitest failures... these appear to be crashes in nsGenericHTMLElement::GetEditorInternal.
2009-01-09 11:35:24 -05:00
Benjamin Smedberg
d0b948876f
Bug 396185 - Make nsIFrame derivatives and helper abstract classes use a different dynamic-cast system than nsISupports:
...
* we know all types frames may be cast to at compile time, so instead of extensible GUID IIDs, use a big enum (see nsQueryFrame::FrameIID)
* eliminate all vestiges of refcounting, since frames aren't refcounted
Some frames (SVG frames in particular) still implement nsISupports-derived interfaces, for example nsISVGValue. There is a FrameIID for nsISVGValue that lets you go from a frame to the XPCOM interface, but you can't query back.
r+sr=roc
This patch locally causes two REFTEST-UNEXPECTED-PASS for Bidi stuff. It's possible that I accidentally fixed a bug, but I'm not sure, so I'm going to wait for the tinderboxes to confirm my local results.
2008-11-05 14:25:30 -05:00
Dave Camp
d4f7aed7eb
Backed out changeset e31d0d3c28fd (bug 441359)
2009-01-05 01:27:51 -08:00
Dave Camp
c3150e042c
Bug 441359: Run scripts/css through the classifier. r+sr=sicking
2009-01-04 23:40:29 -08:00
Olli Pettay
8dfbd47e9e
Bug 462517, r=biesi, sr=roc
2008-12-03 12:53:50 +02:00
Boris Zbarsky
164afa9a0a
Bug 458100. Make sure that we'll also instantiate the plug-in even if the nsPluginDocument has no presshell yet at OnStartRequest time. Thanks to Evgeny <evg.salmin@gmail.com> for pointing out the problem. r+sr=jst
2008-10-07 14:53:23 -04:00
Arpad Borsos
cce478d087
Bug 454482 - Clean up unused variables in content/; r+sr=peterv
2008-09-24 06:37:56 +02:00
Dave Townsend
3919bf60ce
Bug 391728: No placeholder for disabled plugins. r=jst, r=josh, sr=bz
2008-09-09 16:43:21 +01:00
Johnny Stenback
0089661672
Fixing bug 445520. Make sure plugin instantiation attempts that happen before a plugin frame has been reflown re-instantiates the plugin so that updates to the plugin src etc actually do happen in those cases as well. r=cbiesinger@gmail.com, sr=bzbarsky@mit.edu
2008-08-15 08:30:26 -07:00
Boris Zbarsky
baa891763c
Bug 113934. Backend and some minimal front end for dragging tabs between windows. r=gavin, r+sr=jst
2008-08-11 09:38:43 +01:00
Dave Townsend
3af4f7166c
Backed out changeset eb6fda3be5bb
2008-08-10 17:15:00 +01:00
Boris Zbarsky
1b9a4caf46
Bug 113934. Backend and some minimal front end for dragging tabs between windows. r=gavin, r+sr=jst
2008-08-07 19:15:40 -04:00
Johnny Stenback
aff325c8ef
Fixing bug 438830. Prevent instantiation of a plugin if it's already been instantiated. r+sr=jonas@sicking.cc
2008-06-27 12:54:27 -07:00
jst@mozilla.org
41cc75cf2e
Fixing bug 430802 (and bug 405357). Prevent plugin instantiation code re-entrancy. r=jwatt@jwatt.org, sr=bzbarsky@mit.edu, a=beltzner
2008-05-06 13:49:30 -07:00
jonas@sicking.cc
50e225dd71
Putting back since orange wasn't affected
2008-05-01 15:43:42 -07:00
jonas@sicking.cc
727cd3dafb
Backing out to try to fix orange
2008-05-01 12:03:48 -07:00
jst@mozilla.org
dc5e4ea192
Fixing bug 431280. Sniff the URI file extension for a mimetype to use if the one given in an embed elements type attribute isn't a supported type. r+sr=bzbarsky@mit.edu, a=dsicore@mozilla.com
2008-04-30 14:23:32 -07:00
mats.palmgren@bredband.net
6091930e10
Don't trigger "Missing plugin" UI for supported image or document types when there's no URI. b=429157 r+sr=jst a1.9=beltzner
2008-04-16 13:07:11 -07:00
bent.mozilla@gmail.com
7e4ac1fa4a
Bug 425551 - "Fix async DOM events from keeping link elements alive past unlink". r+sr=sicking, a=beltzner.
2008-04-11 10:29:06 -07:00
jonas@sicking.cc
b8a008856b
Backing out due to test failures
2008-04-08 20:20:39 -07:00
jonas@sicking.cc
5e5ddf5b3f
Bug 425551: Improve script blocker code and use it to prevent async DOM events from keeping link elements alive past CC. Patch by bent. r/sr=sicking
2008-04-08 19:06:27 -07:00
jst@mozilla.org
fe790e64b1
Fixing bug 393756. Get the plugin mimetype from the file extension in the URI if no other type information is available. r+sr=bzbarsky@mit.edu
2008-03-27 16:12:18 -07:00
jst@mozilla.org
e0c4d37c12
Fixing bug 425013. Make Firefox show the plugin finder UI for unsupported object tags w/o alternate content. r+sr=bzbarsky@mit.edu
2008-03-26 16:04:57 -07:00
jst@mozilla.org
14b80d26bc
Landing followup fix for bug 402983 and re-enabling the new stricter file URI security policies. r+sr=bzbarsky@mit.edu
2008-03-22 09:50:47 -07:00
jst@mozilla.org
2df3e44ea1
Fixing bug 395110. Give type attribute higher priority when deciding what plugin to use. r=jonas@sicking.cc, sr=bzbarsky@mit.edu
2008-03-17 17:36:14 -07:00
jonas@sicking.cc
da4cab3b75
Bug 401155: Make it safe to create a wrapper for an object at any time by making sure wrapping doesn't write script. Also set up a service that tries to keep track of when it's safe to execute script. r=bz sr=jst
2008-03-14 16:08:57 -07:00
dcamp@mozilla.com
cc10446d26
Bug 394485: Prevent loading blocklisted urls loaded from <object> tags. r=jonas, sr=biesi, blocking1.9=stuart
2008-03-12 14:52:47 -07:00
jonas@sicking.cc
47a2cff0da
Back out due to failed mochitest
2008-03-10 19:18:15 -07:00
jonas@sicking.cc
4b8f2cdf7a
Bug 401155: Make sure plugins and XBL constructors not execute when creating a script wrapper if it's not safe to execute scripts. r=bz sr=jst.
2008-03-10 17:49:19 -07:00
jst@mozilla.org
16e1932c1a
Fixing bug 410946. Prevent plugins from being destroyed while there's code from those plugins on the stack. r+sr=bzbarsky@mit.edu, a=beltzner@mozilla.com
2008-02-28 18:06:00 -08:00
dtownsend@oxymoronical.com
266b3971a8
Bug 409025: OBJECTs with type attribute but no data attribute do not fire plugin errors. r=biesi, sr=jst
2008-02-26 16:29:02 -08:00
dtownsend@oxymoronical.com
1c40cc488a
Bug 388445: Notify user when plugin on page is blocklisted. r=gavin, sr=jst
2007-12-27 03:49:26 -08:00
bzbarsky@mit.edu
f1f789125b
Propagate the node's principal as the owner to <object> loads as needed, and allow <object data="javascript:"> to work. Bug 353334 and bug 300263, r=biesi, sr=jst, a=dsicore
2007-12-03 13:57:17 -08:00
bzbarsky@mit.edu
187831740f
Handle image embeds with a null URI. Bug 402320, r+sr=jst
2007-11-07 09:25:57 -08:00
gavin@gavinsharp.com
673e7ccc5d
Back out the patch for bug 393845 because it appears the regressions it caused are worse than the bug it fixed (a=biesi)
2007-09-25 13:41:07 -07:00
bzbarsky@mit.edu
d8af0891f1
Make disabling image loading in prefs also block images loaded via <object>. Bug 393503, r+sr=biesi, a=jst
2007-09-14 11:57:59 -07:00
cbiesinger@gmx.at
3f5cf3a38d
393845 Stop plugins from content instead of from layout. Also, reget the frame after stopping the plugin, as that may process events. Finally, make the object frame not touch member variables after stopping the plugin. r+sr=bz a=jst
2007-09-05 13:33:18 -07:00
bzbarsky@mit.edu
a65e95ce03
Make <object> do the same text/plain sniffing that <iframe> does, and use the content type hint if the type comes back as application/octet-stream or sniffed-binary. Bug 389677, r+sr=biesi, a=sicking
2007-08-20 20:26:12 -07:00
bzbarsky@mit.edu
96ec7b6a69
Make NS_CheckContent(Load|Process)Policy only take a principal, not a principal and a uri. Get the URI from the principal. Bug 391438, r+sr+a=sicking
2007-08-09 22:35:08 -07:00
bzbarsky@mit.edu
73073f4f3b
Use the principal doing the load to decide on the URI to pass to content policies. This will help do the right thing for data:/javascript:/about:blank, as well as allow us to skip content policies altogether if the caller is system. Bug 388597, r+sr=sicking
2007-08-07 18:16:09 -07:00
cbiesinger@gmx.at
6b6b771bea
Bug 390385 make sure that plugins don't get instantiated before the first Reflow of the objectframe, so that the first NPP_SetWindow call contains the right dimensions r+sr=bz
2007-08-06 17:32:14 -07:00
cbiesinger@gmx.at
3416bd0451
364235 don't instantiate plugins for URLs with unknown schemes unless we also have a type specified r+sr=bz
2007-08-06 11:02:14 -07:00
cbiesinger@gmx.at
9b77299b29
382378 call NotifyContentObjectWrapper after the plugin has been instantiated so that scriptable plugins work correctly r+sr=bz
2007-08-02 14:15:05 -07:00
cbiesinger@gmx.at
15ca9d3c96
322170 ignore errors from OnStartRequest (NPP_NewStream) for Shockwave on the Mac r=jst sr=bz
2007-08-02 13:16:49 -07:00
