My Browser Builds (Part 4)

[Mima]

(duplicated post, ignore)

Edited May 3, 2023 by Mima
dupe

[silverni]

 

8 hours ago, Mima said:

You only need to copy the relevant errors from the console. If the same error repeats over and over, then don't copy them all; just copy a single instance of it and tell us that that error is repeating. Another pro-tip is that most of the warnings (like the CSP errors marked as warnings) you see is mostly irrelevant and can be ignored.

You can do that, but I think this forum software should support "spoiler" tags which should automatically hide the contents of the console output you've pasted. Put it within a code block for better readability and nest it within a spoiler tag for best results.

Please don't. That's bad netiquette.

Btw, you can access the "F12 Browser Console" with the F12 key, which will toggle the developer tools at the bottom of your browser. Not sure why you didn't get that from the very start...

In my NM28, F12 does nothing.

Regards

[roytam1]

so things are merged in upstream, while misskey.io problem goes to new issue: https://repo.palemoon.org/MoonchildProductions/UXP/issues/2229

and it seems that curent tip build having issue on gmail.

Edited May 4, 2023 by roytam1

[Mima]

17 hours ago, silverni said:

In my NM28, F12 does nothing.

NM doesn't have devtools? Does Ctrl-Shift-J do anything? Does the Developer Tools submenu show up from the Tools menu of the menubar?

17 hours ago, silverni said:

<screenshot>

I don't think there's anything remotely relevant here other than the error at the very bottom... And even that doesn't seem relevant because it shows up for me too while logged out. Btw when I said "find relevant errors", I mean find those which come from the website's code itself, like the jquery error shown in the screenshot. NS_ERROR_CONSOLE errors are mostly irrelevant as they're just errors related to internal browser code of New Moon itself. Same goes with the browser.xul warnings and l10n.js error.

Can you try enabling layout.css.report_errors from about:config, clear the console output, and refresh the page while logged in to that dashboard? I'm suspecting this issue has more to do with CSS than JavaScript. It will probably produce a lot more output than before though (which is why this about:config option is disabled by default).

[Saphir]

7 hours ago, roytam1 said:

And it seems that curent tip build having issue on gmail.

@roytam1what exactly happens/is breaking? 

[roytam1]

2 hours ago, Saphir said:

@roytam1what exactly happens/is breaking? 

clicking on any mail from mail list doesn't open / can't back to mail list when clicking on back button on gmail's toolbar.

but after Ctrl-F5 I can't reproduce them, weird.

Edited May 4, 2023 by roytam1

[Mima]

54 minutes ago, roytam1 said:

but after Ctrl-F5 I can't reproduce them, weird.

Gmail itself probably got updated and the cache is causing conflicts, which is probably why the bug no longer happens for you after bypassing the cache. It happens. :P

[silverni]

5 hours ago, Mima said:

NM doesn't have devtools? Does Ctrl-Shift-J do anything? Does the Developer Tools submenu show up from the Tools menu of the menubar?

I don't think there's anything remotely relevant here other than the error at the very bottom... And even that doesn't seem relevant because it shows up for me too while logged out. Btw when I said "find relevant errors", I mean find those which come from the website's code itself, like the jquery error shown in the screenshot. NS_ERROR_CONSOLE errors are mostly irrelevant as they're just errors related to internal browser code of New Moon itself. Same goes with the browser.xul warnings and l10n.js error.

Can you try enabling layout.css.report_errors from about:config, clear the console output, and refresh the page while logged in to that dashboard? I'm suspecting this issue has more to do with CSS than JavaScript. It will probably produce a lot more output than before though (which is why this about:config option is disabled by default).

error console screenshot 1.3MB , link from Onedrive:

https://1drv.ms/u/s!AmvyhfGl8qmbgXddu90uCBgyvC5G?e=NuWXT8

Regards

 

Edited May 4, 2023 by silverni

[Mima]

1 hour ago, silverni said:

error console screenshot 1.3MB

Well that wasn't much help for me either. I can't see anything obvious that would cause the page to go blank like that. There are some familiar errors like min() and unprefixed user-select not being recognized (as UXP doesn't support them yet) and therefore dropped, but they really shouldn't break the page (and those errors appear while logged out anyway, so most likely irrelevant)...

Anyway you did try testing in a fresh profile, right? Otherwise your only option might just be to contact the webmaster about it since I can't really reproduce it (I don't have an account) here in my trunk build of Pale Moon, so I can't provide much help than this I'm afraid.

Edited May 4, 2023 by Mima

[silverni]

8 minutes ago, Mima said:

Well that wasn't much help for me either. I can't see anything obvious that would cause the page to go blank like that. There are some familiar errors like min() and unprefixed user-select not being recognized (as UXP doesn't support them yet) and therefore dropped, but they really shouldn't break the page (and those errors appear while logged out anyway, so most likely irrelevant)...

Anyway you did try testing in a fresh profile, right? Otherwise your only option might just be to contact the webmaster about it since I can't really reproduce it (I don't have an account) here in my trunk build of Pale Moon, so I can't provide much help than this I'm afraid.

The error does not occur in versions of the browsers until February, it only affects the latest versions of the browsers; it is a problem of the browsers, not of the site.

Regards

 

[Mima]

4 minutes ago, silverni said:

it is a problem of the browsers, not of the site.

You can't really be sure about that, as we don't really have enough details to come to the conclusion that it's a regression caused by the platform updates.

Anyway, last workaround I can think of: try disabling WebComponents by setting dom.webcomponents.enabled to false via about:config. We've enabled WC by default since March because most of it is working, but the implementation is still not a full 100% complete (more like 95% right now) If that works, then that would mean the website has a pretty poor fallback when the browser doesn't 100% support the WebComponents package, which isn't good practice.

[roytam1]

1 hour ago, Mima said:

Well that wasn't much help for me either. I can't see anything obvious that would cause the page to go blank like that. There are some familiar errors like min() and unprefixed user-select not being recognized (as UXP doesn't support them yet) and therefore dropped, but they really shouldn't break the page (and those errors appear while logged out anyway, so most likely irrelevant)...

Anyway you did try testing in a fresh profile, right? Otherwise your only option might just be to contact the webmaster about it since I can't really reproduce it (I don't have an account) here in my trunk build of Pale Moon, so I can't provide much help than this I'm afraid.

upstream filed an issue for css min()/max()/clamp(): https://repo.palemoon.org/MoonchildProductions/UXP/issues/2230

[silverni]

8 hours ago, Mima said:

Well that wasn't much help for me either. I can't see anything obvious that would cause the page to go blank like that. There are some familiar errors like min() and unprefixed user-select not being recognized (as UXP doesn't support them yet) and therefore dropped, but they really shouldn't break the page (and those errors appear while logged out anyway, so most likely irrelevant)...

Anyway you did try testing in a fresh profile, right? Otherwise your only option might just be to contact the webmaster about it since I can't really reproduce it (I don't have an account) here in my trunk build of Pale Moon, so I can't provide much help than this I'm afraid.

I reported the bug to give way and time for the developers to fix it, but it is not a big deal to me.
I have two versions of NM28 on disk, the one from February and the one from April, and I launch one or the other depending on the circumstances.
By the way from WinXP I can boot a Win7 virtual machine, and in this one I can boot Palemoon, Firefox, and other NT6 browsers.
Regards

 

[roytam1]

New build of Serpent/UXP for XP!

Test binary:
Win32 https://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20230506-3219d2d-uxp-725b27a0f-xpmod.7z
Win64 https://o.rthost.win/basilisk/basilisk52-g4.8.win64-git-20230506-3219d2d-uxp-725b27a0f-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.rthost.win/basilisk/basilisk52-g4.8.win32-git-20230506-3219d2d-uxp-725b27a0f-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.rthost.win/palemoon/palemoon-28.10.6a1.win32-git-20230506-d849524bd-uxp-725b27a0f-xpmod.7z
Win32 IA32 https://o.rthost.win/palemoon/palemoon-28.10.6a1.win32-git-20230506-d849524bd-uxp-725b27a0f-xpmod-ia32.7z
Win32 SSE https://o.rthost.win/palemoon/palemoon-28.10.6a1.win32-git-20230506-d849524bd-uxp-725b27a0f-xpmod-sse.7z
Win64 https://o.rthost.win/palemoon/palemoon-28.10.6a1.win64-git-20230506-d849524bd-uxp-725b27a0f-xpmod.7z

Official UXP changes picked since my last build:
- Remove unmaintained WebGL conformance test suite. (733451d1b)
- Revert "Issue #61 - Place Skia in libxul" (260f8e154)
- Issue #61 - Follow-up: Fix gkmedias.dll link bustage related to Skia. (6e2f70a70)
- Issue #61 - Follow-up: Fix xul.dll link bustage related to Skia. (4584069f2)
- No issue - Correct handling of async (arrow) functions declared inside constructors (ffe6d48af)
- No issue - [MailNews] Fix build bustage due to recent MIME API changes (2048f4df5)
- Issue #1451 - Split hunspell library out of xul. (cbbc15ad8)
- Issue #1451 - Fix hunspell build bustage in FreeBSD. (d75c6c181)
- Issue #1691 - Part 1: Provide a way of associating a private value with a script or module. (1a6b3a822)
- Issue #1691 - Part 2: Implement call import and import meta in the parser. https://bugzilla.mozilla.org/show_bug.cgi?id=1427610 https://bugzilla.mozilla.org/show_bug.cgi?id=1484948 (a8ab41b4c)
- Issue #1691 - Part 3: Finish implementing import meta. https://bugzilla.mozilla.org/show_bug.cgi?id=1427610 (3f4985ce6)
- Issue #1691 - Part 4: Finish implementing call import. https://bugzilla.mozilla.org/show_bug.cgi?id=1499140 (7a6dab0b3)
- Issue #1691 - Part 5: Don't pre-create module metadata object when compiling. https://bugzilla.mozilla.org/show_bug.cgi?id=1489477 (c965fbdd1)
- Issue #1691 - Part 5b: Stop modules from entraining the top-level JSScript. https://bugzilla.mozilla.org/show_bug.cgi?id=1489477 (5c09a3b3c)
- Issue #1691 - Part 6a: Support private values which contain pointers to cycle-collected C++ objects https://bugzilla.mozilla.org/show_bug.cgi?id=1342012 (1b811bd71)
- Issue #1691 - Part 6b: Initial browser support for dynamic import from module scripts. https://bugzilla.mozilla.org/show_bug.cgi?id=1342012 Factor out script fetch options from script load request classes. https://bugzilla.mozilla.org/show_bug.cgi?id=1480720 Remove support for version parameter from script loader. https://bugzilla.mozilla.org/show_bug.cgi?id=1428745 (86e0057ea)
- Issue #1691 - Part 6c: Fix a few issues with the earlier commits. Need PNK_CALL_IMPORT and PNK_IMPORT_META in the node children list or it will abort. When porting ScriptFetchOptions support, I had 2 constructors, I picked the wrong one. Missed adding the preference javascript.options.dynamicImport to all.js. (9a19e9d16)
- Issue #1691 - Part 6d: Optimize handling of internally-created Promise objects. https://bugzilla.mozilla.org/show_bug.cgi?id=1358879 The patch uses a different test: PROMISE_FLAG_DEFAULT_REJECT_FUNCTION which doesn't exist in our codebase. It was added in this bug, which appears to be partly complete: https://bugzilla.mozilla.org/show_bug.cgi?id=1313049 (c3daecc21)
- Issue #1691 - Part 6e: Fix problems due to divergent codebases. This gets basic dynamic import working. Fix a problem in Part 5b where Mozilla used toGCThing() and we don't. Fix a problem in Part 4 where runtime() returns nullptr in our codebase since it runs on JS Helper thread. We need to get the runtime via runtimeFromAnyThread() instead. (e7d0b58e5)
- Issue #1691 - Part 7a: Add a JS API to get the private value for the calling script or module. https://bugzilla.mozilla.org/show_bug.cgi?id=1342012 (62be54216)
- Issue #1691 - Part 7b: Make load request element optional. https://bugzilla.mozilla.org/show_bug.cgi?id=1342012 (ba01d99e3)
- Issue #1691 - Part 7c: Refactor ModuleScript into ClassicScript class and LoadedScript base class so we can represent all scripts that can perform dynamic import. https://bugzilla.mozilla.org/show_bug.cgi?id=1342012 (41eda05e5)
- Issue #1691 - Part 7d: Allow dynamic import in cases where there's no referencing script or module. Support dynamic import from classic scripts by creating ClassicScript objects and associating them with the compiled. https://bugzilla.mozilla.org/show_bug.cgi?id=1342012 This patch is incomplete, some code in ScriptLoader::EvaluateScript() could not be applied for missing dependencies. Part 7e will apply the missing dependencies and finish the patch. (d5d7bb5e4)
- Issue #1691 - Part 7e: Dependencies for required to finish part 7d. https://bugzilla.mozilla.org/show_bug.cgi?id=1331662 Reimplement EvaluateString using the ExecutionContext class. https://bugzilla.mozilla.org/show_bug.cgi?id=1316078 Extract redudant code into StartOffThreadParseTask. Use an ExclusiveContext instead of a JSContext in XDR functions. Add a script decoder as a valid off-main-thread parse-task. https://bugzilla.mozilla.org/show_bug.cgi?id=900784 Add nsJSUtils functions for encoding and decoding the bytecode. https://bugzilla.mozilla.org/show_bug.cgi?id=1316081 Add XDRIncrementalEncoder to replace delazified LazyScript in the encoded XDR buffer. Add an XDRIncrementalEncoder instance on the ScriptSource. Expose a new JSAPI to incrementally encode bytecode when it is generated. https://bugzilla.mozilla.org/show_bug.cgi?id=1334091 XDR function use the sourceObject instead of the enclosingScript as argument. (993476283)
- Issue #1691 - Part 7f: Split up compile and execute so we can use ClassicScript. https://bugzilla.mozilla.org/show_bug.cgi?id=1342012 Refactor nsJSUtils::ExecutionContext to separate compilation and execution steps and allow extraction of compiled JSScript. https://bugzilla.mozilla.org/show_bug.cgi?id=1366773 Move buffer argument from JS::StartIncrementalEncoding to JS::FinishIncrementalEncoding. (c69a47c39)
- Issue #1691 - Part 8: Fix --enable-debug builds and continue dynamic module import changes. https://bugzilla.mozilla.org/show_bug.cgi?id=1342012 Support import from timeout handlers by associating the initiating script with the compiled JSScript. Fix error message that covers all import() failures that don't throw a JS exception. https://bugzilla.mozilla.org/show_bug.cgi?id=1331662 Partial - Replace nsJSUtils::EvaluateString calls by ExecutionContext scopes. Left EvaluateString() in nsXBLProtoImplField.cpp until ExecutionContext errors can be fixed. (1109559a5)
- Issue #1691 - Part 9: Make import() work when the active script is in another document. https://bugzilla.mozilla.org/show_bug.cgi?id=1342012 Associate event handler with active script when they are compiled. (be916ef7c)
- Issue #1691 - Part 10: Add and use method to annotate CC crashes with a class name. https://bugzilla.mozilla.org/show_bug.cgi?id=1277260 Make PtrInfo into a class and mark it final. Also fix an erroneous debug assert because mBaseURL not set in one code path. (7e056a6f7)
- Issue #1691 - Part 11: Fix incorrect reference counting in ModuleScript class. (edb82ec9f)
- Issue #1691 - Part 12: Fix return value in ExecScript() and debug assert in ParseTask. https://bugzilla.mozilla.org/show_bug.cgi?id=1331662 Replace nsJSUtils::EvaluateString calls by ExecutionContext scopes. ExecutionContext: The mRetValue is not used even though it is checked in various asserts. This is how it is in the Mozilla code even years later, only the passed in value is used. ParseTask: JoinDecode() and JoinCompile() run the same code but the type is different causing a debug assert when it checks the type Script vs ScriptDecode. (a892a7590)
- Issue #1691 - Part 13: Remove MUST_MATCH_TOKEN* macros in Parser. https://bugzilla.mozilla.org/show_bug.cgi?id=1501928 This was helpful in debugging one of the crashes, the macros made debugging a nightmare. (bc895eba9)
- Issue #1691 - Part 14: Fix a debug assert and memory leak. PNK_IMPORT_META and PNK_CALL_IMPORT are binary nodes... They function similar to list nodes, so they worked there, but debug mode asserts because of the wrong type. https://bugzilla.mozilla.org/show_bug.cgi?id=1342012 Fixed a memory leak due to missing code in SourceScriptObject::finalize(). Thanks FranklinDM! (79ea755b7)
- Issue #1691 - Part 15: Make |new import()| a syntax error. https://bugzilla.mozilla.org/show_bug.cgi?id=1508672 (f2b179156)
- Issue #1691 - Part 16: Dynamically imported modules can throw any value as an exception. https://bugzilla.mozilla.org/show_bug.cgi?id=1508672 (1ac1e90b6)
- Bug 1432272 - Make Fetch API use the global's base URL instead of the entry document's base URL (62e2fbc4e)
- Issue #2218 - Part 1: Add nsExpirationTracker::NotifyEndTransaction(Locked) callbacks for subclasses to know when an aging iteration is complete (fa787f4a6)
- Issue #2218 - Part 2: Make SurfaceCache free ImageSurfaceCache objects outside of the lock (93644fd33)
- Issue #2073 - Follow-up: Use internal Move instead of std::move for consistency (b7f217e5a)
- Issue #2213 - Follow-up: Fix derieved Generator prototypes after previous change (a27e8bf98)
- Issue #2173 - Follow-up: Use common ancestor of PNK_(OPT)DOT in ASTSerializer::expression (e31a058ce)
- Issue #1894 - Follow-up: Fix return value ordering in IonMonkey nullish coalescing (aece6260f)
- Issue #1285 - Follow-up: Correctly handle follow-up compilations of RegExp with named capturing groups (0a74a5758)
- Issue #2221 - Enable link-time optimization for Spidermonkey by default if building it shared. (b592711bb)
- Issue #1691 - Follow-up: Fix CallSelfHostedFunction deprot introduced by Part 4. (45b34592b)
- Issue #1862 - Follow-up: Replace deprecated Harfbuzz functions with current ones. (51426d105)
- Issue #1691 - Follow-up: use error message with no arguments for bad import statements (73a99c155)
- Issue #2142 - Fold BytecodeEmitter::checkTypeSet into BytecodeEmitter::emitCheck (014ebe8ae)
- Issue #2142 - Extend newSuperCall for JSOP_SPREADSUPERCALL (6f56a494e)
- Issue #2142 - Change InitPropertyOperation to accept a PropertyName directy and use DefineDataProperty (627d0da7d)
- Issue #2142 - Add predicate functions count_if and any_of to ListNode iterator (3b9b111b2)
- Issue #2142 - Ensure 'await' is always a restricted identifier when parsing modules (faed047e3)
- Issue #2142 - Reduce calls to FindReservedWord when checking for forbidden identifiers during parsing (6abb54ff1)
- Issue #2142 - Add internal option for fields, but always true (afa7bddc9)
- Issue #2142 - Implement syntax for public/private fields and computed field names (51db22ff2)
- Issue #2142 - Improve TokenPos handling in BCE (ab7721e4d)
- Issue #2142 - Add PropertyEmitter, ObjectEmitter, ClassEmitter, LexicalScopeEmitter, DefaultEmitter (1b89be6d0)
- Issue #2142 - Add FunctionEmitter, FunctionScriptEmitter, and FunctionParamsEmitter with current methods (e6335ded8)
- Issue #2142 - Handle fields in derived classes (235ca7794)
- Issue #2142 - Factor out PropertyName parsing from Parser::propertyName() (014953c53)
- Issue #2142 - Implement ASI for fields (447261cf8)
- Issue #2142 - Use JSOP_INITPROP for field initializers (849ab4417)
- Issue #2142 - Restrict contents of direct eval in fields (8c6750014)
- Issue #2142 - Fix several scoping issues in field initializers (bcb6203e4)
- Issue #2142 - Pass through arguments in synthesized constructors for derived classes (1031b1fc4)
- Issue #2142 - Emit field keys in correct order (f374ab472)
- Issue #2142 - Don't treat PNK_NAME specially emitAssignmentOrInit (f0b06f5ad)
- Issue #2142 - Set anonymous function name in field initializer (c20611496)
- Issue #2142 - Optimize .initializers scoping and emitter (e0b5528c2)
- Issue #2142 - Support SuperProperty in field initializers (b187006a7)
- Issue #2142 - Parse and process static class fields (c3b19191f)
- Issue #2142 - Track isFieldInitializer on JSScript instead of Scope (22a9d46ef)
- Issue #2142 - Implement class static block (1fac189ae)
- Issue #2142 - Remove the temporary fields option (ba9647a12)
- Issue #2097 - Implement logical assignment operators (10951a169)
- Issue #2097 - Handle Logical Assignment in Ion CFG (a7d8cecdb)
- Issue #1691 - Follow-up: Fix videojs and potentially other problems. When rewriting the ExecutionContext code in ScriptLoader I accidentally removed some required code. (a1f787fc8)
- Issue #2225 - Implement Element.replaceChildren (ae50e9e01)
- Issue #1656 - remove vim control lines from js/src (e6ff567de)
- Issue #1656 - Remove more vim control lines. (d388e478a)
- Issue #1656 - Remove more vim control lines. (c9b8a576d)
- Issue #1656 - Remove more vim control lines. (2f117eeca)
- Issue #1656 - Remove more vim control lines. (a39fd9e75)
- Issue #1656 - Remove more vim control lines. (f1759b33f)
- No Issue - Do not parse or return body for XHRs with HEAD/CONNECT method or content-length=0 (84eadbe3c)
- Issue #1691 - Follow-up: Ship dynamic module imports enabled by default. (5b7f259ab)

Official Pale-Moon changes picked since my last build:
- [Pale-Moon] Issue MoonchildProductions/UXP#1451 - Support hunspell as shared library (2be4e3e69)

Official Basilisk changes picked since my last build:
- [Basilisk] Issue MoonchildProductions/UXP#1451 - Support hunspell as shared library (61874576e)

My changes since my last build:
- partly import changes from tenfourfox: #651: M1761233 M1687303 M1633019 M1797336 M1799748 M1801102 (fb91afbb4) (ba3e46b17)
- follow-up Issue #1691 - Part 6b, fix code that codepath is removed by upstream but remaining here (80d11ebc3)
- Issue #2142 - follow-up rev 51db22ff, removing `static` keyword from upstream force-push rev aa6d42e5fc..63a00fdd25 (fa52ac972)

[roytam1]

New build of BOC/UXP for XP!

Test binary:
MailNews Win32 https://o.rthost.win/boc-uxp/mailnews.win32-20230506-5b6a0056-uxp-725b27a0f-xpmod.7z
BNavigator Win32 https://o.rthost.win/boc-uxp/bnavigator.win32-20230506-5b6a0056-uxp-725b27a0f-xpmod.7z

source repo (excluding UXP): https://github.com/roytam1/boc-uxp/tree/custom

Changes:
- [Mail] Issue MoonchildProductions/UXP#1451 - Support hunspell as shared library (98335aaa)
- [Navigator] Issue MoonchildProductions/UXP#1451 - Support hunspell as shared library (5b6a0056)

* Notice: the profile prefix (i.e. parent folder names) are also changed since 2020-08-15 build, you may rename their names before using new binaries when updating from builds before 2020-08-15.

--

New build of HBL-UXP for XP!

Test binary:
IceDove-UXP(mail) https://o.rthost.win/hbl-uxp/icedove.win32-20230506-id-656ea98-uxp-725b27a0f-xpmod.7z
IceApe-UXP(suite) https://o.rthost.win/hbl-uxp/iceape.win32-20230506-id-656ea98-ia-93af9a0-uxp-725b27a0f-xpmod.7z

source repo (excluding UXP):
https://github.com/roytam1/icedove-uxp/tree/winbuild
https://github.com/roytam1/iceape-uxp/tree/winbuild

Changes:
- [IceDove] Issue MoonchildProductions/UXP#1451 - Support hunspell as shared library (b2d9022)
- [IceApe] Issue MoonchildProductions/UXP#1451 - Support hunspell as shared library (5bce618)

for UXP changes please see above.