My Browser Builds (Part 3)

[RainyShadow]

1 hour ago, XPerceniol said:

Also, I tried "folder option" -> "file types" and make .url (internet shortcut) open in New Moon and even that won't work.

Only .url ? There are more items there to be changed.

Look for URL:Hyper Text Transfer Protocol and URL:Hyper Text Transfer Protocol wyth Privacy

 

Also, if you have restored your registry after deleting the Mypal entries, try again but search only in HKEY_CLASSES_ROOT this time and replace all Mypal entries with NewMoon.

 

P.S.

You can also try to toggle browser.shell.checkDefaultBrowser in about:config , then restart your browser.

This may be useful. Check also the following posts, ignore M.T.

Edited May 18, 2021 by RainyShadow

[rereser]

try this:
http://www.pc-tools.net/win32/setbrowser/

[XPerceniol]

Fixed now :)

The installer by @i430VX did the trick and now its the default. However, I still have that raccoon so its still lurking around somewhere, but oh well, time to think about other things. 

Thanks everyone for your help and enjoy your day/evening.

[NotHereToPlayGames]

Navigate to C:\WINDOWS\Installer and look for any folders that have any Mypal files.

Delete those folders.

Reboot and the icons should return to normal.

If not, we may have to look into rebuilding the icon cache, it's been a while since I've done that so we'll cross that bridge if still required.

 

ps - too late for this warning, but I never use "installers", everything on my computer is from manually extracted "installers" and self-created registry settings for programs that aren't natively "portable".

[grey_rat]

roytam1,

I tested h/w acceleration on Radeon x700. Firefox does not include it. But works in your. Looks like Directx 9.0b support is critical

 

Dave-H,

do you have an h/w acceleration on x600?

 

[Dave-H]

Sadly not.
If I switch to my other graphics card, the Nvidia Quadro 2000, WebGL works fine, so the Firefox settings are presumably OK.
The ATI card just doesn't support it.
I have DirectX 9.0c installed.

[grey_rat]

DirectX 9.0b

Quote

R420Edit

While heavily based upon the previous generation, this line included extensions to the Shader Model 2 feature-set. Shader Model 2b, the specification ATI and Microsoft defined with this generation, offered somewhat more shader program flexibility.

 

Dave-H,

x600 supports WebGL, but not Angle, only OpenGL . You need to switch:

webgl.disable-angle - true

webgl.angle.try-d3d11 - false

webgl.force-enabled - true

Even the Radeon 9500 can WebGL

 

If WebGL through OpenGL will not work, you can do s/w WebGL, it will not h/w, but will work (necessary SSE2)

webgl.disable-angle - true

webgl.angle.try-d3d11 - false

webgl.force-enabled - true

gfx.prefer-mesa-llvmpipe - true

and add opengl32.dll to your browser folder   https://msfn.org/board/topic/182647-my-browser-builds-part-3/page/11/?tab=comments#comment-1199702

 

 

[NotHereToPlayGames]

How do you test these and know which true/false settings to use?

I generally DISABLE hardware acceleration as a rule-of-thumb but that's been based on too many people that report issues with it enabled.

[Dave-H]

54 minutes ago, grey_rat said:

DirectX 9.0b

 

Dave-H,

x600 supports WebGL, but not Angle, only OpenGL . You need to switch:

webgl.disable-angle - true

webgl.angle.try-d3d11 - false

webgl.force-enabled - true

Even the Radeon 9500 can WebGL

 

If WebGL through OpenGL will not work, you can do s/w WebGL, it will not h/w, but will work (necessary SSE2)

webgl.disable-angle - true

webgl.angle.try-d3d11 - false

webgl.force-enabled - true

gfx.prefer-mesa-llvmpipe - true

and add opengl32.dll to your browser folder   https://msfn.org/board/topic/182647-my-browser-builds-part-3/page/11/?tab=comments#comment-1199702

 

 

Thank you!

The third setting was already correct, but I changed the first two and WebGL now works with the ATI card!

[grey_rat]

Quote

I generally DISABLE hardware acceleration as a rule-of-thumb but that's been based on too many people that report issues with it enabled.

Typically, hardware acceleration problems occurred due to Direct2D. But Direct2D doesn't work for WindowsXP.
Only "compositing" can work in WindowsXP

Quote

How do you test these and know which true/false settings to use?

about:config and about:support

 

Dave-H,

For Radeon 9500 - x600 can enable OpenGL compositing. In K-Meleon, PM27 and Firefox browsers 45-47.

layers.prefer-opengl - true

layers.acceleration.force-enabled - true or layers.acceleration.force - true

In km - normal. In PM and Firefox - shifted up (fixed by full-screen mode and browser.fullscreen.autohide - false

PM28 (Firefox52) crash at start

 

[Dave-H]

layers.prefer-opengl - true seems to be fine on FF 52.9 ESR.
layers.acceleration.force-enabled causes it to crash on startup though.

[grey_rat]

Dave-H, only K-Meleon, PM27 and Firefox 45-47.

I asked Feodor2 to fix OpenGL, it did it, but then it broke it https://github.com/Feodor2/Mypal/issues/126#issuecomment-660646967

 

roytam1, is it possible to fix (without crashing) OpenGL compositing in NM28, Basilisk and Serpent?

[roytam1]

10 hours ago, grey_rat said:

Dave-H, only K-Meleon, PM27 and Firefox 45-47.

I asked Feodor2 to fix OpenGL, it did it, but then it broke it https://github.com/Feodor2/Mypal/issues/126#issuecomment-660646967

 

roytam1, is it possible to fix (without crashing) OpenGL compositing in NM28, Basilisk and Serpent?

no clue what/how is Feodor2 doing in his repo, it is hard to trace specified change since he always rebasing/recommiting changes.

[roytam1]

Advance Notice: since there is no windows-related UXP upstream changes, UXP-based builds will not be produced tomorrow.

[roytam1]

New NewMoon 27 Build!

32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210522-1e24b22ef-xpmod.7z
32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210522-1e24b22ef-xpmod-sse.7z
32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210522-1e24b22ef-xpmod-ia32.7z

64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20210522-1e24b22ef-xpmod.7z

source repo: https://github.com/roytam1/palemoon27

repo changes since my last build:
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1166504 - Make nsMultiplexInputStream cloneable. r=bkelly,froydnj (e3410f3a9)
 - Bug 1161240 - Make sure that NS_CloneInputStream correctly deals with null input; r=froydnj (b8afd03ec)
 - Bug 1059081 - Fix non-unified bustage. a=bustage (135927526)
 - Bug 1059081 - More non-unified bustage fixage. (d8fd4ad65)
 - Bug 1059081 - Add a threadsafe wrapper for persistent nsMultiplexStream queues. r=nfroyd (1060af97f)
 - Bug 1163029 - Use NamedDecl::getName instead of expensive NamedDecl::getNameAsString. r=ehsan (97a8a32a6)
 - Bug 1156084 - Disallow AddRef() and Release() calls on the return value of methods returning XPCOM objects; r=jrmuizel (96f6567c8) (8aa25c2df)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1163861 - Pass window IDs in nsILoadInfo (r=ckerschb) (4cd150a88)
 - fix of extra of Bug 1137287 - Part 2: (69fbdfbe0)
 - Bug 1159280 - TSan: data race netwerk/protocol/websocket/WebSocketChannel.cpp:3156 WebSocketChannel::Close, r=mcmanus (306796e71)
 - resources in browser, not global (f55b0a903)
 - Bug 1163268 - Remove UI options for recording with memory and jit optimizations. r=vp (56e82c90a)
 - Bug 1163384 - Add 'devtools-toolbarbutton' class to performance tool buttons to make them correct themed on Linux. r=jsantell (c3ce44795)
 - Bug 1141313 - Reenable several disabled performance tool tests that are now resolved. r=vp (3585f6bc1)
 - Bug 1146239 - Show the recording as loading in the performance sidebar after it has stopped; r=jsantell (4ee8200ea)
 - Bug 1146239 - Show the recording as loading in the performance details pane after it has stopped; r=jsantell (137a5f810)
 - Bug 1160313 - Remove temporary retro-mode for performance tools. r=vp (0131c940d)
 - Bug 1164281 - Make a new pref devtools.performance.ui.experimental enabled on nightly to show experimental options. r=vp (1a3f90404)
 - Bug 1162583 - Disable realtime rendering in performance tools when e10s is not on. r=vp (8f138f0dc) (8ed22681d)
- remove nsINetUtil_ESR_38, fixup and import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1150366 - Remove duplicate d3.js. r=jsantell (71f991da6)
 - Bug 1175352 - Refactor LoadInfo arguments to be more self contained (r=bent) (3f7d08164)
 - Bug 1181533 - Add nsCRTGlue.h and prtime.h includes to nsDownloadManager.cpp to fix --disable-safe-browsing build bustage. r=froydnj (af338edf1)
 - Bug 905127 - Part 1 - Make some functions from nsNetUtil not inline. r=jduell Bug 905127 - Part 2 - remove unnecessary nsNetUtil.h includes r=jduell Bug 905127 - Make some functions from nsNetUtil not inline. r=jduell (bc47d5b3a)
 - Bug 1117650 - Part 1: Move all CORS tests into dom/security/test. r=sicking (6f727387c)
 - Bug 1117650 - Part 2: Move all MixedContent tests into dom/security/test. r=tanvi (73f273829)
 - Bug 1004703 - ignore 'unsafe-inline' if nonce- or hash-source specifed - tests (r=sstamm) (ebf411159)
 - Bug 921493 - CSP: test whitelisting of scheme-relative sources (r=dveditz) (f0618b189)
 - Bug 1139569 - Optimize update function for element editors in markup view;r=mratcliffe (c60d38dca)
 - Partil of Bug 1139644 - Make markupview use sdk timers instead of individual timers for each window;r=pbrosset (8bc1a4d66)
 - Bug 1146568 - Avoid unsafe CPOW usage warnings in browser_markupview_tag_edit_08.js; r=bgrins Making use of the devtools test frame-script to set the test node's attributes instead of using a CPOW. (61d7eed64)
 - Bug 1146568 - Avoid unsafe CPOW usage in markupview tests, in assertAttributes helper; r=bgrins (efc492d5d)
 - Bug 858038 - Allow moving elements in the markupview by drag/drop; r=pbrosset (51d79c15c)
 - Bug 1139644 - Flash only relevant attributes in markup view when changed;r=pbrosset (29807fa46)
 - Bug 1147128 - Make sure attribute shows up in markup view after removing and setting to the previous value;r=mratcliffe (1947f7417)
 - Bug 1147325 - Clear box model timer on markup view destroy. r=pbrosset Prevents 'this._inspector.toolbox is null' spam in mochitest-dt (1d6808b4a)
 - Bug 921102 - 1 - Linkify URIs in the inspector; r=tromey, miker This first part adds a parser for node attributes which, given some node information and an attribute name, generates a small AST-like array of objects that tells which parts of the attribute (if any) are links, and what they link to. Using this, the markup-view generates the right HTML structure to display these parts as links. This part 1 doesn't yet allow users to follow these links. (c500e270e)
 - Bug 1139667 - CSP tests for fetch(). r=ckirschb (df030457f)
 - Bug 1086999 - CSP: Asterisk (*) wildcard should not allow blob:, data:, or filesystem: when matching source expressions - tests (r=sstamm) (dee0f2239)
 - Bug 1117650 - Part 3: Move all CSP tests into dom/security/test. r=sstamm (29c95475c)
 - Bug 1146566 - 1 - Use devtools common frame-script in markupview tests and add helper; r=bgrins (b10e42468)
 - Bug 1146566 - 2 - Avoid using CPOWs in test browser_markupview_tag_edit_03.js; r=bgrins This change updates the browser_markupview_tag_edit_03.js test by making use of the getDomElementInfo message listener to retrieve information about the tested nodes without having to go through CPOWs. (07f0ee3ff)
 - Bug 1036324 - Adds option to walker.parents() to not traverse DocShellTreeItems of different types (24123a5d2)
 - Bug 921102 - 2 - Markup-view tests for attribute links; r=miker (f6da81799)
 - Bug 901250 - Add scroll into view menu item for the inspector. r=bgrins (bfc7518b1)
 - Bug 994055 - Part 1: Add a toggle sidebar panel button to the inspector. r=miker This adds a new icon, next to the searchbox (like in the debugger) to expand or collapse the sidebar panel in the inspector. The state is *not* persisted yet when the toolbox is closed. The button is hidden when the toolbox host switches to 'side'. (87730e20a)
 - Bug 994055 - Part 2: Add tests for the inspector sidebar toggle button. r=miker (2872c3fbd)
 - Bug 921102 - 3 - Open/copy markup-view attribute links; r=bgrins This part adds contextual menu items that become enabled when the user right clicks on an attribute that has a link. Depending on the nature of the link, a new tab will be opened or a node selected. The user can also choose to copy the link in the clipboard. (9b93485f6)
 - Bug 921102 - 4 - Tests for the open/copy links on markup-view attributes; r=bgrins (861ba51f6) (1e24b22ef)