Jump to content

My Browser Builds (Part 2)


Recommended Posts

On 12/7/2019 at 2:06 PM, msfntor said:

Yes, look to about:support page in RoyTam1's Firefox 45, under Graphics section: "Supports Hardware H264 Decoding    No; Failed to create H264 decoder" ...

This is quite normal for the browser running on Windows XP, and it will persevere even when you manually install h264 decoding support in that browser (installing Adobe Primetime CDM or standalone LAV Filters DLLs); that setting only checks for a GPU-provided h264 decoder, it implies you have WMF (Vista SP2+) and a compatible gfx card; my old (end of 2006) Vista laptop I'm posting with doesn't come with such a card (with a H/W h264 decoder), so, while I don't need either APCDM nor LAV Filters for decoding purposes, the browser uses Vista's native WMF decoder only in software mode, thus heavily taxing my equally old Core 2 Duo processor... :(

Link to comment
Share on other sites


On 12/7/2019 at 11:33 PM, msfntor said:
On 12/8/2019 at 6:51 PM, msfntor said:

Could you download Zenmate (or Browsec), that are free extensions, and get your impressions here, please? Thank you!

I'm not on XP (as you know already), but the last "legacy" version of the Browsec VPN (actually, proxy) extension (v3.16.16) does install and work as expected under Serpent 52.9.0:

ABaNXbW.jpg

BTW, I used CAA to access it (caa:addon/browsec) ;)

Edited by VistaLover
Link to comment
Share on other sites

9 hours ago, VistaLover said:
  On 12/7/2019 at 1:06 PM, msfntor said:

Yes, look to about:support page in RoyTam1's Firefox 45, under Graphics section: "Supports Hardware H264 Decoding    No; Failed to create H264 decoder" ...

 

9 hours ago, VistaLover said:

This is quite normal for the browser running on Windows XP, and it will persevere even when you manually install h264 decoding support in that browser (installing Adobe Primetime CDM or standalone LAV Filters DLLs); that setting only checks for a GPU-provided h264 decoder, it implies you have WMF (Vista SP2+) and a compatible gfx card; my old (end of 2006) Vista laptop I'm posting with doesn't come with such a card (with a H/W h264 decoder), so, while I don't need either APCDM nor LAV Filters for decoding purposes, the browser uses Vista's native WMF decoder only in software mode, thus heavily taxing my equally old Core 2 Duo processor... :(

Yes, Windows XP here, so what do you think about WMF items in my about:config (media.wmf.enabled in default, true this one, others in false)?...Set it to false? or reset it? No problem?

Thank you @VistaLover for your detailled, always interesting comments!

Link to comment
Share on other sites

I use Win10 at work but use XP x86, XP x64, and 2003 x86 at home.

I seem to prefer a build that nobody here talks about so I finally created an account versus just watching and reading (often quite an entertaining read).

I use Roytam's ARCTIC FOX build pretty much exclusively at home (a portable Chromium v48 runs my YouTube-to-bigscreen-TV box).

I hope that ARCTIC FOX doesn't get lost in all of this.

 

LOVE THE WORK, ROYTAM.

Please don't let any of the "back-and-forth" discourage you in any way :whistle:

Link to comment
Share on other sites

11 hours ago, VistaLover said:

BTW, I used CAA to access it (caa:addon/browsec)

I don't have CAA (Classic Add-ons Archive: https://github.com/JustOff/ca-archive ) extension, but why downloading it, cause I've downloaded from legacycollector WaybackMachine link for Browsec (this same version browsec-3.16.16.xpi): https://web.archive.org/web/20191029180612/https://legacycollector.org/firefox-addons/603434/index.html

I can download these (Browsec, Zenmate) - then window after click on icon confirm that all is good, but it's not true cause all links show insecure page - "Peer's Certificate has expired" problem. "Secure Connection Failled - The connection to the server was reset while the page was loading" ...on all websites

privacy.firstparty.isolate have true (user set)

security.tls.version.max have 4 (default)

network.proxy.type have 5 (default)

network.proxy.proxy_over_tls true (default) - all other network.proxy are default, but:

network.proxy.no_proxies_on have user set string: localhost, localdomain, .localdomain, local, .local, 127.0.0.1, 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16, www.google-analytics.com - all these on Serpent 55, but not set by me, but by Browsec I think (that is disabled of course).

In my Windows XP, I've deleted in Registry under Windows NT whole AeDebug key ...

So I don't use these or another proxies, but I don't bother much about...

Thank you!

Edited by msfntor
network.proxy.type
Link to comment
Share on other sites

9 hours ago, msfntor said:

XP here, so what do you think about WMF items in my about:config (media.wmf.enabled in default, true this one, others in false)?...Set it to false? or reset it? No problem?

It doesn't matter, because Windows Media Foundation doesn't exist in XP. Microsoft added that feature in Vista.

On Win 7, I often set media.wmf.enabled to false for testing (so the browser won't look for it, forcing it to use alternate media decoders such as Primetime). You might also set it false if your Windows version has a WMF bug you find intolerable; otherwise I'd leave it at true for normal browsing.

Link to comment
Share on other sites

9 hours ago, ArcticFoxie said:

I use Win10 at work but use XP x86, XP x64, and 2003 x86 at home.

I seem to prefer a build that nobody here talks about so I finally created an account versus just watching and reading (often quite an entertaining read).

I use Roytam's ARCTIC FOX build pretty much exclusively at home (a portable Chromium v48 runs my YouTube-to-bigscreen-TV box).

I hope that ARCTIC FOX doesn't get lost in all of this.

 

LOVE THE WORK, ROYTAM.

Please don't let any of the "back-and-forth" discourage you in any way :whistle:

I don't think that @roytam1's Arctic Fox for XP, FF 45 for SSE PCs, or K-Meleon builds are affected at all. Arctic Fox already has its own branding.

Link to comment
Share on other sites

The "branding" should be treaded lightly.

I used GreenBrowser + Proxomitron exclusively from 2003/2004 all the way thru 2017/2018.

Tabbed browsing using the IE6 engine long before tabbed IE7 showed up (company forced IE at the time).

Company would prevent "installs" but I could run "portable" apps without issue (don't tell the IT guy).

That worked for me for a VERY long time.

But...  Proxomitron was "high maintenance".

 

I was fine with all of the "maintenance" - it was the web MY WAY and not the way that "script kiddies" threw it at me.

But...  It became too much of a hassle to edit and tweak for a couple credit card websites at the time.

 

That led me to abandoning GreenBrowser + Proxomitron.

My search led me through Maxthon, SlimBrowser, Vivaldi, Chrome, Chromium, Iron, Opera, Firefox, Flock, Waterfox, Icefox, IceDragon, Avant, Sleipnir, among many others I'm sure that I'm forgetting as I look back.

 

Sleipnir and Chromium lasted the longest as far as my exclusives at the time.

I eventually landed on a precursor (?) to New Moon.

I forget the name of it but the "branding" was a RACCOON waiving his paw.

Sounds "trivial" perhaps, but the RACCOON turned me away from that browser.

Despite its functionality meeting ALL of my needs.

The "branding" cut my time short with my trial of that browser.

 

My home CPU is an i7 and came with Win 10 LTSB - would never do anything 10 without it being LTSB.

But I opted to keep XP x64 instead.

My computer can "handle" more resource-hungry web browsers - but it is NOT my "preference".

 

So I ended up still hunting and stumbled into Roytam's builds.

Ran v28 for a while, but reverted in favor of v27.

Then arrived Arctic Fox Win32 - been my exclusive ever since.

 

Edit:  I'll add this.  One of my gripes against v28 is because I do *not* 'want' to update each and every week.  That borders again on "high maintenance".  I would prefer kind of knowing which weekly release is best associated with the "stable" upstream versus the "nightly" upstream.  I don't really have a need to update my browser each and every week.

Edited by ArcticFoxie
update history
Link to comment
Share on other sites

35 minutes ago, ArcticFoxie said:

I don't really have a need to update my browser each and every week.

no you don't really need to update it every week. just update it when you feel you need an update for the browser you use.

even I build every week, I don't update my "daily-used" copy every week, even I preliminary tested them in separated position.

Link to comment
Share on other sites

New build of Serpent/UXP for XP!

Test binary:
Win32 https://o.rths.ml/basilisk/basilisk52-g4.5.win32-git-20191221-fd382bb-uxp-d6baead6c-xpmod.7z
Win64 https://o.rths.ml/basilisk/basilisk52-g4.5.win64-git-20191221-fd382bb-uxp-d6baead6c-xpmod.7z

source code that is comparable to my current working tree is available here: https://github.com/roytam1/UXP/commits/custom

IA32 Win32 https://o.rths.ml/basilisk/basilisk52-g4.5.win32-git-20191221-fd382bb-uxp-d6baead6c-xpmod-ia32.7z

source code that is comparable to my current working tree is available here: https://github.com/roytam1/UXP/commits/ia32

NM28XP build:
Win32 https://o.rths.ml/palemoon/palemoon-28.9.0a1.win32-git-20191221-806d807dc-uxp-d6baead6c-xpmod.7z
Win64 https://o.rths.ml/palemoon/palemoon-28.9.0a1.win64-git-20191221-806d807dc-uxp-d6baead6c-xpmod.7z

Official UXP changes since my last build:
- Remove Pale Moon from the Unified XUL Platform repository (4e34c5746)
- Remove Pale Moon 7zstub (094e74ff6)
- Update metadata in UXP 7z-stub installer (e6a3515a1)
- Update 7z-stub README to reflect the current state of its use. (e7d4713e0)
- Remove Basilisk from the Unified XUL Platform repository (06494f307)
- Issue #1323 - Move XULRunner back to topsrcdir (e2de507e0)
- Bug 1317376 - Part 1: Remove unreachable code and remnants from the self-hosted implementation. (e23b013ad)
- Bug 1317376 - Part 2: Detect Promise self-resolution when resolving through the Promise resolving fast path. (f589ef816)
- Bug 336705 - Part 1: Support creating and resolving Promises without resolve/reject functions. (cb732e5fd)
- Bug 1336705 - Part 2: Add self-hosting intrinsics for resolving/rejecting Promises and adding reactions. (1fd726c6b)
- Bug 1343481 - Part 1: Remove {JSFunction,JSScript,LazyScript}.isGenerator() method. (9163aaebb)
- Bug 1343481 - Part 2: Stop using StarGegerator for async function. (f07d5707c)
- Bug 1343481 - Part 3: Add JSOP_AWAIT and rename {yieldIndex,yieldOffset} to {yieldAndAwaitIndex,yieldAndAwaitOffset}. (3a3de55aa)
- Bug 1343481 - Part 4: Add Add GeneratorObject.{isAfterYield,isAfterAwait}. (73c873200)
- Bug 1343481 - Part 5: Rename AsyncFunction-related names in Promise.cpp to explicitly say Async Function. (4f88fc851)
- Bug 1343481 - Part 6: Add native functions wrapper for GetInternalError and GetTypeError. (cba3f6142)
- Bug 1343481 - Part 7: Add BytecodeEmitter::emitDotGenerator and make yield/await nodes unary. (79b5eb14b)
- Bug 1316098 - Optimize out result object allocation for await/return in async function. (d0d9a4f43)
- Bug 1344753 - Update for-of stack depth in ControlFlowGenerator::processWhileOrForInLoop. (f5fa6cfea)
- Bug 1317389: Change property attributes for generator and async functions to match ES2015/2017. (f87b1b885)
- Bug 1331092 - Part 0: Define NOMINMAX to avoid compile error from min/max macro on windows. (5e1f00240)
- Bug 1331092 - Part 1: Add Symbol.asyncIterator. (82f9efff9)
- Bug 1331092 - Part 2: Implement Async Generator except yield*. (d5086ac3a)
- Bug 1331092 - Part 2: Implement Async Generator except yield*. (8e2f6a759)
- Bug 1331092 - Part 6: Support JSOP_TOASYNCGEN in JIT. (bbd1fef78)
- Bug 1331092 - Part 7: Implement Async Generator yield*. (ef44324d9)
- Bug 1331092 - Part 8: Support JSOP_TOASYNCITER in JIT. (580dfb2af)
- Bug 1331092 - Part 9: Implement for-await-of. (5bdfdb678)
- Bug 1331092 - Part 11: Await on the innerResult.value when innerResult.done is true in yield*. (55728c964)
- Bug 1355399 - Switch property retrieval in Async-from-Sync Iterator prototype methods. (63eee6325)
- Bug 1364608 - Stash rval in AsyncIteratorClose. (dd1cbde15)
- Bug 1379525 - Part 1: Await on the value before yielding or returning inside async generator. (38c1d558a)
- Bug 1379525 - Part 2: Properly handle rejection in async-from-sync iteration. (4a5ed79c6)
- Bug 1390082 - Implement AsyncGeneratorQueue with simpler array operations. (e0eee318c)
- Bug 1352312 - Enable Async Iteration. (c9bddfca7)
- Bug 1454285 - Part 1: Specify the current scope when emitting await and .generator. (c66d25512)
- Bug 1454285 - Part 2: Disallow using innermostEmitterScope while the value does not match the bytecode environment. (6c3e42ac6)
- Merge pull request #1327 from g4jc/async_iteration (07d0bcbf1)
- No Issue - Fix indentation and account for system libevent in ipc/chromium/moz.build (a3e93d08d)
- Issue #1328 - Part 1: Change default monospace font to Consolas for all localizations where this would be set to Courier New. (342f94295)
- Issue #1328 - Part 2: Change tests that require Courier New explicitly. (0f67e7b22)
- Issue #1328 - Part 3: Add fuzz to reftests and fix misc other tests. (eea149e73)
- Merge branch 'master' of https://github.com/MoonchildProductions/UXP (11ce27f0d)
- Issue #1328 - Follow-up: Increase font size for changed monospace fonts. (8220776c9)
- Issue #1322 - Part 1: Remove the DOM Promise guts. (fb96b43b5)
- Issue #1322 - Part 2: Remove --enable-sm-promise buildconfig switch. (cdea310f1)
- Issue #1322 - Part 3: Remove DOM promise interface gunk (d46126240)
- Issue #1322 - Part 4: Fix WebIDL.py parser for line endings and wrong IDL name (c7ae364a8)
- Merge pull request #1330 from MoonchildProductions/DOM-promise-removal (3332f584c)
- Issue #1316 - Reduce compiler warnings in ICU (368170bdc)
- Enable layers acceleration by default on GTK platforms. (ce250e9cc)
- Issue #1219 - Align computed DOM styles with mainstream behvior. (d6baead6c)

Official Pale-Moon changes since my last build:
- Remove Platform Code (846fcb459)
- Move browser/ to palemoon/ (93c20bef5)
- Overlay Pale Moon 28 (1f08c8017)
- Version bump (Start Pale Moon 29) (509cbf622)
- Add Comm Build System (75dd33577)
- Add platform submodule (e573bfa52)
- Add a mach stub (20ea829d0)
- Update commit pointer (1afb8b2ec)
- Re-create README.md (8c1e4fb00)
- Add 7zstub (b2577a078)
- Merge branch 'master' of https://github.com/MoonchildProductions/Pale-Moon (d8661ecb1)
- Make Pale Moon build in a comm configuration (07a6051d2)
- Issue #1697 - Add executable bit to scripts. (980687b89)
- Merge pull request #1698 from trav90/buildsystem-work (9e6597608)
- Issue #1697 - Correct a couple of paths for Mac build (9d6b4534e)
- Merge pull request #1699 from adeshkp/mac-build-split-platform (af8643918)
- Update gitignore to account for someone being clever with ntfs junctions or other filesystem links ;) (af431d801)
- Update platform commit pointer (806d807dc)

Official Basilisk changes since my last build:
- Add git stuff (db902e7)
- Add Comm Build System (3d5a69a)
- Update readme (5a343e9)
- Add executable bit to scripts (744b3b2)
- Add platform submodule (0e3a8df)
- Update readme (c2eb348)
- Update readme ONE MORE TIME (ab111e9)
- Update gitignore to account for someone being clever with ntfs junctions or other filesystem links ;) (67f840a)
- Make Basilisk build in a comm configuration (fd382bb)

My changes since my last build:
- The front-end(PM/BK/etc.)/back-end(UXP Goanna engine) separation works is not happened in my repo for the time being, but for tracking their changes I put my portions over their repo in my working copy. If the rebranding really happens here, front-end/back-end separation works will happen here and there may be an unified front-end repo for my front-end works.

Edited by LoneCrusader
fix NM win32 link
Link to comment
Share on other sites

New build of BOC/UXP for XP!

Test binary:
MailNews Win32 https://o.rths.ml/boc-uxp/mailnews.win32-20191221-beb2221f-uxp-d6baead6c-xpmod.7z
Browser-only Suite Win32 https://o.rths.ml/boc-uxp/bnavigator.win32-20191221-beb2221f-uxp-d6baead6c-xpmod.7z

source patch (excluding UXP): https://o.rths.ml/boc-uxp/boc-uxp-src-xpmod-20191123.7z

Official repo changes since my last build:
- Update gitignore so that Pale Moon and Basilisk will not be tracked if they are symlinked/junction'd into the tree (855250ff)
- Update commit pointer (beb2221f)

For UXP changes please see above.

Link to comment
Share on other sites

new ArcticFox win32 test build is uploaded: http://o.rths.ml/gpc/files1.rt/arcticfox-27.9.19.win32-20191221.7z

- Bug 1134518 - Add test to ensure we don't lose shistory entries when switching a browser's remoteness r=Mossop (f7bd1b4e2)
- Bug 1134518 - Use the globalMM for ContentTask's frame script to prevent it from being loaded twice into a docShell swapped to another browser window r=bustage (da4f3899c)
- Bug 883609 - Porting tests to new Session Restore backup mechanism;r=ttaubert (f53b11a6f)
- Bug 1134518 - Add xpcshell test for shistory caps on clean shutdown and fix existing tests r=Yoric (3ed696a4c)
- Bug 1077738: Retain whether history entries are set to persist in session history and restore that. r=smacleod (a0be00dc4)
- update (2959bddc8)
- Merge pull request #19 from wicknix/master-good (cda043440)
- Merge pull request #20 from rmottola/master (ed9f72c2e)
- fix duplicated IsWin10OrLater() function (02510748c)
- add missing bit of Bug 1134280 - Get rid of Tag() - patch 1 - Is{HTML,XUL,MathML,SVG}Element and IsAnyOf{HTML,XUL,MathML,SVG}Elements, r=smaug (4b86381cc)
- Bug 960959 - Use RunState to keep track of the SessionFile's closed status r=yoric (a4cb3aaa3)
- remove telemetry initialization part, not needed anyway: removes warning (dd0ef756d)
- Bug 1134518 - Send a state object instead of a string to SessionWorker.write() r=Yoric (bd5b96e4d)
- Bug 1157235 - Remove unused sessionstore-state-write notification r=Yoric (884843b11)
- Bug 1150529 - Remove code for expired telemetry histograms r=Yoric (b86b58e55)
- add specific stuff that could hurt mac (156375c79)
- clean (904cbaa94)
- Bug 1073502 - Restore closed windows even when there are no open windows on startup r=yoric (2c12eea64)
- Bug 940777 - Simplify about:blank handling in session restore (r=ttaubert) (5298de7ed)
- Bug 1135498 - Get rid of the unused browser.__SS_restore_data property r=smacleod (a87fabb3f)
- Bug 923315 - Ensure that sessionstore.js writes don't interfere with browser_upgrade_backup.js r=yoric (31c6a52c0)
- Bug 883609 - Porting tests to new Session Restore backup mechanism;r=ttaubert (7c04c1156)
- Bug 1107941 - Removing old upgrade backups if too many exist. Max upgrade backups is defined in preferences. Tested with mochi tests. r=Yoric (857fa2844)
- Bug 1134518 - Cap shistory entries in the SessionWorker when shutting down r=Yoric (f0a9f43ac)
- Bug 1100223 - Make calling loadURI() on pending tabs work as expected by marking the tab as no longer pending and simply waiting for the restoration to finish r=billm (5f4f766b7)
- Bug 1152341 - Failure to read one of the session file candidates shouldn't stop us from trying further r=Yoric (ff1bccd2f)
- Bug 1034075 - Add more details to SessionFile AsyncShutdown blocker. r=ttaubert (fc0aace02)
- Bug 1031298 - Add a histogram for counting unrecoverable sessions. r=yoric (95b377fef)
- more stuff to do (044d0dbff)
- revert default of showQuitWarning because it exposes a bug which corrupts the profile or session storage (714e4eb78)
- Bug 1056443 - Use correct URL in location bar when switching processes (r=ttaubert) (7e6a54793)
- Bug 1109875 - When unloading a frameLoader flush all pending state changes by sending a final update message r=billm (4264bb7d1)
- Bug 1161928 - Move epoch handling from ContentRestore.jsm to content-sessionStore.js r=billm (9bc20bf8a)
- Bug 1145942 - Use the frame message manager to revive crashed tabs upon navigation r=billm (a2ee3704d)
- Bug 1161928 - Require an epoch (managed in the parent) included in every message sent by the frame script to get rid of TabState.flush() calls in restoreTab() r=billm (11a126155)
- Bug 1169892 - Avoid extra round-trip when a pending tab is reloaded r=billm (569786f54)
- Bug 1125411 - OSX: turn CoreLocation provider on, r=cpeterson Also, expose this setting in about:config (fe7fb5620)
- Bug 1136956 - Set CoreLocation provider to non-release only in firefox.js. - r=cpeterson (7e971cf02)
- update (110be9f3f)
- Merge remote-tracking branch 'upstream/devel' into fix-winbuild (11f9fe3b2)

Edited by roytam1
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...