My Browser Builds (Part 2)

[RainyShadow]

5 hours ago, we3fan said:

Hi guys, I was just curious, does Good Old Youtube 1.16.0 still work for you on XP?

it stopped working for me recently...

Edited December 25, 2020 by RainyShadow

[roytam1]

New build of Serpent/UXP for XP!

Test binary:
Win32 https://o.rthost.win/basilisk/basilisk52-g4.7.win32-git-20201226-97b3dce-uxp-62568d94d-xpmod.7z
Win64 https://o.rthost.win/basilisk/basilisk52-g4.7.win64-git-20201226-97b3dce-uxp-62568d94d-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-g4.7.win32-git-20201226-97b3dce-uxp-62568d94d-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-28.10.2a1.win32-git-20201226-78b687d98-uxp-62568d94d-xpmod.7z
Win64 https://o.rthost.win/palemoon/palemoon-28.10.2a1.win64-git-20201226-78b687d98-uxp-62568d94d-xpmod.7z

Official UXP changes since my last build:
- Issue #1695 - Add clamping to websocket polling timeouts. (2fa993b56)
- [network] Update port blacklist. (3c0236666)
- Issue #1696 - Propagate flex sizes to the table wrapper (ac4c22f09)
- [toolkit] Handle corner case confusion of downloaded files without extension. (78f114fc4)
- Issue #1695 - Restore Sleep/Wake timer that was erroneously removed. (da0c073a7)
- Revert "Issue #1695 - Add clamping to websocket polling timeouts." (408477a87)
- Issue #1697 - Reinstate the performance timing code removed in error. (0187d3e4e)
- Merge pull request 'Reinstate the performance timing code removed in error' (#1698) from adesh/UXP:fix-performace-api into master (8500ebf53)
- Revert "Issue #1686 - Align a keybinding definition with the others" (05569cb6d)
- Issue #1700 - Apply background color instead of inset shadow for findbar input. (911cbcd1a)
- Issue #1701 - Implement Intl.PluralRules API (25b6703e9)
- Issue #1693 - Update NSPR to 4.29 (149d2ffa7)
- Issue #1693 - Update NSS to 3.59.1.1 (029bcfe18)
- Issue #1693 - Additional configuration for NSS lib update. (ae775e493)
- Clobber for NSS update. (6ae313b51)
- Issue #1681 - Hard-code FLAC and AAC media type associations for .flac and .aac (a8e90975d)
- Revert "Issue #1681 - Hard-code FLAC and AAC media type associations for .flac and .aac" (837432de6)
- Issue #1681 - Hard-code FLAC media type association for .flac (62568d94d)

Official Basilisk changes since my last build:
- Update back-end branch pointer. (2e26040)
- Explicitly disable the legacy DBM format. (97b3dce)

Official Pale-Moon changes since my last build:
- Issue #1854 - Account for users disabling off-line and memory caching. (5b38337f5)
- Back-end branch pointer update (Unstable 2020-12-21) (f4ace7065)
- NSS update, clobber and version bump (0388d9143)
- Explicitly disable the legacy DBM format. (78b687d98)

My changes since my last build:
- Reverted "Reverted Issue UXP#1694 and Basilisk#31 related changes"
- Reverted Issue #1693 related changes since I'll maintain my own NSS and NSPR
- Reverted "Explicitly disable the legacy DBM format" changes since I want to keep using DBM

[roytam1]

New build of BOC/UXP for XP!

Test binary:
MailNews Win32 https://o.rthost.win/boc-uxp/mailnews.win32-20201226-355db4de-uxp-62568d94d-xpmod.7z
BNavigator Win32 https://o.rthost.win/boc-uxp/bnavigator.win32-20201226-355db4de-uxp-62568d94d-xpmod.7z

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

* 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.win32-20201226-id-eed0566-uxp-62568d94d-xpmod.7z
IceApe-UXP(suite) https://o.rthost.win/hbl-uxp/iceape.win32-20201226-id-eed0566-ia-41157bf-uxp-62568d94d-xpmod.7z

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

for UXP changes please see above.

[roytam1]

New NewMoon 27 Build!

32bit https://o.rthost.win/palemoon/palemoon-27.9.7.win32-git-20201226-5e5396a2f-xpmod.7z
32bit SSE https://o.rthost.win/palemoon/palemoon-27.9.7.win32-git-20201226-5e5396a2f-xpmod-sse.7z
32bit noSSE https://o.rthost.win/palemoon/palemoon-27.9.7.win32-git-20201226-5e5396a2f-xpmod-ia32.7z

64bit https://o.rthost.win/palemoon/palemoon-27.9.7.win64-git-20201226-5e5396a2f-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 1140148 - Update reftest manifests for Android 4.3 emulator; r=dminor (69b7ed099)
 - Bug 742176 - Remove/reduce fuzz annotations for tests that no longer appear to be affected by bad overflow areas. r=ehsan (8fac68e26)
 - pieces of older patches (6d4b30aed)
 - Bug 1153574 - Re-enable Mulet reftests taskcluster-specific disables. r=ahal, a=test-only (72ed7dbc2)
 - part of Bug 1138442 - Disable reftests on Mulet in parity with B2G/B2G (8baf07b3c)
 - Bug 1073209 - Eliminate usage of CreateSamplingRestrictedDrawable on OS X. r=jrmuizel (86bee3fbc)
 - Bug 1148889 - Fuzz some android tests. r=mstange (f63be045e)
 - Mark unaccelerated antialiasing reftest failures with APZ as fuzzy. (bug 1157716, r=mstange) (4c0c0f034)
 - Bug 1156197. Just use the return value of the cols DOM property for determining the default number of rendered columns for a textarea. r=ms2ger (16ab3064f)
 - uniform emoji tests here (5a1e9802c)
 - mulet was missing (433a50d93)
 - Bug 1141511 - Disable some gtk2Widget-dependant reftests on Mulet. r=ahal (80d3cef05)
 - Bug 1141607. Fix the initial value of outline-color to actually be correct. r=dbaron (309275f24)
 - part of Bug 1138442 - Disable reftests on Mulet in parity with B2G/B2G (b7a3bb189)
 - Bug 1141511 - Disable some gtk2Widget-dependant reftests on Mulet. r=ahal (1afac0742)
 - Bug 1153845 - For CSS filters, make the primitive subregions big enough to contain the whole pre-filter visual overflow rect. r=roc (6d56dd6f9)
 - annotate (090064379)
 - boh (39f7c5aba)
 - Bug 1170158 - Apply GTK2 reftest exceptions to GTK3. r=karlt (1f1299503)
 - Bug 1165185 - Try to avoid invalidations when scrolling transformed elements. r=roc (d4693eb11)
 - Bug 1165185 - Nudge layer prescale to integers. r=roc (2ac6881fc)
 - remove spurious file (58fc9059b)
 - Bug 1145730 - Restrict the async transforms applied to mask layers to those caused by fixed-position adjustment. r=kats (ebdcc1223)
 - Bug 1146024 - Fix up input routing for APZ on Fennec. r=kats (85e8b249b)
 - Bug 1146024 - Followup to fix non-android bustage on a CLOSED TREE. r=bustage (1ccd4a58f)
 - Bug 1133905. Add some reftests. (89c9f9c8c)
 - Bug 1133905. Make layout place scrollbars at the scroll position clamping scrollport size when one is set so that scrollbars are positioned properly on b2g. r=mstange (841279606)
 - Bug 1155359 - Add a reftest; r=gw280 (a6e7aeb78)
 - Bug 1155359 follow-up: Disable the test on B2G and Mulet (d709b7c54)
 - Bug 1157726: Updated algorithm to match ICU implementation of support for bidi isolates. r=smontagu (96d9801e1)
 - Bug 1154892 - mozinstall fails to install on Mac with paths which contain white spaces. r=chmanchester (1bddadcdd)
 - Bug 1151908 - Updating our character property data with new Bidi info from Unicode 6.3. r=smontagu (c34f88172)
 - Bug 817406 part 2 - Tests for propagating 'direction' to the viewport. (3bc7bfa40)
 - Bug 1139306. Adjust heights of scrollable area so scrollbar sizes almost match in reftests now that RTL scrollbars correctly show in these reftests. r=smontagu (9c77c96ed)
 - Bug 817406 part 3 - Increase the fuzz factors slightly on a few tests for bug 1133905. (22d8a4cbc)
 - Bug 1148582. Part 3 follow up. Properly test if a scroll frame uses async scrolling when determining if it needs its clip put on its layer. r=mstange (13beb6a24)
 - Fix tests that depend on synchronous scrolling. (bug 1140293, r=mstange,masayuki,kgilbert) (578d2ff1e)
 - fix order because of mispatch (75279e185)
 - Bug 1162009 - Add TOUCH_CANCEL sending on e10s. r=kats (0fc5574ac)
 - Bug 1162179 - Don't create APZCTreeManagers for windows which aren't multiprocess. r=dvander (135d115e0)
 - Enable APZ for all Desktop windows if E10s is enabled. (bug 1163115, r=kats) (354729297)
 - Fix wheel events not applying root scroll delta overrides. (bug 1147223 part 1, r=kats) (6d650c449)
 - fix patch order (08b9ad9bd)
 - Improve bad APZ wheel pref detection in about:support. (bug 1147223 part 2, r=kats) (292fa3634)
 - Bug 1151890 - Explicitly set a null MessageLoop for the APZ controller thread on OS X. r=mstange (a4fe37d78)
 - Bug 1148868 - Put the APZ event thread on Mac behind a pref layers.async-pan-zoom.separate-event-thread that defaults to false. r=dvander (827a7cca7)
 - Backed out changeset c5af69669855 (bug 1124084) for various regressions. (28d328c8d)
 - Bug 1141819 - Don't implicitly dispatch a runnable to the main thread in RasterImage::StartDecoding. r=baku (1d1233319)
 - Bug 1124088 - Rename decode-on-draw to decode-only-on-draw. r=tn (83c24cdb7)
 - Bug 1149893 - Add a pref that makes us decode all images immediately. r=baku (6432ae587)
 - Bug 1156401 - Fix ContextMenu handling when building Fennec with C++ APZ. r=kats, r=botond (dd784945c)
 - Bug 1168629. Always make sure that there is at least one layer with the metrics for the root scroll frame/root element. r=botond,kats (06f19aef3)
 - Bug 1154478 - Rename gfxPrefs::LayoutEventRegionsEnabled to LayoutEventRegionsEnabledDoNotUseDirectly. r=kats (68790b756)
 - Don't expand displayports beyond the maximum texture size. (bug 1135907, r=kats) (805866f66)
 - Bug 1138444 - Add Mulet detection for reftests. r=jgriffin (cc9378556)
 - Use widgets for APZ checks, rather than gfxPrefs. (bug 1162064, r=kats,mstange) (40d2823a5)
 - Bug 1151993 - Apply shrink-wrapping to block frames that are orthogonal to their containing block. r=smontagu (9fdd76008)
 - Bug 1156021 - Use containing block inline-size (as appropriate for writing mode) to compute text-indent when specified as percentage. r=dbaron (c7fcb4193)
 - Bug 1152941 - Always use unconstrained block-size when reflowing an orthogonal child block, to avoid incomplete reflow across orthogonal boundary. r=dbaron (5f4748c5a)
 - Bug 1163238 - Don't set available block-size to unconstrained for orthogonal-block reflow state until after InitConstraints etc. r=dholbert (07ca784bf)
 - Bug 1155798: Drop end-of-line whitespace from C++ files in layout/tables. (whitespace-only, no review, DONTBUILD) (6ad9c50d8)
 - Bug 1155880 part 1: Rename nsITableLayoutStrategy::ComputeColumnWidths and associated comments to use "ISize" instead of "Width". r=jfkthame (c3bec937b)
 - Bug 1155880 part 2: Rename nsTableColFrame Get/Set/ResetFinalWidth methods to use ISize instead of Width. r=jfkthame (1d2bfd990)
 - Bug 1155880 part 3: Rename nsTableFrame::GetColumnWidth to use "ISize" instead of "Width". r=jfkthame (8a0d7a3c8)
 - Bug 1159101 part 1 - Alter dependencies among headers of table struct frames. r=roc (b943ee1c3)
 - Bug 1159101 part 2 - Add GetTableFrame() for table inner frames. r=roc (6641affa7)
 - Bug 1145036 part 1 - Rename mBorderSpacingX/Y to mBorderSpacingCol/Row. r=heycam (7e91504af)
 - Bug 1145036 part 2 - Rename nsTableFrame::GetCellSpacingX/Y to GetCol/RowSpacing. r=roc (b439569de)
 - Bug 1159101 part 3 - Replace most of nsTableFrame::GetTableFrame() usage with new GetTableFrame() methods. r=roc (4bdf27d00)
 - Bug 1159990 - Add TableArea struct and replace nsIntRect with it for table damage area. r=roc (ebca44c12) (fc21d5d09)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - bug 1157569 part 1 - Rename BC_BORDER_{TOP,RIGHT,BOTTOM,LEFT}_HALF* to BC_BORDER_{START,END}_HALF*. r=roc (2ef82ce76)
 - Bug 1157569 part 2 - Convert output parameters of GetColorAndStyle/GetPaintStyleInfo in nsTableFrame to use pointers. r=roc (c58bbe734)
 - Bug 1157569 part 3 - Merge two GetColorAndStyle functions in nsTableFrame. r=roc (009df4d97)
 - Bug 1157569 part 4 - Replace mTableIsLTR with mTableWM in BCMapCellInfo & BCPaintBorderIterator. r=roc (9784a46eb)
 - Bug 1157569 part 5 - Convert GetColorAndStyle, GetPaintStyleInfo, and CompareBorders in nsTableFrame to accept writing mode and logical side. r=roc (aaddc4a80)
 - Bug 1157569 part 6 - Rename methods and fields in BCMapCellInfo from physicals to logicals. r=roc (aeaa0b18c)
 - Bug 1157569 part 7 - Rename methods and fields in nsTableColFrame from physicals to logicals. r=roc (fd4e1e667)
 - Bug 1157569 part 8 - Move some code in BCPaintBorderIterator::SetDamageArea for less computation. r=roc (634f19ca7)
 - missing bit of Bug 895096 - Part 1 (a96e3d35c)
 - Bug 1157569 part 9 - Rename methods and fields in nsTableRowFrame from physicals to logicals. r=roc (d097f1e4f)
 - Bug 1157569 part 10 - Add operator+= for LogicalMargin. r=jfkthame (c651b685f)
 - Bug 1159101 part 4 - Override GetWritingMode() in nsTable{{Col,Row}{,Group},Cell}Frame. r=roc (34d90d10b)
 - Bug 1159101 part 5 - Initialize writing mode of desired mode of table cell frame from its table frame on a CLOSED TREE. r=roc (8bc6978ca)
 - Bug 1157569 part 11 - Rename methods and fields in nsTableCellFrame from physicals to logicals. r=roc (bcfd4514b)
 - Bug 1157569 part 12 - Remove useless m{Start,End}Side from BCMapCellInfo. r=roc (1df59d670)
 - Bug 1159127 - Always redirect inserting col group frame to first-in-flow table frame. r=mats (2e41bbeeb)
 - Bug 1167765: Misc whitespace cleanup in nsTableFrame.cpp. whitespace/rewrapping-only, no review, DONTBUILD (f256939c9)
 - Bug 1167696 part 1: Remove nsTableReflowState's unused constructor & nsPresContext arg. r=mats (5b8beadb9)
 - Bug 1167696 part 2: Remove nsTableReflowState's frame constructor-arg; get it from passed-in reflow state instead. r=mats (205dad74f)
 - Bug 1155412: Use mozilla::Maybe instead of hardcoded placement-new, for reflow state created in nsTableOuterFrame::OuterBeginReflowChild() and used by its caller. r=dbaron (ebf9e9891)
 - Bug 1147834: Use abstract coordinates in nsHTMLReflowState, r=jfkthame (30e644a14)
 - Bug 1169432 part 2: Refactor nsTableFrame::Init. r=mats (779928597)
 - Bug 1169432 part 3: Use nsTArray::Contains instead of nsTArray::IndexOf(), for brevity, in nsTableFrame. r=mats (aa6aee3c4)
 - Bug 1169432 part 4: Convert NS_ASSERTION(false, ...) to NS_ERROR(...) in nsTableFrame. r=mats (9cbb2f53f)
 - Bug 1169432 part 5: Use range-based 'for' loops when walking frame lists, in nsTableFrame. r=mats (bd4e7d0d4)
 - Bug 1171328 - Convert nsTableFrame::GetChildAreaOffset() and its friends to use LogicalMargin. r=dholbert (196955d0c)
 - Remove unneeded code from SmoothScrollAnimation. (bug 1139220 part 4, r=kats) (68c6af9f8)
 - Use stronger typing for GetScrollWheelDelta. (bug 1139220 part 5, r=kats) (4669d1f38)
 - When APZ is enabled, only apply the root scroll delta override to root frames. (bug 1155800, r=kats) (2a738ee82)
 - Bug 1168629. Clear frame metrics on the root layer if we aren't setting new frame metrics because the layer may be recycled and still have frame metrics from last time. r=mstange (dc0889fae)
 - Bug 1158424 - Rename FrameMetrics::mIsRoot to mIsRootContent. r=kats (0005fb016)
 - Bug 1148078: Replace the word "Logical" with "Flex-Relative" in some vars & functions in flexbox layout. r=mats (601497c55)
 - Bug 1148298 part 1: Convert most flexbox-layout IsAxisHorizontal() calls into IsMainAxisHorizontal()/IsCrossAxisHorizontal(). r=mats (ff2aaa5b3)
 - Bug 1148298 part 3: Make flexbox helper GetBaselineOffsetFromOuterCrossEdge take a FlexboxAxisTracker instead of an explicit axis. r=mats (009dd9f7e)
 - Bug 1148298 part 4: Convert flexbox helper-function IsAxisHorizontal() to be a private implementation detail of FlexboxAxisTracker. r=mats (901f830d6)
 - Bug 1148298 part 2: Replace flexbox static helper-function GetSizeProBug 1148298 part 2: Replace flexbox static helper-function GetSizePro (3d2940f88)
 - Bug 1158290 part 1: Use AvailableBSize instead of AvailableHeight, throughout flex layout. r=mats (49e2075f8)
 - Bug 1148294 part 1: Determine flex axes (in FlexboxAxisTracker) using the flex container's writing mode. r=mats (67f12cf58)
 - Bug 1148294 part 2: Add one reftest for writing-mode & direction properties' influence on flex axes. (85e0ede9d)
 - Bug 1148294 part 3: Add more reftests for writing-mode & direction properties' influence on flex axes (as modified copies of first test). (fac965a24)
 - Bug 1148294 followup: fix spec links in reftests flexbox-writing-mode-* to point to TR instead of ED spec version. (no review) (84333c728)
 - Bug 1158290 part 2: Use logical (not physical) skipsides & borderpadding when chipping away from available BSize. r=mats (987b4edfa)
 - Bug 1152913 part 1: Make FlexboxAxisTracker store the WritingMode & representation of how flex axes map to it. r=mats (675737981)
 - Bug 1152951: Delete the copy-constructor and reassignment operator for FlexboxAxisTracker & PositionTracker state-management classes. r=mats (71c88e7e9)
 - Bug 1152913 part 2: Convert flexbox 'IsAxisHorizontal' checks to use new logical-axis member data. r=mats (122b434a5)
 - Bug 1158290 part 3: Convert GenerateFlexLines to use logical axes & sizes. r=mats (367cb1c01)
 - Bug 1158290 part 4: Clarify documentation & naming of ClampFlexContainerMainSize (now called ResolveFlexContainerMainSize). r=mats (5af11ff23)
 - Bug 1149383 part 1: Make flexbox writing-mode tests use different widths vs. heights to reveal bug with vertical writing-modes. (test-only) (c4782feaa)
 - Bug 1149383 part 2: Reorder width/height decls in flexbox-writing-mode reftests for consistency. (test-only) (bba907b70)
 - Bug 1158290 part 5: Use logical coords & axes in methods that computeBug 1158290 part 5: Use logical coords & axes in methods that compute (b78726528)
 - Bug 1158290 part 6: Use new GET_[MAIN|CROSS]_COMPONENT_LOGICAL macros, to reduce explicit width/height usage. r=mats (25d2e20a3)
 - Bug 1152913 part 3: Make flexbox's PositionTracker classes track whether axis is reversed, & directly convert flex item final sizes/positions to logical coords. r=mats (1feee7e6d)
 - Bug 1174507 - Replace the frame flag CONTAINS_RELATIVE_HEIGHT with CONTAINS_RELATIVE_BSIZE, and adjust callsites appropriately. r=dholbert (0e981d5ac)
 - Clip async scrollframes based on the displaylist clip, not the composition bounds. (bug 1148582 part 4, r=tn) (f2b1196ed)
 - Bug 1148582 - Add mask layers to FrameMetrics for ancestor scroll frame clips. r=mattwoodrow (008239305)
 - Bug 1148582 - Factor mask layer creation out of ContainerState::SetupMaskLayer. r=mstange (1d739614e)
 - piece of Bug 1138442 (e98de18dd)
 - Bug 1173307 - Convert nsTableCellFrame to work with logical coordinates. r=dholbert (81d0aa811)
 - Bug 1174700 - patch 1 - Convert nsTableRowFrame and nsTableRowGroupFrame to work with logical coordinates. r=dholbert (09ee9c814)
 - Bug 1174700 - patch 2 - Convert nsTableFrame::SetColumnDimensions to work with logical-order iteration over the table, so that column coordinates and collapsed borders are correct. r=dholbert (7957e377b)
 - Bug 1129040 - Provide a way for content processes to query the chrome side blocklist service. r=billm (5ea750b8b)
 - Bug 1174711 - patch 1 - Rename nsIPercentHeightObserver to nsIPercentBSizeObserver, and update related frame methods to match. r=dholbert (7a17fbe4f)
 - Bug 1174711 - patch 2 - Convert nsTableFrame to work with logical coordinates. r=dholbert (329776abe)
 - Bug 1174711 - patch 3 - Rename a couple more frame-state bits from physical to logical. r=dholbert (d61c9624d)
 - Bug 1174711 - patch 4 - Rename mSpecialHeightReflow to mSpecialBSizeReflow, and update comments to match. r=dholbert (49969e512)
 - Bug 1174711 patch 5 - Copy inline-size rather than width from prev-in-flow when initializing nsTableFrame. r=dholbert (7afb6c80a)
 - pointer style (52cd6f9af)
 - pointer style (ba99b3f74)
 - Bug 1149797 - Use the loop's *static* block object when freshening a loop's block object, then copy in values from the old cloned block. Using the old cloned block directly isn't valid when the cloned block might be extended with additional variables created by eval or added by nested function statements. r=shu (deba22790)
 - Bug 1165486 - Cleanup: use standard object allocation functions when allocating scope objects. (r=terrence) (b35a4d7aa)
 - pointer style (74ea6d085)
 - Bug 1165486 - Add StaticNonSyntacticScopeObjects and teach scope iterators about it. (r=luke) (18a526ed3)
 - Bug 1144371: Implement DEBUG-only JS shell function |dumpStringRepresentation|. r=jandem (35524cf42)
 - Bug 1148963 - OdinMonkey: add CompileOptions::lazyParsingDisabled and testing function setDiscardSource (r=bz) (53f9c8fe6)
 - pointer style (edaa05d69)
 - Bug 1165486 - Rename hasPollutedGlobalScope to hasNonSyntacticScope. (r=luke) (505fc8d05)
 - pointer style (3c4435038)
 - Bug 1165486 - Remove PollutedGlobalScopeOption in favor of using the static scope chain to detect non-syntactic scopes. (r=luke) (73ca126f5)
 - Bug 1150106 - Async loadSubscript - Part 1, add `async` to loadSubscriptWithOptions r=bholley,mcrr8 (a0ffeed0d)
 - Bug 1150106 - Add an async version of loadSubscript - Part 2, tests r=bholley (6aeaa93e4)
 - Bug 1165486 - Use JS::CompileForNonSyntacticScope in Gecko where we used to set polluted global scope. (r=bz, rs=bholley) (073332ac9)
 - Bug 1165486 - Split JS::Compile into JS::Compile and JS::CompileForNonSyntacticScope. (r=luke) (0b82c73c4)
 - Bug 1165486 - Restructure function and script cloning in light of PollutingGlobal scope changes. (r=Waldo) (b4e39652c)
 - Bug 1141865 - Part 7: Make new.target work in generator functions. (r=jorendorff, r=jandem) (a9f5ce1ad)
 - Bug 1165486 - Detect with scopes at parse time using the static scope chain for non-function scripts. Also cache static scope properties on SharedGlobalContext. (r=efaust) (a7d4283cb)
 - Bug 1176070 - Optimize nsTableFrame's FirstInFlow() lookups from calls to GetColumnISize(). r=dholbert (beed47b93)
 - fix misspatch of Bug 1141862 - Part 6 (d091f70a2)
 - Bug 1169736 - Temporarily disallow eval and arrow functions inside derived class constructors. (r=jorendorff) (6cf8a2db1)
 - Bug 1150855 - Disallow method syntax without curly brackets. r=efaust (26086ee95)
 - part of Bug 1150855 - Remove uses of the curly syntax. (bfe86ea19)
 - Bug 1169853 - Require semicolon after export default AssignmentExpression. r=shu (580ff7e30)
 - Bug 1174009 - Remove a couple of unused AutoRooter types; r=sfink (113aee16a) (b3541682d)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1176414 - Clean up/rename various column index variables in nsTableRowFrame and nsTableFrame. r=dholbert (d8e33d549)
 - Bug 1176105 - Remove the (largely gutted) nsTableIterator class, and replace with simple frame-list iteration. r=dholbert (46a15df19)
 - Bug 1176555 - Replace explicit bit-twiddling of frame state flags by human-readable nsIFrame state-manipulation methods in table layout code. r=dholbert (aef9fb035)
 - Bug 1176354 - Rename nsTableRowFrame::GetBSize to differentiate better from nsIFrame::BSize. r=dholbert (9b67203a0)
 - Bug 1177600 - Properly adjust table row and cell positions when their containing block-size changes in vertical-rl writing mode. r=dholbert (413148d55)
 - Bug 1177925: Use range-based for loops for nsFrameList iteration, in various places in layout. r=tn (6db450285)
 - Bug 1148582 - Recycle mask layers attached to FrameMetrics. r=mattwoodrow (91360fbd8)
 - Bug 1148582 - Apply async transforms to (ancestor) mask layers correctly. r=botond (d6fca66fb)
 - Bug 1143575. Add some #includes to avoid unified-build issues on Windows. r=nical (ff78b76a3)
 - Bug 1143575. Add some #includes to avoid more unified-build issues on Windows. r=nical (99e0548f4)
 - Bug 1181832 - Keep gfxPrefs.h out of header files. r=kats (341f0aea0)
 - Bug 1181832 - Include gfxPrefs.h from nsImageLoadingContet.cpp to fix bustage on a CLOSED TREE. r=bustage (d61a18ab9)
 - Bug 1137557 - Part 0: TextEventDispatcher shouldn't forward keyboard events coming from TextInputProcessor to the parent process. r=smaug (87f70fcc4)
 - Bug 1172405 - Fix build error when adding files to dom/workers and gfx/layers. r=nical (8e114b8bb)
 - Bug 1061393 - Export display list info to layer scope viewer. r=kamidphish (60fd247f0)
 - Bug 1072313 - Check for more leaks in the layers code. r=mstange (aacbb2e1c)
 - Bug 1198979 - GC unused IPDL struct member. r=jmuizelaar (79c3113fb)
 - add missing semicolon (17ba58b45)
 - Bug 1153348 - Add an analysis to prohibit operator bools which aren't marked as either explicit or MOZ_IMPLICIT; r=jrmuizel (903dca970)
 - Bug 1170388 - Restrict the static analysis error given about raw pointers to refcounted objects inside a lambda, to the case where the raw pointer is captured. r=ehsan (cbbd3174c)
 - Bug 1180993 - Part 1: Add an analysis to help catch unused return values of specific types. r=ehsan (2420face4)
 - Bug 1180993 - Part 2: Add the MOZ_MUST_USE attribute to mfbt, and use it to verify usage of already_AddRefed. r=ehsan (b72785a60)
 - Bug 1123907 - Part 1: Add an analysis to ensure that a class marked MOZ_NEEDS_NO_VTABLE_TYPE cannot be instantiated by a class with a VTable; r=ehsan (0f5730c04)
 - Bug 1123907 - Part 2: Add MOZ_NEEDS_NO_VTABLE_TYPE to Attributes.h, and use it to verify the EntryType argument of nsTHashtable; r=ehsan (2e80fc581)
 - Bug 1159433 - Part 1: Add an analysis to ensure that some template arguments cannot be non-memmovable types; r=ehsan (3602e950f)
 - uniform MOZ_MUST_USE, it will be renamed in patch 1267550 (3b5ebd4c5)
 - add missing part of Bug 1187073 - Use MOZ_INHERIT_TYPE_ANNOTATIONS_FROM_TEMPLATE_ARGS to validate the usage of AlignedStorage2. r=ehsan (3b451bfb7)
 - Bug 1159433 - Part 2: Add the MOZ_NON_MEMMOVABLE and MOZ_NEEDS_MEMMOVABLE_TYPE macros to MFBT; r=froydnj (171c23ec4)
 - Bug 1159433 - Part 3: Make nsTArray_CopyChooser only accept memmovable argument types by default; r=froydnj (40cb4aa67)
 - Bug 1159433 - Part 4: Mark nsTAutoString_CharT and nsAutoArrayBase as non-memmovable; r=froydnj (7dc6bd220) (75dd89485)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1158456 - Remove control characters from composition string, and add dom.compositionevent.allow_control_characters pref to control it. r=masayuki (5d94de2fd)
 - Bug 1154053 - Limit concurrency of e10s memory reporting. r=erahm (87b9d2335) (5e5396a2f)

[roytam1]

New regular/weekly KM-Goanna release:
https://o.rthost.win/kmeleon/KM76.3.1-Goanna-20201226.7z

Changelog:

In-tree changes:
- kplugins: jsplugin, spellcheck: follow-up after chaos mode imported in GRE (9b574223)
- sln: build WinEmbed after kmeleon to fix racing condition on program database (3295b6dc)

Out-of-tree changes:
* update Goanna3 to git cafd4d45b..5e5396a2f:
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1140148 - Update reftest manifests for Android 4.3 emulator; r=dminor (69b7ed099)
 - Bug 742176 - Remove/reduce fuzz annotations for tests that no longer appear to be affected by bad overflow areas. r=ehsan (8fac68e26)
 - pieces of older patches (6d4b30aed)
 - Bug 1153574 - Re-enable Mulet reftests taskcluster-specific disables. r=ahal, a=test-only (72ed7dbc2)
 - part of Bug 1138442 - Disable reftests on Mulet in parity with B2G/B2G (8baf07b3c)
 - Bug 1073209 - Eliminate usage of CreateSamplingRestrictedDrawable on OS X. r=jrmuizel (86bee3fbc)
 - Bug 1148889 - Fuzz some android tests. r=mstange (f63be045e)
 - Mark unaccelerated antialiasing reftest failures with APZ as fuzzy. (bug 1157716, r=mstange) (4c0c0f034)
 - Bug 1156197. Just use the return value of the cols DOM property for determining the default number of rendered columns for a textarea. r=ms2ger (16ab3064f)
 - uniform emoji tests here (5a1e9802c)
 - mulet was missing (433a50d93)
 - Bug 1141511 - Disable some gtk2Widget-dependant reftests on Mulet. r=ahal (80d3cef05)
 - Bug 1141607. Fix the initial value of outline-color to actually be correct. r=dbaron (309275f24)
 - part of Bug 1138442 - Disable reftests on Mulet in parity with B2G/B2G (b7a3bb189)
 - Bug 1141511 - Disable some gtk2Widget-dependant reftests on Mulet. r=ahal (1afac0742)
 - Bug 1153845 - For CSS filters, make the primitive subregions big enough to contain the whole pre-filter visual overflow rect. r=roc (6d56dd6f9)
 - annotate (090064379)
 - boh (39f7c5aba)
 - Bug 1170158 - Apply GTK2 reftest exceptions to GTK3. r=karlt (1f1299503)
 - Bug 1165185 - Try to avoid invalidations when scrolling transformed elements. r=roc (d4693eb11)
 - Bug 1165185 - Nudge layer prescale to integers. r=roc (2ac6881fc)
 - remove spurious file (58fc9059b)
 - Bug 1145730 - Restrict the async transforms applied to mask layers to those caused by fixed-position adjustment. r=kats (ebdcc1223)
 - Bug 1146024 - Fix up input routing for APZ on Fennec. r=kats (85e8b249b)
 - Bug 1146024 - Followup to fix non-android bustage on a CLOSED TREE. r=bustage (1ccd4a58f)
 - Bug 1133905. Add some reftests. (89c9f9c8c)
 - Bug 1133905. Make layout place scrollbars at the scroll position clamping scrollport size when one is set so that scrollbars are positioned properly on b2g. r=mstange (841279606)
 - Bug 1155359 - Add a reftest; r=gw280 (a6e7aeb78)
 - Bug 1155359 follow-up: Disable the test on B2G and Mulet (d709b7c54)
 - Bug 1157726: Updated algorithm to match ICU implementation of support for bidi isolates. r=smontagu (96d9801e1)
 - Bug 1154892 - mozinstall fails to install on Mac with paths which contain white spaces. r=chmanchester (1bddadcdd)
 - Bug 1151908 - Updating our character property data with new Bidi info from Unicode 6.3. r=smontagu (c34f88172)
 - Bug 817406 part 2 - Tests for propagating 'direction' to the viewport. (3bc7bfa40)
 - Bug 1139306. Adjust heights of scrollable area so scrollbar sizes almost match in reftests now that RTL scrollbars correctly show in these reftests. r=smontagu (9c77c96ed)
 - Bug 817406 part 3 - Increase the fuzz factors slightly on a few tests for bug 1133905. (22d8a4cbc)
 - Bug 1148582. Part 3 follow up. Properly test if a scroll frame uses async scrolling when determining if it needs its clip put on its layer. r=mstange (13beb6a24)
 - Fix tests that depend on synchronous scrolling. (bug 1140293, r=mstange,masayuki,kgilbert) (578d2ff1e)
 - fix order because of mispatch (75279e185)
 - Bug 1162009 - Add TOUCH_CANCEL sending on e10s. r=kats (0fc5574ac)
 - Bug 1162179 - Don't create APZCTreeManagers for windows which aren't multiprocess. r=dvander (135d115e0)
 - Enable APZ for all Desktop windows if E10s is enabled. (bug 1163115, r=kats) (354729297)
 - Fix wheel events not applying root scroll delta overrides. (bug 1147223 part 1, r=kats) (6d650c449)
 - fix patch order (08b9ad9bd)
 - Improve bad APZ wheel pref detection in about:support. (bug 1147223 part 2, r=kats) (292fa3634)
 - Bug 1151890 - Explicitly set a null MessageLoop for the APZ controller thread on OS X. r=mstange (a4fe37d78)
 - Bug 1148868 - Put the APZ event thread on Mac behind a pref layers.async-pan-zoom.separate-event-thread that defaults to false. r=dvander (827a7cca7)
 - Backed out changeset c5af69669855 (bug 1124084) for various regressions. (28d328c8d)
 - Bug 1141819 - Don't implicitly dispatch a runnable to the main thread in RasterImage::StartDecoding. r=baku (1d1233319)
 - Bug 1124088 - Rename decode-on-draw to decode-only-on-draw. r=tn (83c24cdb7)
 - Bug 1149893 - Add a pref that makes us decode all images immediately. r=baku (6432ae587)
 - Bug 1156401 - Fix ContextMenu handling when building Fennec with C++ APZ. r=kats, r=botond (dd784945c)
 - Bug 1168629. Always make sure that there is at least one layer with the metrics for the root scroll frame/root element. r=botond,kats (06f19aef3)
 - Bug 1154478 - Rename gfxPrefs::LayoutEventRegionsEnabled to LayoutEventRegionsEnabledDoNotUseDirectly. r=kats (68790b756)
 - Don't expand displayports beyond the maximum texture size. (bug 1135907, r=kats) (805866f66)
 - Bug 1138444 - Add Mulet detection for reftests. r=jgriffin (cc9378556)
 - Use widgets for APZ checks, rather than gfxPrefs. (bug 1162064, r=kats,mstange) (40d2823a5)
 - Bug 1151993 - Apply shrink-wrapping to block frames that are orthogonal to their containing block. r=smontagu (9fdd76008)
 - Bug 1156021 - Use containing block inline-size (as appropriate for writing mode) to compute text-indent when specified as percentage. r=dbaron (c7fcb4193)
 - Bug 1152941 - Always use unconstrained block-size when reflowing an orthogonal child block, to avoid incomplete reflow across orthogonal boundary. r=dbaron (5f4748c5a)
 - Bug 1163238 - Don't set available block-size to unconstrained for orthogonal-block reflow state until after InitConstraints etc. r=dholbert (07ca784bf)
 - Bug 1155798: Drop end-of-line whitespace from C++ files in layout/tables. (whitespace-only, no review, DONTBUILD) (6ad9c50d8)
 - Bug 1155880 part 1: Rename nsITableLayoutStrategy::ComputeColumnWidths and associated comments to use "ISize" instead of "Width". r=jfkthame (c3bec937b)
 - Bug 1155880 part 2: Rename nsTableColFrame Get/Set/ResetFinalWidth methods to use ISize instead of Width. r=jfkthame (1d2bfd990)
 - Bug 1155880 part 3: Rename nsTableFrame::GetColumnWidth to use "ISize" instead of "Width". r=jfkthame (8a0d7a3c8)
 - Bug 1159101 part 1 - Alter dependencies among headers of table struct frames. r=roc (b943ee1c3)
 - Bug 1159101 part 2 - Add GetTableFrame() for table inner frames. r=roc (6641affa7)
 - Bug 1145036 part 1 - Rename mBorderSpacingX/Y to mBorderSpacingCol/Row. r=heycam (7e91504af)
 - Bug 1145036 part 2 - Rename nsTableFrame::GetCellSpacingX/Y to GetCol/RowSpacing. r=roc (b439569de)
 - Bug 1159101 part 3 - Replace most of nsTableFrame::GetTableFrame() usage with new GetTableFrame() methods. r=roc (4bdf27d00)
 - Bug 1159990 - Add TableArea struct and replace nsIntRect with it for table damage area. r=roc (ebca44c12) (fc21d5d09)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - bug 1157569 part 1 - Rename BC_BORDER_{TOP,RIGHT,BOTTOM,LEFT}_HALF* to BC_BORDER_{START,END}_HALF*. r=roc (2ef82ce76)
 - Bug 1157569 part 2 - Convert output parameters of GetColorAndStyle/GetPaintStyleInfo in nsTableFrame to use pointers. r=roc (c58bbe734)
 - Bug 1157569 part 3 - Merge two GetColorAndStyle functions in nsTableFrame. r=roc (009df4d97)
 - Bug 1157569 part 4 - Replace mTableIsLTR with mTableWM in BCMapCellInfo & BCPaintBorderIterator. r=roc (9784a46eb)
 - Bug 1157569 part 5 - Convert GetColorAndStyle, GetPaintStyleInfo, and CompareBorders in nsTableFrame to accept writing mode and logical side. r=roc (aaddc4a80)
 - Bug 1157569 part 6 - Rename methods and fields in BCMapCellInfo from physicals to logicals. r=roc (aeaa0b18c)
 - Bug 1157569 part 7 - Rename methods and fields in nsTableColFrame from physicals to logicals. r=roc (fd4e1e667)
 - Bug 1157569 part 8 - Move some code in BCPaintBorderIterator::SetDamageArea for less computation. r=roc (634f19ca7)
 - missing bit of Bug 895096 - Part 1 (a96e3d35c)
 - Bug 1157569 part 9 - Rename methods and fields in nsTableRowFrame from physicals to logicals. r=roc (d097f1e4f)
 - Bug 1157569 part 10 - Add operator+= for LogicalMargin. r=jfkthame (c651b685f)
 - Bug 1159101 part 4 - Override GetWritingMode() in nsTable{{Col,Row}{,Group},Cell}Frame. r=roc (34d90d10b)
 - Bug 1159101 part 5 - Initialize writing mode of desired mode of table cell frame from its table frame on a CLOSED TREE. r=roc (8bc6978ca)
 - Bug 1157569 part 11 - Rename methods and fields in nsTableCellFrame from physicals to logicals. r=roc (bcfd4514b)
 - Bug 1157569 part 12 - Remove useless m{Start,End}Side from BCMapCellInfo. r=roc (1df59d670)
 - Bug 1159127 - Always redirect inserting col group frame to first-in-flow table frame. r=mats (2e41bbeeb)
 - Bug 1167765: Misc whitespace cleanup in nsTableFrame.cpp. whitespace/rewrapping-only, no review, DONTBUILD (f256939c9)
 - Bug 1167696 part 1: Remove nsTableReflowState's unused constructor & nsPresContext arg. r=mats (5b8beadb9)
 - Bug 1167696 part 2: Remove nsTableReflowState's frame constructor-arg; get it from passed-in reflow state instead. r=mats (205dad74f)
 - Bug 1155412: Use mozilla::Maybe instead of hardcoded placement-new, for reflow state created in nsTableOuterFrame::OuterBeginReflowChild() and used by its caller. r=dbaron (ebf9e9891)
 - Bug 1147834: Use abstract coordinates in nsHTMLReflowState, r=jfkthame (30e644a14)
 - Bug 1169432 part 2: Refactor nsTableFrame::Init. r=mats (779928597)
 - Bug 1169432 part 3: Use nsTArray::Contains instead of nsTArray::IndexOf(), for brevity, in nsTableFrame. r=mats (aa6aee3c4)
 - Bug 1169432 part 4: Convert NS_ASSERTION(false, ...) to NS_ERROR(...) in nsTableFrame. r=mats (9cbb2f53f)
 - Bug 1169432 part 5: Use range-based 'for' loops when walking frame lists, in nsTableFrame. r=mats (bd4e7d0d4)
 - Bug 1171328 - Convert nsTableFrame::GetChildAreaOffset() and its friends to use LogicalMargin. r=dholbert (196955d0c)
 - Remove unneeded code from SmoothScrollAnimation. (bug 1139220 part 4, r=kats) (68c6af9f8)
 - Use stronger typing for GetScrollWheelDelta. (bug 1139220 part 5, r=kats) (4669d1f38)
 - When APZ is enabled, only apply the root scroll delta override to root frames. (bug 1155800, r=kats) (2a738ee82)
 - Bug 1168629. Clear frame metrics on the root layer if we aren't setting new frame metrics because the layer may be recycled and still have frame metrics from last time. r=mstange (dc0889fae)
 - Bug 1158424 - Rename FrameMetrics::mIsRoot to mIsRootContent. r=kats (0005fb016)
 - Bug 1148078: Replace the word "Logical" with "Flex-Relative" in some vars & functions in flexbox layout. r=mats (601497c55)
 - Bug 1148298 part 1: Convert most flexbox-layout IsAxisHorizontal() calls into IsMainAxisHorizontal()/IsCrossAxisHorizontal(). r=mats (ff2aaa5b3)
 - Bug 1148298 part 3: Make flexbox helper GetBaselineOffsetFromOuterCrossEdge take a FlexboxAxisTracker instead of an explicit axis. r=mats (009dd9f7e)
 - Bug 1148298 part 4: Convert flexbox helper-function IsAxisHorizontal() to be a private implementation detail of FlexboxAxisTracker. r=mats (901f830d6)
 - Bug 1148298 part 2: Replace flexbox static helper-function GetSizeProBug 1148298 part 2: Replace flexbox static helper-function GetSizePro (3d2940f88)
 - Bug 1158290 part 1: Use AvailableBSize instead of AvailableHeight, throughout flex layout. r=mats (49e2075f8)
 - Bug 1148294 part 1: Determine flex axes (in FlexboxAxisTracker) using the flex container's writing mode. r=mats (67f12cf58)
 - Bug 1148294 part 2: Add one reftest for writing-mode & direction properties' influence on flex axes. (85e0ede9d)
 - Bug 1148294 part 3: Add more reftests for writing-mode & direction properties' influence on flex axes (as modified copies of first test). (fac965a24)
 - Bug 1148294 followup: fix spec links in reftests flexbox-writing-mode-* to point to TR instead of ED spec version. (no review) (84333c728)
 - Bug 1158290 part 2: Use logical (not physical) skipsides & borderpadding when chipping away from available BSize. r=mats (987b4edfa)
 - Bug 1152913 part 1: Make FlexboxAxisTracker store the WritingMode & representation of how flex axes map to it. r=mats (675737981)
 - Bug 1152951: Delete the copy-constructor and reassignment operator for FlexboxAxisTracker & PositionTracker state-management classes. r=mats (71c88e7e9)
 - Bug 1152913 part 2: Convert flexbox 'IsAxisHorizontal' checks to use new logical-axis member data. r=mats (122b434a5)
 - Bug 1158290 part 3: Convert GenerateFlexLines to use logical axes & sizes. r=mats (367cb1c01)
 - Bug 1158290 part 4: Clarify documentation & naming of ClampFlexContainerMainSize (now called ResolveFlexContainerMainSize). r=mats (5af11ff23)
 - Bug 1149383 part 1: Make flexbox writing-mode tests use different widths vs. heights to reveal bug with vertical writing-modes. (test-only) (c4782feaa)
 - Bug 1149383 part 2: Reorder width/height decls in flexbox-writing-mode reftests for consistency. (test-only) (bba907b70)
 - Bug 1158290 part 5: Use logical coords & axes in methods that computeBug 1158290 part 5: Use logical coords & axes in methods that compute (b78726528)
 - Bug 1158290 part 6: Use new GET_[MAIN|CROSS]_COMPONENT_LOGICAL macros, to reduce explicit width/height usage. r=mats (25d2e20a3)
 - Bug 1152913 part 3: Make flexbox's PositionTracker classes track whether axis is reversed, & directly convert flex item final sizes/positions to logical coords. r=mats (1feee7e6d)
 - Bug 1174507 - Replace the frame flag CONTAINS_RELATIVE_HEIGHT with CONTAINS_RELATIVE_BSIZE, and adjust callsites appropriately. r=dholbert (0e981d5ac)
 - Clip async scrollframes based on the displaylist clip, not the composition bounds. (bug 1148582 part 4, r=tn) (f2b1196ed)
 - Bug 1148582 - Add mask layers to FrameMetrics for ancestor scroll frame clips. r=mattwoodrow (008239305)
 - Bug 1148582 - Factor mask layer creation out of ContainerState::SetupMaskLayer. r=mstange (1d739614e)
 - piece of Bug 1138442 (e98de18dd)
 - Bug 1173307 - Convert nsTableCellFrame to work with logical coordinates. r=dholbert (81d0aa811)
 - Bug 1174700 - patch 1 - Convert nsTableRowFrame and nsTableRowGroupFrame to work with logical coordinates. r=dholbert (09ee9c814)
 - Bug 1174700 - patch 2 - Convert nsTableFrame::SetColumnDimensions to work with logical-order iteration over the table, so that column coordinates and collapsed borders are correct. r=dholbert (7957e377b)
 - Bug 1129040 - Provide a way for content processes to query the chrome side blocklist service. r=billm (5ea750b8b)
 - Bug 1174711 - patch 1 - Rename nsIPercentHeightObserver to nsIPercentBSizeObserver, and update related frame methods to match. r=dholbert (7a17fbe4f)
 - Bug 1174711 - patch 2 - Convert nsTableFrame to work with logical coordinates. r=dholbert (329776abe)
 - Bug 1174711 - patch 3 - Rename a couple more frame-state bits from physical to logical. r=dholbert (d61c9624d)
 - Bug 1174711 - patch 4 - Rename mSpecialHeightReflow to mSpecialBSizeReflow, and update comments to match. r=dholbert (49969e512)
 - Bug 1174711 patch 5 - Copy inline-size rather than width from prev-in-flow when initializing nsTableFrame. r=dholbert (7afb6c80a)
 - pointer style (52cd6f9af)
 - pointer style (ba99b3f74)
 - Bug 1149797 - Use the loop's *static* block object when freshening a loop's block object, then copy in values from the old cloned block. Using the old cloned block directly isn't valid when the cloned block might be extended with additional variables created by eval or added by nested function statements. r=shu (deba22790)
 - Bug 1165486 - Cleanup: use standard object allocation functions when allocating scope objects. (r=terrence) (b35a4d7aa)
 - pointer style (74ea6d085)
 - Bug 1165486 - Add StaticNonSyntacticScopeObjects and teach scope iterators about it. (r=luke) (18a526ed3)
 - Bug 1144371: Implement DEBUG-only JS shell function |dumpStringRepresentation|. r=jandem (35524cf42)
 - Bug 1148963 - OdinMonkey: add CompileOptions::lazyParsingDisabled and testing function setDiscardSource (r=bz) (53f9c8fe6)
 - pointer style (edaa05d69)
 - Bug 1165486 - Rename hasPollutedGlobalScope to hasNonSyntacticScope. (r=luke) (505fc8d05)
 - pointer style (3c4435038)
 - Bug 1165486 - Remove PollutedGlobalScopeOption in favor of using the static scope chain to detect non-syntactic scopes. (r=luke) (73ca126f5)
 - Bug 1150106 - Async loadSubscript - Part 1, add `async` to loadSubscriptWithOptions r=bholley,mcrr8 (a0ffeed0d)
 - Bug 1150106 - Add an async version of loadSubscript - Part 2, tests r=bholley (6aeaa93e4)
 - Bug 1165486 - Use JS::CompileForNonSyntacticScope in Gecko where we used to set polluted global scope. (r=bz, rs=bholley) (073332ac9)
 - Bug 1165486 - Split JS::Compile into JS::Compile and JS::CompileForNonSyntacticScope. (r=luke) (0b82c73c4)
 - Bug 1165486 - Restructure function and script cloning in light of PollutingGlobal scope changes. (r=Waldo) (b4e39652c)
 - Bug 1141865 - Part 7: Make new.target work in generator functions. (r=jorendorff, r=jandem) (a9f5ce1ad)
 - Bug 1165486 - Detect with scopes at parse time using the static scope chain for non-function scripts. Also cache static scope properties on SharedGlobalContext. (r=efaust) (a7d4283cb)
 - Bug 1176070 - Optimize nsTableFrame's FirstInFlow() lookups from calls to GetColumnISize(). r=dholbert (beed47b93)
 - fix misspatch of Bug 1141862 - Part 6 (d091f70a2)
 - Bug 1169736 - Temporarily disallow eval and arrow functions inside derived class constructors. (r=jorendorff) (6cf8a2db1)
 - Bug 1150855 - Disallow method syntax without curly brackets. r=efaust (26086ee95)
 - part of Bug 1150855 - Remove uses of the curly syntax. (bfe86ea19)
 - Bug 1169853 - Require semicolon after export default AssignmentExpression. r=shu (580ff7e30)
 - Bug 1174009 - Remove a couple of unused AutoRooter types; r=sfink (113aee16a) (b3541682d)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1176414 - Clean up/rename various column index variables in nsTableRowFrame and nsTableFrame. r=dholbert (d8e33d549)
 - Bug 1176105 - Remove the (largely gutted) nsTableIterator class, and replace with simple frame-list iteration. r=dholbert (46a15df19)
 - Bug 1176555 - Replace explicit bit-twiddling of frame state flags by human-readable nsIFrame state-manipulation methods in table layout code. r=dholbert (aef9fb035)
 - Bug 1176354 - Rename nsTableRowFrame::GetBSize to differentiate better from nsIFrame::BSize. r=dholbert (9b67203a0)
 - Bug 1177600 - Properly adjust table row and cell positions when their containing block-size changes in vertical-rl writing mode. r=dholbert (413148d55)
 - Bug 1177925: Use range-based for loops for nsFrameList iteration, in various places in layout. r=tn (6db450285)
 - Bug 1148582 - Recycle mask layers attached to FrameMetrics. r=mattwoodrow (91360fbd8)
 - Bug 1148582 - Apply async transforms to (ancestor) mask layers correctly. r=botond (d6fca66fb)
 - Bug 1143575. Add some #includes to avoid unified-build issues on Windows. r=nical (ff78b76a3)
 - Bug 1143575. Add some #includes to avoid more unified-build issues on Windows. r=nical (99e0548f4)
 - Bug 1181832 - Keep gfxPrefs.h out of header files. r=kats (341f0aea0)
 - Bug 1181832 - Include gfxPrefs.h from nsImageLoadingContet.cpp to fix bustage on a CLOSED TREE. r=bustage (d61a18ab9)
 - Bug 1137557 - Part 0: TextEventDispatcher shouldn't forward keyboard events coming from TextInputProcessor to the parent process. r=smaug (87f70fcc4)
 - Bug 1172405 - Fix build error when adding files to dom/workers and gfx/layers. r=nical (8e114b8bb)
 - Bug 1061393 - Export display list info to layer scope viewer. r=kamidphish (60fd247f0)
 - Bug 1072313 - Check for more leaks in the layers code. r=mstange (aacbb2e1c)
 - Bug 1198979 - GC unused IPDL struct member. r=jmuizelaar (79c3113fb)
 - add missing semicolon (17ba58b45)
 - Bug 1153348 - Add an analysis to prohibit operator bools which aren't marked as either explicit or MOZ_IMPLICIT; r=jrmuizel (903dca970)
 - Bug 1170388 - Restrict the static analysis error given about raw pointers to refcounted objects inside a lambda, to the case where the raw pointer is captured. r=ehsan (cbbd3174c)
 - Bug 1180993 - Part 1: Add an analysis to help catch unused return values of specific types. r=ehsan (2420face4)
 - Bug 1180993 - Part 2: Add the MOZ_MUST_USE attribute to mfbt, and use it to verify usage of already_AddRefed. r=ehsan (b72785a60)
 - Bug 1123907 - Part 1: Add an analysis to ensure that a class marked MOZ_NEEDS_NO_VTABLE_TYPE cannot be instantiated by a class with a VTable; r=ehsan (0f5730c04)
 - Bug 1123907 - Part 2: Add MOZ_NEEDS_NO_VTABLE_TYPE to Attributes.h, and use it to verify the EntryType argument of nsTHashtable; r=ehsan (2e80fc581)
 - Bug 1159433 - Part 1: Add an analysis to ensure that some template arguments cannot be non-memmovable types; r=ehsan (3602e950f)
 - uniform MOZ_MUST_USE, it will be renamed in patch 1267550 (3b5ebd4c5)
 - add missing part of Bug 1187073 - Use MOZ_INHERIT_TYPE_ANNOTATIONS_FROM_TEMPLATE_ARGS to validate the usage of AlignedStorage2. r=ehsan (3b451bfb7)
 - Bug 1159433 - Part 2: Add the MOZ_NON_MEMMOVABLE and MOZ_NEEDS_MEMMOVABLE_TYPE macros to MFBT; r=froydnj (171c23ec4)
 - Bug 1159433 - Part 3: Make nsTArray_CopyChooser only accept memmovable argument types by default; r=froydnj (40cb4aa67)
 - Bug 1159433 - Part 4: Mark nsTAutoString_CharT and nsAutoArrayBase as non-memmovable; r=froydnj (7dc6bd220) (75dd89485)
- import changes from `dev' branch of rmottola/Arctic-Fox:
 - Bug 1158456 - Remove control characters from composition string, and add dom.compositionevent.allow_control_characters pref to control it. r=masayuki (5d94de2fd)
 - Bug 1154053 - Limit concurrency of e10s memory reporting. r=erahm (87b9d2335) (5e5396a2f)

* Notice: the changelog above may not always applicable to XULRunner code which K-Meleon uses.

A goanna3 source tree that has kmeleon adaption patch applied is available here: https://github.com/roytam1/palemoon27/tree/kmeleon76

[DanR20]

Does anyone know of any sites that still use flash? I haven't been able to find any lately but there's probably still some around.

For those it might be a good idea to consider keeping an older version for Basilisk since from what I've read Adobe put a time bomb on their last version set to explode Jan 1, 2021. I've got v32.0.0.453 just in case, although there is supposedly a flash emulator available somewhere that will hopefully install in these older FF versions.  

[RainyShadow]

3 hours ago, DanR20 said:

Does anyone know of any sites that still use flash? I haven't been able to find any lately but there's probably still some around.

Newgrounds (although they seem to be transitioning to Ruffle emulation), Kongregate, Ninja Kiwi, Armor Games...

[Ben Markson]

7 hours ago, DanR20 said:

For those it might be a good idea to consider keeping an older version for Basilisk since from what I've read Adobe put a time bomb on their last version set to explode Jan 1, 2021. I've got v32.0.0.453 just in case, although there is supposedly a flash emulator available somewhere that will hopefully install in these older FF versions.  

https://msfn.org/board/topic/174085-newest-adobe-flash-and-shockwave-and-java-too/?do=findComment&comment=1187691

So, broadly, I think you will still be able to use flash but only for domains you explicitly whitelist:

mms.cfg

EOLUninstallDisable=1
EnableAllowList=1
AllowListUrlPattern=file:*
AllowListUrlPattern=https://www.newgrounds.com//

...the EnableAllowList will become depreciated and behave as if it is always 1.

Ben.
 

[InterLinked]

10 hours ago, DanR20 said:

Does anyone know of any sites that still use flash? I haven't been able to find any lately but there's probably still some around.

For those it might be a good idea to consider keeping an older version for Basilisk since from what I've read Adobe put a time bomb on their last version set to explode Jan 1, 2021. I've got v32.0.0.453 just in case, although there is supposedly a flash emulator available somewhere that will hopefully install in these older FF versions.  

Plenty. Especially lesser used, obscure, or academic sites. Keeping an old version on hand would be a smart idea. I've also got that version I believe safely downloaded from the Wayback Archive.

The bomb goes off Jan 12, not Jan 1.

[Monroe]

I have a User Agent utility added to New Moon 28. You can add a few user agents to use instead of the Default. However, when you close down NM and reopen NM 28, the user agent is always back to the default user agent. Is there somewhere in New Moon a setting or someway for the user agent to always remain on the the user agent I have checked till I change it myself?

...

[Nemoplus]

Hello everyone :) I was wondering about one thing. I'm using New Moon on my old PC with XP and Facebook works really slow and laggy since they change their layout into new one. Especially Messenger when you have to wait few seconds to see letters on screen after typing - before everything works fine. Is there any way to handle with that for this moment? Mostly I care about Messenger - it's kinda difficult to contact with people with those issues. I'll be glad for any help and thank you for all answers.

[DanR20]

11 hours ago, RainyShadow said:

Newgrounds (although they seem to be transitioning to Ruffle emulation), Kongregate, Ninja Kiwi, Armor Games...

Thanks for those links, I can bookmark them to test out later. Ruffle emulation is what I had read about and should install in these older FF versions.

@Ben Markson

That's good to know about the whilelisting, if it holds then I can reinstall 32.0.0.465 back onto Win7, but will be keeping 32.0.0.453 just in case.

@InterLinked

Not suprising, there are academic sites still using the old Shockwave player so flash will probably be around a while as well.

 

[siria]

FLASH timebomb:
have read here the last clean version is 371 already:
https://msfn.org/board/topic/181941-adobe-flash-player-removal-updates/?do=findComment&comment=1192177
And that once updated, it's near impossible to downgrade again to an older Flash version...

Monroe / useragent addon:
those UA resets to default string are usually done actively by the addon-script themselves. Try get another one, or alternatively modify its script?
In K-Meleon the OLD macro uses this pref to decide whether or not to reset:
kmeleon.plugins.macros.useragent.reset_on_exit
Well possible other browers use a pref too, just no clue which.
Another alternative, reset the global override-pref at every session start automatically to a string of your choice, defined in ...profile/user.js (can be changed during session, only gets reset at session starts)
.

Edited December 26, 2020 by siria

[DanR20]

2 hours ago, siria said:

FLASH timebomb:
have read here the last clean version is 371 already:
https://msfn.org/board/topic/181941-adobe-flash-player-removal-updates/?do=findComment&comment=1192177
And that once updated, it's near impossible to downgrade again to an older Flash version...

If Ben Markson's solution works then having the last version shouldn't matter??  For me since I keep the flash dll in the basilisk\browser\plugins folder whitelisting will have no effect, it will only for the IE activex version so I'm going to download version 371 for basilisk, thanks for the link. And thanks for nothing Adobe. :--) 

[NotHereToPlayGames]

@VistaLover

3 hours ago, siria said:

FLASH timebomb:
have read here the last clean version is 371 already:
https://msfn.org/board/topic/181941-adobe-flash-player-removal-updates/?do=findComment&comment=1192177
And that once updated, it's near impossible to downgrade again to an older Flash version...

It's shennanigans like this that are one of about two hundred reasons why NOTHING on my computer "auto-updates". 

Although on the other hand, I also don't recall seeing any Flash Content from my browsing habits for YEARS.

I do have a few Soduku sites and crossword sites on the computer at work, but I don't think they are Flash but can't say for sure from home.

Edited December 26, 2020 by ArcticFoxie