Jump to content

Leaderboard

Popular Content

Showing content with the highest reputation on 11/13/2024 in all areas

  1. Unfortunately, my time and resources are limited. Researching and writing such articles is very time-consuming. Any help, any testing of programs, any posting of experiences already made regarding such security programs for Windows XP is therefore always welcome.
    2 points
  2. I for one pretty much assumed that you do. But it still was a gigantic surprise that you posted the link without manually removing the vvveeerrryyy lllooonnnggg aaannnddd uuunnnnnneeeccceeeessssssaaarrryyy pppooorrrtttiiiooonnnsss ooofff ttthhheee UUURRRLLL.
    1 point
  3. @roytam1 eBay must have recently changed their layout. Pages with items lead now to a 100% CPU utilisation on my single core machine, even with uBO. The actual causer seems to be the item images and in some cases also the item description. Weeks ago, it was working fine. Now, surfing eBay with New Moon 28 is no fun anymore. BTW, same issue in Mypal 68.
    1 point
  4. If you call me Mr. OT Police, then I regret to inform you at this point that you are the Mr. OT Master.
    1 point
  5. I have made a small change in the All about: custom button to reduce the size of the tooltip popup by half. Now it looks like this: Is it better that way?
    1 point
  6. I sincerely hope that constantly posting offtopic isn't a form of senility. In the last few pages, one doesn't get anything about the actual topic here. @roytam1 must have a very thick skin.
    1 point
  7. @feodor2 I am not sure if you aware of this. The logger of uBlock Origin 1.60 does not work in Mypal 68.14.4b when running single-process mode. Neither in the officially last compatible version 1.46 of uBlock Origin. In multiprocess mode, the logger is running fine.
    1 point
  8. Your assumption is unfortunately wrong. If you had read my article carefully, you would have noticed that the All about: custom button is not a fork of an extension but of an old custom button. When I create or fork custom buttons, then a well-functioning toolbar button is the main target. So, using the All about: custom button means using the toolbar button. That's why it is called custom button and not extension. I could implement such functionality you requested but I am not interested in. Personally, I am a fan of toolbar buttons with multiple and meaningful functions.
    1 point
  9. That basically kind of makes my point. MSFN will not teach you how to debloat XP, Vista, 7, 10, 11... People that truly want to learn that don't land here at MSFN. Sorry, they just don't. (At least I certainly don't see it!) Most of us forget that XP was also a very HEAVY operating system with a lot of "bloat". We just had 20 years to get it to our liking and only want to spend 20 minutes to get our next OS "to our liking".
    1 point
  10. So you promise you will never spread lies about me ever again?
    1 point
  11. 1 point
  12. I have no idea what you're talking about, and you refuse to even introduce yourself.
    1 point
  13. No, in English they use Capslock or Shift to capitalise words.
    1 point
  14. I'm sorry, who are you? Remind me. What do you do? Are you a developer? Do you write browsers or the such?
    1 point
  15. The early internet that we knew was a big library where everyone shared info and content. The modern internet is a shopping centre, and browsers money makers machines. Just fire the start page and it is already making money from you....do a search, visit a page, everything.... and remember.,,.... if you computer goes slow or run out of memory... its cause you need to buy new hardware for internet browsing... Adblocking is just the top of the big iceberg...
    1 point
  16. New build of post-deprecated Serpent/moebius for XP! * Notice: This repo will not be built on regular schedule, and changes are experimental as usual. ** Current moebius patch level should be on par with 52.9, but some security patches can not be applied/ported due to source milestone differences between versions. Test binary: Win32 https://o.rthost.win/basilisk/basilisk55-win32-git-20241102-df498d40f-xpmod.7z Win64 https://o.rthost.win/basilisk/basilisk55-win64-git-20241102-df498d40f-xpmod.7z repo: https://github.com/roytam1/basilisk55 Repo changes: - import from UXP: Issue #2646 - Parse "variations" keywords. (8d536c0c) (d3002349e) - import from UXP: Issue #2648 - Apply CSP to documents sent through multipart/x-mixed-replace (6fe326c7) (fa67c714e) - import from UXP: Issue #2648 - Inherit the content disposition of the multipart response (5c0c0c02) (31da70f0e) - import from UXP: Bug 1920800 - Remove oldImages from check for flushed old principal handle (9ba1057e) (df498d40f)
    1 point
  17. New build of BOC/UXP for XP! Test binary: MailNews Win32 https://o.rthost.win/boc-uxp/mailnews.win32-20241102-7bda12e7-uxp-4397f384a7-xpmod.7z BNavigator Win32 https://o.rthost.win/boc-uxp/bnavigator.win32-20241102-7bda12e7-uxp-4397f384a7-xpmod.7z source repo (excluding UXP): https://github.com/roytam1/boc-uxp/tree/custom * 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.win32-20241102-id-656ea98-uxp-4397f384a7-xpmod.7z IceApe-UXP(suite) https://o.rthost.win/hbl-uxp/iceape.win32-20241102-id-656ea98-ia-93af9a0-uxp-4397f384a7-xpmod.7z source repo (excluding UXP): https://github.com/roytam1/icedove-uxp/tree/winbuild https://github.com/roytam1/iceape-uxp/tree/winbuild
    1 point
  18. New build of Serpent/UXP for XP! Test binary: Win32 https://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20241102-3219d2d-uxp-4397f384a7-xpmod.7z Win64 https://o.rthost.win/basilisk/basilisk52-g4.8.win64-git-20241102-3219d2d-uxp-4397f384a7-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-20241102-3219d2d-uxp-4397f384a7-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-28.10.7a1.win32-git-20241102-d849524bd-uxp-4397f384a7-xpmod.7z Win32 IA32 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20241102-d849524bd-uxp-4397f384a7-xpmod-ia32.7z Win32 SSE https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20241102-d849524bd-uxp-4397f384a7-xpmod-sse.7z Win64 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win64-git-20241102-d849524bd-uxp-4397f384a7-xpmod.7z Official UXP changes picked since my last build: - Revert "Issue #2645 - Fix type confusion for `ResumeKind`" (8bbcb50b62) - Issue #2646 - Parse "variations" keywords. (8d536c0cc7) - Issue #2648 - Apply CSP to documents sent through multipart/x-mixed-replace (6fe326c7bb) - Issue #2648 - Inherit the content disposition of the multipart response (5c0c0c02d9) - Bug 1920800 - Remove oldImages from check for flushed old principal handle (9ba1057e54) No official Pale-Moon changes picked since my last build. No official Basilisk changes picked since my last build. Update Notice: - You may delete file named icudt*.dat inside program folder when updating from old releases. * Notice: From now on, UXP rev will point to `custom` branch of my UXP repo instead of MCP UXP repo, while "official UXP changes" shows only `tracking` branch changes.
    1 point
  19. 1 point
  20. Excuse me, but this is extremely naive point. FSB is KGB that change nickname to drop the heat. Its like in GTA when you have 6 star chase and you drive in paint shop to change paint color and engine number. NPC in game obey the script. But you are alive human been, and not primitive scripted NPC bot, you should understand simple things, right? FSB\KGB is the most prominent terroristic organization in the world. It stands behind 90% percent of organized world terrorism, from ISIS, to Al-Quaida, Irish National Liberation Army, Taliban, HAMAS. All of this is made possible just cause KGB\FSB. Chekists came to power at the end of 1917, then never give up power since that year. They still in charge of Kremlin. Security Services of Democratic countries exist to prevent crime and terrorism. FSB\KGB is a major terroristic organization of the world - it exist to create chaos, corruprion, wars and terrorism across the world. P.S. Here is a good read about FSB one "special operation": https://en.wikipedia.org/wiki/Blowing_Up_Russia
    1 point
  21. Don't underestimate the competence of the French intelligence services: DGSE, DGSI, and the DST.
    1 point
  22. I think Durov already started talking, it's just we don't see the full picture because of the nature of the investigation, they tell us about 000.1%, and now he needs to avoid drinking russian tea.
    1 point
  23. Today’s Russian papers on Pavel Durov’s detention in Paris: * “If Telegram crashes, how will [our army] fight?" Courtesy and copyright of BBC.UK. (translated papers with English subt.). Interesting! https://video.twimg.com/amplify_video/1827954149629337600/vid/avc1/1280x720/YeHbAsETO6SsYZLr.mp4
    1 point
  24. I completely agree. Image how difficult will be for somebody, who has never used Pentium I with Windows 98, to understand how fast Windows was back in the day in a hardware that was 100 times slower than the one we have nowadays. Now... ¿how is this possible? or ¿what is happening? Possible solution: ¿Time to get back to SERVICE PACKS? It feels like a BETA OS Since Windows 10 1607. Windows as a Service isn't the path to follow, MS. This is completely no sense: https://winclassic.net/thread/1652/list-metro-ified-components-windows Current Windows status can be only understood in one way: nowadays hardware, even the cheapest laptop, comes with tons of RAM and disk space to waste... It's a bit sad how all has changed after w8.1.... we are in the last days of computers. Sometimes I think if MS is only testing Windows on a highend PC, even if this is true, the OS feels like a beta; slow gui, lag... tons of background processes... It was one day an (almost) perfect Operating System and not this kind of beta "Service"
    1 point
  25. somewhat off-topic: since I got some free time because of work-from-home (as covid delta and omicron pandemic in hong kong at the moment), I tried to make fun by putting newmoon 28 app into SP55 engine. with some code changes, it is somewhat working. (but not packagable in the moment)
    1 point
  26. New regular/weekly KM-Goanna release: https://o.rthost.win/kmeleon/KM76.4.5-Goanna-20220205.7z Changelog: In-tree changes: - update XULRunner-sdk patch as of NM27 git rev a988f766 (127240e9) Out-of-tree changes: * update Goanna3 to git d8ae51153...a988f766e: - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 870173 - Upgrade to clang 3.3. r=ehsan (f654c5710a) - add missing and previous stuff (4d3e5dbf07) - Bug 1155476 - Update sccache to 155c926 (c72624ae94) - goanna->gecko and permissions (a6aa2bb828) - Bug 1203857 - Allow to build standalone js against jemalloc4. r=gps (cfe96f97a6) - Bug 1186424: Remove AsmJS min heap length optimization; r=luke (199d840dcf) - Bug 1198224 - OdinMonkey: hash the signatures, not their addresses (r=bbouvier) (c66aa93445) - Bug 1178793 - Make asm.js atomic primitives return Int, not Intish. r=luke (cb64cb1ff5) - Bug 1159473 - Add Mac-specific debug logging code. r=spohl (180e51c1f7) - fix mispatch (3bbd00e849) - Bug 1186158 - Receive notifications of sandbox violations in the browser on OS X. r=spohl (e22b97dd79) - Bug 1152585 - Implement GetClientOffset in nsChildView to fix child offset calculation when drawInTitlebar=false r=mstange (89d4ee9d52) - [Adapted not to drop 10.5] Bug 1172664 - Use [NSApplication setPresentationOptions:] instead of SetSystemUIMode for hide os chrome. r=smichaud Bug 1188322 - Always hide menubar as well as dock for fullscreen on OS X whatever the screen is. r=mstange (872ef3e90f) - Bug 1180564: Pref off NSTextInput protocol methods in anticipation of removing them. Also re-organize them so they're all in one spot and fix an unrelated warning. r=masayuki (aef2c0bb7e) - add back some crash reporter stuff (75dfba90be) - Bug 1188425 - Add a pref to allow insecure text input, thereby avoiding crashes at [ChildView keyDown:] in non-release builds. r=spohl (e089d0ccea) - Bug 1193062 - Give synthesized NSEvents a timestamp that is in the right space. r=smichaud (452f7992ca) - Bug 1193977 - Bind the texture unit we want to use when compositing the OSX titlebar. r=jrmuizel (063830bda1) - Bug 976722 - Make -moz-appearance:toolbox influence the sheet attachment edge. r=roc (2e87d2e340) - Bug 1154676 - Add a null-check for the presshell. r=mstange (a41d28effc) - Bug 1154322 - Always use DrawTargetCG when painting directly into a window on OS X (unaccelerated drawRect rendering). r=jrmuizel (af405884eb) - Bug 1148868 - Fix up nsChildView::ConfigureAPZControllerThread(). r=kats (497a262ad8) - Bug 1141917 - Remove unused member function RectTextureImage::UpdateFromDrawTarget(). r=mstange (36182de410) - Bug 1193062 - Make nsChildView send PanGestureInput events into APZ. r=kats, r=smichaud (696b0a9886) - Bug 1110888 follow-up: Mark SetPluginFocused as override (26805abd52) - Bug 1016035 - Move some code around. r=kats (c786056e9b) (abb8daa3f) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1178833 - Fix for unified builds when a new file is added to layout/base. r=mstange (bf961c00fd) - Bug 1178847 - Add a MobileViewportManager to manage setting the CSS viewport on B2G. r=botond,tn (937ba486fa) - Bug 1178847 - Don't use the MobileViewportManager code on Mulet or Fennec yet. r=botond (3061b2c617) - Bug 1168950. Add the document to DoProcessRestyles profile label. r=dbaron (2347b76411) - Bug 1187792 - Don't set the SPCSPS unless meta-viewport or zooming is enabled. r=botond (a8f5ed1006) - Bug 1193062 - Don't double-send target APZC confirmations that might race each other. r=kats (de48467401) - Bug 1186004 - Add a pref to enable/disable APZ zooming behaviour. r=botond (fc3653c979) - Bug 1193062 - Add a PAN_MOMENTUM state. r=kats (10c8bb1a5e) - Bug 1148350 - Have ConvertToGecko use a live screen-to-apzc transform to avoid transform mismatches. r=botond (9eee9ea6f2) - Bug 1169844 - Remove incorrect assertion and properly untransform wheel events for overscroll situations. r=botond (9b101b53bb) - Bug 1169690 - Accelerate flings correctly in the presence of overscroll handoff. r=kats (49d55c05f4) - Bug 1158424 - Extract a BreadthFirstSearch() helper function. r=kats (6a49498185) - Bug 1158424 - Add APZCTreeManager::FindRootContentApzcForLayersId(). r=kats (d6c7ecf781) - Bug 1158424 - Fix APZCTreeManager::GetMultitouchTarget(). r=kats (bf8571a4ca) - Bug 1174205 - Make sure to acquire the tree lock before finding the multitouch target. r=botond (bdbd98f2c7) - Bug 1211660 - Remove dead obj argument from two functions. r=Waldo. (efba5a2f8d) - Bug 1158424 - Undeprecate HasNoParentWithSameLayersId(). r=kats (d994ea28f8) - Fix AsyncCompositionManager not comparing layer tree IDs when finding scrollbar targets. (bug 1151169, r=kats) (c7dbc4776a) - Bug 1163259. Make sure we don't look at the parent ref layer when looking for a layer with the frame metrics for a scrollbar. r=kats (8b99d2c830) - Bug 1164340. Convert the scrollbar thumb ratio to a unitless ratio properly for non-zoomable scrollframes. r=botond (27f56cab56) - Bug 1164340. Fix up some scrollbar related comments in AsyncCompositionManager. r=botond (b9d540bea9) - Bug 1164767. Modify async scrollbar positioning code to deal with containerless root scrollables. r=botond (501d07592e) - Bug 1158424 - Remove FrameMetrics::IsRootScrollable() (it just duplicated IsRootContent()). r=kats (672fd034b3) - Bug 1120683 - Properly handle unapplying 3D projective transforms throughout APZ code. r=botond (a46612811a) - Bug 1148868 - Make APZ wheel handling handle pixel scrolling on OS X. r=dvander, r=kats (da3bd34135) - Bug 1193062 - Add fields to PanGestureInput and ScrollWheelInput. r=kats (716ba34e18) - Bug 1189565 - Send resize events when backing scale factor changes. r=mstange (4e81a7bbf7) - Bug 1193062 - Add nsCocoaUtils::ModifiersForEvent. r=smichaud (da67b22ff7) - Bug 1016035 - Remove the ability to swipe vertically. r=kats (fa315e0b97) - Bug 1016035 - Refactor the code that decides whether we want to start a swipe. r=kats (3091961328) - Bug 1016035 - Make the threshold in AxisPhysicsMSDModel::IsFinished controllable by the caller. r=kip (76df4d85a2) - Bug 1016035 - More swipe refactoring. r=kats (2abcd05b33) - Bug 1016035 - Implement the swipe animation ourselves instead of calling the NSEvent trackSwipe API. r=kats (4842b2a728) - Bug 1016035 - Move swipe tracking code into DispatchAPZWheelInputEvent. r=kats (bb1052b87c) - Bug 1016035 - Also mark widgetWheelEvents that are handled by APZ with mCanStartSwipe. r=kats (21809b902d) - Bug 1016035 - Put PanGestureInput events into a queue if they can end up as a swipe. r=kats (2ebac59047) - Bug 1016035 - Split up MaybeTrackScrollEventAsSwipe into SendMayStartSwipe and TrackScrollEventAsSwipe. r=kats (ced10d9671) - Bug 1016035 - Send MayStartSwipe event before sending the wheel event. r=kats (b3346b81fb) - Fix some fields missing in the widget event IPC glue. (bug 1139711, r=kats) (1298b25705) - Bug 1016035 - Add mCanTriggerSwipe and TriggersSwipe(). r=masayuki (664a059972) - Bug 1016035 - Make APZEventState report defaultPrevented=true when the event started a swipe. r=kats (128d627ac7) - Bug 1016035 - Set scroll overflow information on potential swipe start events that have been processed by APZ. r=masayuki (b5931f6320) - Bug 1206801 - fix broken CONFIG['DEBUG'] checks in moz.build files; r=mshal (292d5ac709) - Bug 1186424: Refactor AsmJSValidate.cpp in preparation for parallelization; r=luke (4868bc66ca) - Bug 1129267 - Fix by backing out patch for bug 1118615. r=mstange (ea8aaba04e) - Bug 1182411 Part 1: Make plugin quirks available to the Parent as well as the Child. r=jimm (1f2551a6b7) - Bug 1228462 - "PluginQuirks.* uses DOS CRLF". r=jmathies (b569a6c6c8) - Bug 1152300 - When focus shifts from a native plugin window to content, content needs to forward the request for focus change to chrome. r=enn (5bd811fadd) - Bug 1185529: On Windows, ensure that NPAPI child window has the correct parent before setwindow is called. r=jimm (9d94fdf2ff) - Bug 1182411 Part 2: Change winless popup surrogate to have its parent set in the chrome process. r=jimm (adee9bb012) (e9ba40cf6) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1144112 - Rearrange code a little to simplify it. No functional changes intended. r=botond (585aec1682) - Bug 1144112 - When tagging a block as fast-motion with touch-action enabled, make sure to set the allowed behaviors on it as well. r=botond (6c969faab6) - Bug 1172541 - Don't skip touch event delivery during smooth scroll. r=kats (c1e81e308e) - Bug 1194876 - Extract a TouchCounter class. r=botond (742458342a) - Bug 1159455 - Fix assorted debugging strings in APZ code. r=botond (4f608f0e0a) - Bug 1194876. Reset the input state on an APZC if a new touch block targets a new APZC while there were touches still active. r=botond (481002eaa9) - Bug 1194876 - Do a better job of resetting input state. r=botond (9fc5a3a605) - Bug 1143575. Fix build by ensuring ColorLayerComposite uses namespace gfx. r=nical (429d9f9abe) (16a3a03bc) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1193062 - Add PanGestureBlockState. r=kats (a8da88a934) - Bug 1193062 - Make PanGestureInput transform processing work like ScrollWheelInput processing. r=kats (fb17c120e8) - Bug 1180780 - Enable handling of pixel-scrolling wheel events with APZ on Windows and Linux. r=mstange (41d8fd5ed2) - missing part of Bug 1013412 (4fa14a56d9) - Make sure that confirmed APZ wheel targets are always scrollable. (bug 1154132, r=kats) (c35e3a34ca) - breakage fixes (96a1e3c36a) - Bug 1153711 - Do not discard a component of a fling if an APZC further in the handoff chain has room to scroll in that direction. r=kats (4f35dfd913) - Bug 1193062 - Set correct axis velocities when using PanGestureInput events. r=kats (7036040c68) - Clamp APZ wheel animations to their scrollable range. (bug 1152011 part 2, r=botond) (45c4925fee) - Bug 1193062 - CanScrollWithWheel needs to use ParentLayerCoords for the scroll delta. r=kats (20a3c4ac6b) - Bug 1193062 - Make OverscrollHandoffChain::FindFirstScrollable and AsyncPanZoomController::CanScroll able to deal with PanGestureInput events. r=kats (d87ec270db) - Bug 1193062 - Add mHandledByAPZ on PanGestureInput and ScrollWheelInput, and sync the information to the WidgetWheelEvent. r=kats (6e72f5885f) - Bug 1016035 - Delay the processing of a PanGestureInput block until we know whether it's a swipe. r=kats (77d1dceb04) - Bug 1145084 - Ensure that the 300ms content timeout in the APZ code also applies to the touch-action allowed behaviours. r=botond (45f58fe247) - Bug 1193062 - Make AllowScrollHandoff work for both ScrollWheelInput and PanGestureInput blocks. r=kats (d335008b0a) - Bug 1193062 - Remove mPanGestureState. r=kats (774bb3dedf) - Bug 1193062 - Use ScrollSource::Wheel for pan gesture events. r=kats (e9328b0bc4) - Bug 1193062 - Don't use PanGestureInput events for instant wheel scrolling. r=kats (7d89c5dc2e) - Bug 1175564 - don't lock y-axis if pan displacement is 0. r=kats (372d76c914) - Bug 1193062 - Process pan gesture deltas in begin+end events. r=kats (f29072bf8f) - Bug 1193062 - Fix UntransformVector w coordinate checks. r=kip (73b459e459) - Bug 1016035 - Replay the queue to the swipe tracker once the swipe start confirmation arrives. r=kats (9891444d4f) - Bug 1016035 - Don't wait for content to say that we need to swipe if APZ has enough information. r=kats (2b94c25bd3) - Bug 1016035 - Swallow the rest of the scroll gesture after swiping without APZ. r=kats (f9250388f9) - Fix crashes when a wheel transaction has a null confirmed apzc. (bug 1147249, r=kats) (ef898a9f3a) - Bug 1176001 - Turn unconditional log into a manually-enabled log. r=botond (1d866b1ffa) - Don't have a default repaint interval of "never" for APZ animations. (bug 1152138, r=botond) (7e276ba862) - Bug 1152051 - Ensure that destroying an overscroll animation always clears the state Axis tracks about it. r=Cwiiis (1273aac828) - Request content repaints during APZ wheel animations. (bug 1152138, r=botond) (ba3e24011b) - Bug 1164557 - Use COORDINATE_EPSILON in IsZero(). r=kats (c07e506607) - Bug 1164557 - Weed out spurious calls to Axis::OverscrollBy() caused by rounding error r=kats (15f8d18a7e) - Bug 1186164 - When clearing the overscroll animation state, make sure# etOverscroll() continues to reflect the correct direction of overscroll. r=Cwiiis (282c63383c) - Bug 1198900 - Don't call OnTouchCancel when the current block is not a touch block. r=botond (b724d8d3ef) - Bug 788873 - Only rebind a GLXPixmap if the texture has changed. r=karlt (e1f72858ee) - Bug 1169270 - Minor style fixes. r=dvander (494449170d) - Bug 1145702 - Make the assertion in Axis::GetOverscroll() more informative. r=Cwiiis (dff257bbbd) - Bug 1152051 - During an overscroll animation, detect a peak even if a sample has a velocity of exaclty zero. r=Cwiiis (a5aad5349f) - Bug 1201101 - Enable axis-locking over multiple APZCs. r=botond (21574b80ec) - test of Bug 1143522 - Convert Layer::mClipRect to ParentLayerIntRect. r=botond (8f729ca8b0) - Bug 1159398 - Set the controller thread when running APZ gtests. r=dvander (4349b6722b) - Bug 1148350 - Add a test. r=botond (72e3c0716d) - Bug 1152051 - Gtest. r=Cwiiis,kats (bf337c0d6d) - missing of Bug 1158122 - Remove most occurences of nsIntRect in gfx/tests. r=nical (a328773d9d) - missing test of Bug 1160566 - Make FrameMetrics.mCompositionBounds private (091db42632) - Bug 1163845 - Unify handling of time in APZ gtests. r=kats (2366690fe7) - Fix APZ scroll wheel animations often prematurely ending on the first frame. (bug 1157409, r=kats) (78cc0e2292) - Bug 1169689 - Make sure we don't end up with a spurious velocity if we're not going to be panning. r=botond (8911802134) - Bug 1203744 - Fix the SCREEN blend mode with CompositorOGL when the source data is unpremultiplied. r=dvander (e256f83a14) (d88cb9e20) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1157984 - Part 1: Extend gfx::2d classes to support both float and double precision,r=jrmuizel (d8d0cadcb0) - Bug 1157984 - Part 2: Implement double precision clipping functions in Matrix4x4,r=vlad (f31c243a9d) - Bug 1157984 - Part 3: Correct bounding box transformations to support projections and correct clipping when transforming behind the camera,r=vlad (c9c678905b) - Bug 1157984 - Part 4: Remove gfxRect::TransformBounds,r=vlad (71d076d48b) - Bug 1157984 - Part 5: Test,r=vlad (1c9c711745) - Bug 1159985 - In APZ gtests, allow panning in both directions. r=kats (b4a42c580a) - Bug 1166871 - Add a test. r=botond (b08332347f) - Bug 1159985 - Gtest. r=kats (ad41b809ff) - Bug 1169695 - Make the test timestamp privately held by the MockContentControllerDelayed. r=botond (4f898a778c) - Bug 1169695 - Advance the stored time when sampling the transform for a future frame. r=botond (77d029383e) - Bug 1169695 - Store task run-at times in the task queue. r=botond (4bc228d713) - Bug 1171312 - Add generic tree search algorithms for layers-related data structures. r=botond (675806f3e1) - Bug 1200063 - Make APZCTreeManager the place where GetFrameTime() can be overridden. r=kats (32ba198192) - Bug 1200063 - Share a paint throttler between APZCs in the same layers id. r=kats (a2147d25e6) - Bug 1200063 - Add a generic implementation of nsITimerCallback that's usable with a lambda or other function object. r=kats (6d7ef71d44) - Bug 1200063 - Make sure TaskThrottler sends its pending repaint request eventually, even if it never receives a notification from the previous request. r=kats (6eceede1e2) - Bug 1200063 - Rename MakeAPZCInstance to NewAPZCInstance for consistency. r=kats (6927e90644) - Bug 1144374 - Fix APZ code to deal with layers getting moved to another LayerManager. r=botond (eb682a58cd) - Bug 1193930 - Allow the user to start a pinch while an overscroll animation is still in progress. r=kats (44c2e57123) - Bug 1169690 - Gtests. r=kats (3c206d1cb2) - Bug 1201098 - Consume fling and overscroll velocity per-axis. r=botond (1e05677586) - Bug 1158424 - Clean up uses of HasNoParentWithSameLayersId() in overscroll handoff chain building. r=kats (0ed1ddee59) - Bug 1201277 - Replace functors with lambdas in APZCTreeManager. r=botond (ec5476a598) - Bug 1180030 - Request a fling snap when an overscroll animation completes. r=kats (e789a2ebb3) - Bug 1159985 - If only one component of a fling is in overscroll, continue the fling in the other component. r=kats (81998d723f) - Bug 1031443 - Allow panning in the pinching state even if zooming is disabled;r=botond (4c90350cc6) - Fix APZ not starting and updating wheel animations within an APZC lock. (bug 1152011 part 1, r=botond) (ae0d4d0aa8) - Bug 1145089 - Update documentation for nsEventStatus_eConsumeNoDefault. r=botond (7ca605ad96) - bug 1170988 - Fix skia to build for iOS. r=gw280 (fb375def29) - bug 1171642 - Disable some skia arm assembly for iOS. r=gw280 (004245c173) - fix build on gcc 11 (1c511ac81b) - missing bit of Bug 1188462 - Rename gfx/skia/trunk directory to gfx/skia/skia. (ca8f138b28) - Bug 1180030 - Do not request a fling snap if the fling will overscroll on its first sample. r=kats (55f89d55b5) - Bug 1167721 - Ensure we trigger a repaint when dropping velocity to zero in CancelAnimation. r=botond (d04fbb2a50) - Bug 1158340 - Refactor RedistributeDisplayPortExcess. r=kats (27725c3ee4) - Bug 1189565 - Only factor in the async zoom change. r=kats (57e357a038) - Bug 1192466 - Prune header includes for SkiaGLGlue.h - r=bgirard (37d31a1238) - Bug 1004489 - propagate Cairo font AA settings to Skia font. r=gw280 (9adc439b19) - Bug 1171528 - Remove overflowed nscoord_MAX warnings from nsRect. r=dholbert (08599b00e9) - Bug 1202430 - remove NS_GFX and related macros; r=jrmuizel (cb0b79951e) - Bug 1181317 - Switch drawing order of filter feBlend inputs. r=mstange (4914444219) - Bug 1180379. Split out GMAX4500 and block those devices from D2D. r=Bas (c8d1a755ea) - Bug 1073117 - Theme issues with GTK 3.14 - fix gtk button and entry size, r=karlt (0df9ac5c7f) - Bug 1187237 - Correct range slider drawing on GTK3. r=karlt (0d671487da) - Bug 1171696 - Require room for scrollbar sliders and draw troughs only when there is room for a slider or buttons on GTK. r=karlt (71710baeaa) - Bug 1202317. Optimize away identity ColorMatrix nodes. r=mstange (6c57a71e0c) - Bug 1202317. Support PodEqual over fixed-length arrays. r=Waldo (060f20b429) - Bug 1203078 - whitelist gdk-pixbuf image formats. r=dan (5eb43e4235) - Bug 1202317. Optimize feComposite type="arithmetic" with all-zero k values to transparent black. r=mstange (47b2d40678) - Bug 1189443 - Don't round down the margin amounts when inflating the displayport. r=dvander (4afcf087ab) - Bug 1205741 - add xpm and svg to supported gdk-pixbuf formats. r=acomminos,dveditz (4c356c961a) - Bug 1166741 - Workaround for GTK3 bug where file choosers' delegate refcount is not incremented. r=karlt (9d59be7b05) - Bug 1166741 followup: Guard GTK file-picker member-var decl/init statements with same #ifdef that its usages are guarded with, to avoid -Wunused-private-field warning. r=karlt (25bbf34486) - fix build on 10.5/10.6 (9acd104c9f) - namespaces (358de9591c) - Bug 1178842 - Fix unified build for nsDisplayList.cpp when a new file is added to layout/base. r=mstange (aa8ff77615) (a988f766e) * 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
    1 point
  27. New NewMoon 27 Build! 32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20220205-a988f766e-xpmod.7z 32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20220205-a988f766e-xpmod-sse.7z 32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20220205-a988f766e-xpmod-ia32.7z 64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20220205-a988f766e-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 870173 - Upgrade to clang 3.3. r=ehsan (f654c5710a) - add missing and previous stuff (4d3e5dbf07) - Bug 1155476 - Update sccache to 155c926 (c72624ae94) - goanna->gecko and permissions (a6aa2bb828) - Bug 1203857 - Allow to build standalone js against jemalloc4. r=gps (cfe96f97a6) - Bug 1186424: Remove AsmJS min heap length optimization; r=luke (199d840dcf) - Bug 1198224 - OdinMonkey: hash the signatures, not their addresses (r=bbouvier) (c66aa93445) - Bug 1178793 - Make asm.js atomic primitives return Int, not Intish. r=luke (cb64cb1ff5) - Bug 1159473 - Add Mac-specific debug logging code. r=spohl (180e51c1f7) - fix mispatch (3bbd00e849) - Bug 1186158 - Receive notifications of sandbox violations in the browser on OS X. r=spohl (e22b97dd79) - Bug 1152585 - Implement GetClientOffset in nsChildView to fix child offset calculation when drawInTitlebar=false r=mstange (89d4ee9d52) - [Adapted not to drop 10.5] Bug 1172664 - Use [NSApplication setPresentationOptions:] instead of SetSystemUIMode for hide os chrome. r=smichaud Bug 1188322 - Always hide menubar as well as dock for fullscreen on OS X whatever the screen is. r=mstange (872ef3e90f) - Bug 1180564: Pref off NSTextInput protocol methods in anticipation of removing them. Also re-organize them so they're all in one spot and fix an unrelated warning. r=masayuki (aef2c0bb7e) - add back some crash reporter stuff (75dfba90be) - Bug 1188425 - Add a pref to allow insecure text input, thereby avoiding crashes at [ChildView keyDown:] in non-release builds. r=spohl (e089d0ccea) - Bug 1193062 - Give synthesized NSEvents a timestamp that is in the right space. r=smichaud (452f7992ca) - Bug 1193977 - Bind the texture unit we want to use when compositing the OSX titlebar. r=jrmuizel (063830bda1) - Bug 976722 - Make -moz-appearance:toolbox influence the sheet attachment edge. r=roc (2e87d2e340) - Bug 1154676 - Add a null-check for the presshell. r=mstange (a41d28effc) - Bug 1154322 - Always use DrawTargetCG when painting directly into a window on OS X (unaccelerated drawRect rendering). r=jrmuizel (af405884eb) - Bug 1148868 - Fix up nsChildView::ConfigureAPZControllerThread(). r=kats (497a262ad8) - Bug 1141917 - Remove unused member function RectTextureImage::UpdateFromDrawTarget(). r=mstange (36182de410) - Bug 1193062 - Make nsChildView send PanGestureInput events into APZ. r=kats, r=smichaud (696b0a9886) - Bug 1110888 follow-up: Mark SetPluginFocused as override (26805abd52) - Bug 1016035 - Move some code around. r=kats (c786056e9b) (abb8daa3f) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1178833 - Fix for unified builds when a new file is added to layout/base. r=mstange (bf961c00fd) - Bug 1178847 - Add a MobileViewportManager to manage setting the CSS viewport on B2G. r=botond,tn (937ba486fa) - Bug 1178847 - Don't use the MobileViewportManager code on Mulet or Fennec yet. r=botond (3061b2c617) - Bug 1168950. Add the document to DoProcessRestyles profile label. r=dbaron (2347b76411) - Bug 1187792 - Don't set the SPCSPS unless meta-viewport or zooming is enabled. r=botond (a8f5ed1006) - Bug 1193062 - Don't double-send target APZC confirmations that might race each other. r=kats (de48467401) - Bug 1186004 - Add a pref to enable/disable APZ zooming behaviour. r=botond (fc3653c979) - Bug 1193062 - Add a PAN_MOMENTUM state. r=kats (10c8bb1a5e) - Bug 1148350 - Have ConvertToGecko use a live screen-to-apzc transform to avoid transform mismatches. r=botond (9eee9ea6f2) - Bug 1169844 - Remove incorrect assertion and properly untransform wheel events for overscroll situations. r=botond (9b101b53bb) - Bug 1169690 - Accelerate flings correctly in the presence of overscroll handoff. r=kats (49d55c05f4) - Bug 1158424 - Extract a BreadthFirstSearch() helper function. r=kats (6a49498185) - Bug 1158424 - Add APZCTreeManager::FindRootContentApzcForLayersId(). r=kats (d6c7ecf781) - Bug 1158424 - Fix APZCTreeManager::GetMultitouchTarget(). r=kats (bf8571a4ca) - Bug 1174205 - Make sure to acquire the tree lock before finding the multitouch target. r=botond (bdbd98f2c7) - Bug 1211660 - Remove dead obj argument from two functions. r=Waldo. (efba5a2f8d) - Bug 1158424 - Undeprecate HasNoParentWithSameLayersId(). r=kats (d994ea28f8) - Fix AsyncCompositionManager not comparing layer tree IDs when finding scrollbar targets. (bug 1151169, r=kats) (c7dbc4776a) - Bug 1163259. Make sure we don't look at the parent ref layer when looking for a layer with the frame metrics for a scrollbar. r=kats (8b99d2c830) - Bug 1164340. Convert the scrollbar thumb ratio to a unitless ratio properly for non-zoomable scrollframes. r=botond (27f56cab56) - Bug 1164340. Fix up some scrollbar related comments in AsyncCompositionManager. r=botond (b9d540bea9) - Bug 1164767. Modify async scrollbar positioning code to deal with containerless root scrollables. r=botond (501d07592e) - Bug 1158424 - Remove FrameMetrics::IsRootScrollable() (it just duplicated IsRootContent()). r=kats (672fd034b3) - Bug 1120683 - Properly handle unapplying 3D projective transforms throughout APZ code. r=botond (a46612811a) - Bug 1148868 - Make APZ wheel handling handle pixel scrolling on OS X. r=dvander, r=kats (da3bd34135) - Bug 1193062 - Add fields to PanGestureInput and ScrollWheelInput. r=kats (716ba34e18) - Bug 1189565 - Send resize events when backing scale factor changes. r=mstange (4e81a7bbf7) - Bug 1193062 - Add nsCocoaUtils::ModifiersForEvent. r=smichaud (da67b22ff7) - Bug 1016035 - Remove the ability to swipe vertically. r=kats (fa315e0b97) - Bug 1016035 - Refactor the code that decides whether we want to start a swipe. r=kats (3091961328) - Bug 1016035 - Make the threshold in AxisPhysicsMSDModel::IsFinished controllable by the caller. r=kip (76df4d85a2) - Bug 1016035 - More swipe refactoring. r=kats (2abcd05b33) - Bug 1016035 - Implement the swipe animation ourselves instead of calling the NSEvent trackSwipe API. r=kats (4842b2a728) - Bug 1016035 - Move swipe tracking code into DispatchAPZWheelInputEvent. r=kats (bb1052b87c) - Bug 1016035 - Also mark widgetWheelEvents that are handled by APZ with mCanStartSwipe. r=kats (21809b902d) - Bug 1016035 - Put PanGestureInput events into a queue if they can end up as a swipe. r=kats (2ebac59047) - Bug 1016035 - Split up MaybeTrackScrollEventAsSwipe into SendMayStartSwipe and TrackScrollEventAsSwipe. r=kats (ced10d9671) - Bug 1016035 - Send MayStartSwipe event before sending the wheel event. r=kats (b3346b81fb) - Fix some fields missing in the widget event IPC glue. (bug 1139711, r=kats) (1298b25705) - Bug 1016035 - Add mCanTriggerSwipe and TriggersSwipe(). r=masayuki (664a059972) - Bug 1016035 - Make APZEventState report defaultPrevented=true when the event started a swipe. r=kats (128d627ac7) - Bug 1016035 - Set scroll overflow information on potential swipe start events that have been processed by APZ. r=masayuki (b5931f6320) - Bug 1206801 - fix broken CONFIG['DEBUG'] checks in moz.build files; r=mshal (292d5ac709) - Bug 1186424: Refactor AsmJSValidate.cpp in preparation for parallelization; r=luke (4868bc66ca) - Bug 1129267 - Fix by backing out patch for bug 1118615. r=mstange (ea8aaba04e) - Bug 1182411 Part 1: Make plugin quirks available to the Parent as well as the Child. r=jimm (1f2551a6b7) - Bug 1228462 - "PluginQuirks.* uses DOS CRLF". r=jmathies (b569a6c6c8) - Bug 1152300 - When focus shifts from a native plugin window to content, content needs to forward the request for focus change to chrome. r=enn (5bd811fadd) - Bug 1185529: On Windows, ensure that NPAPI child window has the correct parent before setwindow is called. r=jimm (9d94fdf2ff) - Bug 1182411 Part 2: Change winless popup surrogate to have its parent set in the chrome process. r=jimm (adee9bb012) (e9ba40cf6) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1144112 - Rearrange code a little to simplify it. No functional changes intended. r=botond (585aec1682) - Bug 1144112 - When tagging a block as fast-motion with touch-action enabled, make sure to set the allowed behaviors on it as well. r=botond (6c969faab6) - Bug 1172541 - Don't skip touch event delivery during smooth scroll. r=kats (c1e81e308e) - Bug 1194876 - Extract a TouchCounter class. r=botond (742458342a) - Bug 1159455 - Fix assorted debugging strings in APZ code. r=botond (4f608f0e0a) - Bug 1194876. Reset the input state on an APZC if a new touch block targets a new APZC while there were touches still active. r=botond (481002eaa9) - Bug 1194876 - Do a better job of resetting input state. r=botond (9fc5a3a605) - Bug 1143575. Fix build by ensuring ColorLayerComposite uses namespace gfx. r=nical (429d9f9abe) (16a3a03bc) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1193062 - Add PanGestureBlockState. r=kats (a8da88a934) - Bug 1193062 - Make PanGestureInput transform processing work like ScrollWheelInput processing. r=kats (fb17c120e8) - Bug 1180780 - Enable handling of pixel-scrolling wheel events with APZ on Windows and Linux. r=mstange (41d8fd5ed2) - missing part of Bug 1013412 (4fa14a56d9) - Make sure that confirmed APZ wheel targets are always scrollable. (bug 1154132, r=kats) (c35e3a34ca) - breakage fixes (96a1e3c36a) - Bug 1153711 - Do not discard a component of a fling if an APZC further in the handoff chain has room to scroll in that direction. r=kats (4f35dfd913) - Bug 1193062 - Set correct axis velocities when using PanGestureInput events. r=kats (7036040c68) - Clamp APZ wheel animations to their scrollable range. (bug 1152011 part 2, r=botond) (45c4925fee) - Bug 1193062 - CanScrollWithWheel needs to use ParentLayerCoords for the scroll delta. r=kats (20a3c4ac6b) - Bug 1193062 - Make OverscrollHandoffChain::FindFirstScrollable and AsyncPanZoomController::CanScroll able to deal with PanGestureInput events. r=kats (d87ec270db) - Bug 1193062 - Add mHandledByAPZ on PanGestureInput and ScrollWheelInput, and sync the information to the WidgetWheelEvent. r=kats (6e72f5885f) - Bug 1016035 - Delay the processing of a PanGestureInput block until we know whether it's a swipe. r=kats (77d1dceb04) - Bug 1145084 - Ensure that the 300ms content timeout in the APZ code also applies to the touch-action allowed behaviours. r=botond (45f58fe247) - Bug 1193062 - Make AllowScrollHandoff work for both ScrollWheelInput and PanGestureInput blocks. r=kats (d335008b0a) - Bug 1193062 - Remove mPanGestureState. r=kats (774bb3dedf) - Bug 1193062 - Use ScrollSource::Wheel for pan gesture events. r=kats (e9328b0bc4) - Bug 1193062 - Don't use PanGestureInput events for instant wheel scrolling. r=kats (7d89c5dc2e) - Bug 1175564 - don't lock y-axis if pan displacement is 0. r=kats (372d76c914) - Bug 1193062 - Process pan gesture deltas in begin+end events. r=kats (f29072bf8f) - Bug 1193062 - Fix UntransformVector w coordinate checks. r=kip (73b459e459) - Bug 1016035 - Replay the queue to the swipe tracker once the swipe start confirmation arrives. r=kats (9891444d4f) - Bug 1016035 - Don't wait for content to say that we need to swipe if APZ has enough information. r=kats (2b94c25bd3) - Bug 1016035 - Swallow the rest of the scroll gesture after swiping without APZ. r=kats (f9250388f9) - Fix crashes when a wheel transaction has a null confirmed apzc. (bug 1147249, r=kats) (ef898a9f3a) - Bug 1176001 - Turn unconditional log into a manually-enabled log. r=botond (1d866b1ffa) - Don't have a default repaint interval of "never" for APZ animations. (bug 1152138, r=botond) (7e276ba862) - Bug 1152051 - Ensure that destroying an overscroll animation always clears the state Axis tracks about it. r=Cwiiis (1273aac828) - Request content repaints during APZ wheel animations. (bug 1152138, r=botond) (ba3e24011b) - Bug 1164557 - Use COORDINATE_EPSILON in IsZero(). r=kats (c07e506607) - Bug 1164557 - Weed out spurious calls to Axis::OverscrollBy() caused by rounding error r=kats (15f8d18a7e) - Bug 1186164 - When clearing the overscroll animation state, make sure# etOverscroll() continues to reflect the correct direction of overscroll. r=Cwiiis (282c63383c) - Bug 1198900 - Don't call OnTouchCancel when the current block is not a touch block. r=botond (b724d8d3ef) - Bug 788873 - Only rebind a GLXPixmap if the texture has changed. r=karlt (e1f72858ee) - Bug 1169270 - Minor style fixes. r=dvander (494449170d) - Bug 1145702 - Make the assertion in Axis::GetOverscroll() more informative. r=Cwiiis (dff257bbbd) - Bug 1152051 - During an overscroll animation, detect a peak even if a sample has a velocity of exaclty zero. r=Cwiiis (a5aad5349f) - Bug 1201101 - Enable axis-locking over multiple APZCs. r=botond (21574b80ec) - test of Bug 1143522 - Convert Layer::mClipRect to ParentLayerIntRect. r=botond (8f729ca8b0) - Bug 1159398 - Set the controller thread when running APZ gtests. r=dvander (4349b6722b) - Bug 1148350 - Add a test. r=botond (72e3c0716d) - Bug 1152051 - Gtest. r=Cwiiis,kats (bf337c0d6d) - missing of Bug 1158122 - Remove most occurences of nsIntRect in gfx/tests. r=nical (a328773d9d) - missing test of Bug 1160566 - Make FrameMetrics.mCompositionBounds private (091db42632) - Bug 1163845 - Unify handling of time in APZ gtests. r=kats (2366690fe7) - Fix APZ scroll wheel animations often prematurely ending on the first frame. (bug 1157409, r=kats) (78cc0e2292) - Bug 1169689 - Make sure we don't end up with a spurious velocity if we're not going to be panning. r=botond (8911802134) - Bug 1203744 - Fix the SCREEN blend mode with CompositorOGL when the source data is unpremultiplied. r=dvander (e256f83a14) (d88cb9e20) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1157984 - Part 1: Extend gfx::2d classes to support both float and double precision,r=jrmuizel (d8d0cadcb0) - Bug 1157984 - Part 2: Implement double precision clipping functions in Matrix4x4,r=vlad (f31c243a9d) - Bug 1157984 - Part 3: Correct bounding box transformations to support projections and correct clipping when transforming behind the camera,r=vlad (c9c678905b) - Bug 1157984 - Part 4: Remove gfxRect::TransformBounds,r=vlad (71d076d48b) - Bug 1157984 - Part 5: Test,r=vlad (1c9c711745) - Bug 1159985 - In APZ gtests, allow panning in both directions. r=kats (b4a42c580a) - Bug 1166871 - Add a test. r=botond (b08332347f) - Bug 1159985 - Gtest. r=kats (ad41b809ff) - Bug 1169695 - Make the test timestamp privately held by the MockContentControllerDelayed. r=botond (4f898a778c) - Bug 1169695 - Advance the stored time when sampling the transform for a future frame. r=botond (77d029383e) - Bug 1169695 - Store task run-at times in the task queue. r=botond (4bc228d713) - Bug 1171312 - Add generic tree search algorithms for layers-related data structures. r=botond (675806f3e1) - Bug 1200063 - Make APZCTreeManager the place where GetFrameTime() can be overridden. r=kats (32ba198192) - Bug 1200063 - Share a paint throttler between APZCs in the same layers id. r=kats (a2147d25e6) - Bug 1200063 - Add a generic implementation of nsITimerCallback that's usable with a lambda or other function object. r=kats (6d7ef71d44) - Bug 1200063 - Make sure TaskThrottler sends its pending repaint request eventually, even if it never receives a notification from the previous request. r=kats (6eceede1e2) - Bug 1200063 - Rename MakeAPZCInstance to NewAPZCInstance for consistency. r=kats (6927e90644) - Bug 1144374 - Fix APZ code to deal with layers getting moved to another LayerManager. r=botond (eb682a58cd) - Bug 1193930 - Allow the user to start a pinch while an overscroll animation is still in progress. r=kats (44c2e57123) - Bug 1169690 - Gtests. r=kats (3c206d1cb2) - Bug 1201098 - Consume fling and overscroll velocity per-axis. r=botond (1e05677586) - Bug 1158424 - Clean up uses of HasNoParentWithSameLayersId() in overscroll handoff chain building. r=kats (0ed1ddee59) - Bug 1201277 - Replace functors with lambdas in APZCTreeManager. r=botond (ec5476a598) - Bug 1180030 - Request a fling snap when an overscroll animation completes. r=kats (e789a2ebb3) - Bug 1159985 - If only one component of a fling is in overscroll, continue the fling in the other component. r=kats (81998d723f) - Bug 1031443 - Allow panning in the pinching state even if zooming is disabled;r=botond (4c90350cc6) - Fix APZ not starting and updating wheel animations within an APZC lock. (bug 1152011 part 1, r=botond) (ae0d4d0aa8) - Bug 1145089 - Update documentation for nsEventStatus_eConsumeNoDefault. r=botond (7ca605ad96) - bug 1170988 - Fix skia to build for iOS. r=gw280 (fb375def29) - bug 1171642 - Disable some skia arm assembly for iOS. r=gw280 (004245c173) - fix build on gcc 11 (1c511ac81b) - missing bit of Bug 1188462 - Rename gfx/skia/trunk directory to gfx/skia/skia. (ca8f138b28) - Bug 1180030 - Do not request a fling snap if the fling will overscroll on its first sample. r=kats (55f89d55b5) - Bug 1167721 - Ensure we trigger a repaint when dropping velocity to zero in CancelAnimation. r=botond (d04fbb2a50) - Bug 1158340 - Refactor RedistributeDisplayPortExcess. r=kats (27725c3ee4) - Bug 1189565 - Only factor in the async zoom change. r=kats (57e357a038) - Bug 1192466 - Prune header includes for SkiaGLGlue.h - r=bgirard (37d31a1238) - Bug 1004489 - propagate Cairo font AA settings to Skia font. r=gw280 (9adc439b19) - Bug 1171528 - Remove overflowed nscoord_MAX warnings from nsRect. r=dholbert (08599b00e9) - Bug 1202430 - remove NS_GFX and related macros; r=jrmuizel (cb0b79951e) - Bug 1181317 - Switch drawing order of filter feBlend inputs. r=mstange (4914444219) - Bug 1180379. Split out GMAX4500 and block those devices from D2D. r=Bas (c8d1a755ea) - Bug 1073117 - Theme issues with GTK 3.14 - fix gtk button and entry size, r=karlt (0df9ac5c7f) - Bug 1187237 - Correct range slider drawing on GTK3. r=karlt (0d671487da) - Bug 1171696 - Require room for scrollbar sliders and draw troughs only when there is room for a slider or buttons on GTK. r=karlt (71710baeaa) - Bug 1202317. Optimize away identity ColorMatrix nodes. r=mstange (6c57a71e0c) - Bug 1202317. Support PodEqual over fixed-length arrays. r=Waldo (060f20b429) - Bug 1203078 - whitelist gdk-pixbuf image formats. r=dan (5eb43e4235) - Bug 1202317. Optimize feComposite type="arithmetic" with all-zero k values to transparent black. r=mstange (47b2d40678) - Bug 1189443 - Don't round down the margin amounts when inflating the displayport. r=dvander (4afcf087ab) - Bug 1205741 - add xpm and svg to supported gdk-pixbuf formats. r=acomminos,dveditz (4c356c961a) - Bug 1166741 - Workaround for GTK3 bug where file choosers' delegate refcount is not incremented. r=karlt (9d59be7b05) - Bug 1166741 followup: Guard GTK file-picker member-var decl/init statements with same #ifdef that its usages are guarded with, to avoid -Wunused-private-field warning. r=karlt (25bbf34486) - fix build on 10.5/10.6 (9acd104c9f) - namespaces (358de9591c) - Bug 1178842 - Fix unified build for nsDisplayList.cpp when a new file is added to layout/base. r=mstange (aa8ff77615) (a988f766e)
    1 point
  28. as I said before, you have to find a version that is working well in seamonkey first. EDIT: by the way I tried to modify gm-3.31.4 for SM/IA/BN support, but not well tested. you may try this yourself. http://o.rthost.win/boc-uxp/greasemonkey-3.31.4-pm_sm_ia_bn.xpi
    1 point
  29. New regular/weekly KM-Goanna release: https://o.rthost.win/kmeleon/KM76.4.5-Goanna-20220129.7z Changelog: Out-of-tree changes: * update Goanna3 to git f71f51550...d8ae51153: - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1153444 - Fix up Key Pinning Telemetry (r=keeler) (8a136aeb6f) (partly) - Bug 1183065 - Add logging on OneCRL revocation checks (r=Cykesiopka) (d8dbd936f3) - bug 1170303 - treat malformed name information in certificates as a domain name mismatch r=Cykesiopka (733179123e) - Bug 1193021 - clean up reference-counting in security/; r=keeler (df4b26974f) - Bug 602122 - Add a static analysis to find XPCOM classes with duplicate mRefCnt members; r=ehsan (02db112583) - Bug 1180506 - Fix building the clang-plugin with ToT clang; r=mystor (c491ab3b27) - Bug 1180549 - Fix a startup crash when using the clang-plugin with clang>3.5; r=mystor (9ef5796d3f) - Bug 1183485 - Exempt security/sandbox/chromium from MOZ_IMPLICIT enforcement. r=ehsan (39aa87ce29) - Bug 1185188 - Support multiple clang-plugin annotations; r=ehsan (dba7c0697a) - Bug 1185044 - Unify type annotation logic between MOZ_*_CLASS and MOZ_MUST_USE; r=ehsan (425e42c834) - Bug 1187486 - Update the clang plugin to detect templated implicit constructors; r=ehsan (fa22b9627f) - Fix some bad variable names, no bug (79f2eea2b1) - Bug 1187073 - Add the MOZ_INHERIT_TYPE_ANNOTATIONS_FROM_TEMPLATE_ARGS annotation to clang-plugin. r=ehsan (c6c629a7bf) - Bug 1192015 - Refactor allocation analysis in clang plugin to better allow adding new analyses, and support detecting temporary value allocations, r=ehsan (5d56ce510b) - Bug 1192130 - Part 1: Add MOZ_NON_AUTOABLE to restrict using auto in place of certain types, r=ehsan (d2b9712180) - missing part of Bug 1192130 - Part 2: Use MOZ_NON_AUTOABLE to validate the usage of already_AddRefed, r=ehsan (d66417ecb6) - Bug 975246 - Part1 - Implement operator->* in nsRefPtr. r=nfroyd, r=gerald (299c514e39) - Bug 975246 - Part2- Test Invoking via operator->*. r=nfroyd (197e539bfd) - Bug 975246 - Part3 - Using nsRefPtr->* operator. r=nfroyd (55010f772e) - Bug 1056356 - Add support for nsRefPtr<const T>. r=froydnj (d11d8183b8) - fix token to adhere to bug 1135708 (056463d154) (f1e094af8) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1194022 - Add another exception for a token after yield expression without operand and followed by EOL. r=Waldo (6f6806dd55) - Bug 1193777 - Remove pn==null case from ModuleCompiler::fail(). r=Waldo (ad7a00bc9e) - fix debug code (185d538d36) - Bug 1186723 - Make BytecodeEmitter::emitDupAt take a reasonable offset r=arai (c75ba4dac2) - fix one more assert (6572298e7b) - Bug 1149298 - Part 1: Test cases. r=drno (66301726be) - Bug 1149298 - Part 2: When destroying a candidate, ensure that the ice_ctx doesn't continue waiting for it to init. r=drno (62b15139c6) - missing part of Bug 1149298 - Part 3: Fire end of candidates signal when StartGathering has nothing to do, and only call StartGathering once per offer/answer. r=drno (aecf2da75f) - Bug 1143526 - DataStore API should create DOMError with proper error messages, r=fabrice (f7971e9039) - Bug 1149784 - Make nsSubDocumentFrame::GetMarginAttributes use CSSIntSize instead of the unit-less nsIntSize. r=roc (8dd64a1d9e) - missing part of 1022064 (c96b8add9a) - Bug 1138590 - Create a WebChannel for receiving FxA profile change notifications. r=markh (49a2234a21) - Bug 1139657 - Expose a method on FxAccounts for retrieving profile information. r=markh (2e44487ef5) (4d3fc3ae7) - import changes from `dev' branch of rmottola/Arctic-Fox: - No bug. Minor analysis fix for duplicate avoidance logic. Does not seem to matter in practice. r=woof! (57c242e0b8) - pointer style (adeb45154d) - Bug 1167385 - Mark xpcObjectHelper's unsafe references as UNSAFE_REF. r=bholley (2023354f74) - Bug 1185056 - Use the normal CC macros for XPCWrappedNative. r=smaug (7df93c9f0f) - Bug 1141079 - Make jemalloc_config.c a C++ source file. r=ted (d784a65774) - Bug 1141079 - Setup custom jemalloc chunk hooks to keep RSS usage low. r=njn (227f7527c4) - Bug 1191815 - move jemalloc CFLAGS definition to moz.build; r=mshal (fffe74592e) - Bug 1187776 - Add ICU files to B2G desktop. r=glandium (40074ca68c) - Bug 1192277 - remove check for LLVM PR 8927; r=glandium (d12cfe5f28) - Bug 762449 - Make jemalloc's opt.lg_dirty_mult work as documented. r=njn (2765252e9c) - Bug 1138705 - Preserve LastError when calling TlsGetValue. r=njn (403eee3ca3) - Bug 1139357 - Make jemalloc3's --without-export actually work. r=njn (85f53039af) - Bug 1142412 - Update memory/jemalloc to 4.0.0 release. r=njn (bb65146083) - Bug 1142412 - Remove last jemalloc patch by using a workaround when running its configure. r=mshal (990b6cb4ea) - Bug 1201792 - s/MOZ_JEMALLOC3/MOZ_JEMALLOC4/. r=njn (3924d18006) - Bug 1202523 - Use the default chunk size with jemalloc 4. r=njn (268f4d5a7d) - Bug 1174250: Allow the CommitHook to fail without crashing. r=glandium (7cd2b5374e) - Bug 1203840 - Trigger dirty pages purge after CC. r=njn,r=smaug,r=mccr8 (5649197ba0) - Bug 1188469 - remove configure.in workaround for broken egrep in OS X 10.7; r=mshal (91eb406603) (c7924fe23) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1144934 - Add support for NSIS 3.0b1 in configure.in. r=gps (68c2d307ea) - Bug 1146096: configure.in updates for SCTP and musl.libc (sys/queue.h) r=glandium (1ee3be321a) - Bug 1139020: Update SCTP library from upstream to rev 9168 rs=jesup (59aff8e427) - Bug 1139020: rollup of sctp modifications for Android rs=jesup (0c207e42fa) - Bug 1146096: update usrsctp to rev 9209 from upstream rs=jesup (d48529adb5) - Bug 1146096: rollup of sctp modifications for Android rs=jesup (72296c9831) - (configure) Update embedded libpng to version 1.6.28 (9c91526374) - Bug 1138845 - Don't require atk-bridge for gtk+3 builds. r=tbsaunde (b3954d62cc) - Bug 1159273 - Allow loading of tree cairo and system cairo at the same time. r=glandium (4f1a58a49f) - Bug 1138845 - Fixup for bustage. r=me (dff1a71582) - Bug 1159273 - Update Cairo/Pixman renaming to not leak the symbols we use and silence linker warnings. r=jrmuizel (77daf01890) - Bug 1159273 - Update gkmedias symbol defs to catch up with the Cairo/Pixman renames. r=jrmuizel (c69fece38b) - Bug 1159273 - Fix out-dated Cairo usage that may fail to build or work properly when system Cairo is involved. r=jrmuizel (352f69f8f5) - Bug 1159273 - call g_type_init in nsAppShell::Init. r=glandium (61e1857a64) (d8ae51153) * 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
    1 point
  30. I downloaded this once more, eager to see if it would actually work, testing with vlc 0.9.8a and 1.0.5. For 0.9.8a, I renames the file to shell 32.vlc and had to hexedit it in libvlccore.dll. For version 1.0.5, I had to hexedit vlc.exe. I was amazed that they actually load and work. VLC 0.9.8a loads, the menu and dialogs work, it plays music and video, the basic controls (play, pause, fast forward, stop, volume), playlists and the menu and equalizer work. I haven't tested CD's and DVD's yet. I can't seem to get my physical CD Drive to work in NT 4.0 under Virtualbox (ISO's work), although it clearly lists options to use the CD drives in a VM. On the other hand, VLC 1.0.5 is so broken that it is almost unusable. The menu doesn't work, and consequently, the dialogs don't work. The only way to open a file for playing is by right-clicking it and selecting "Play with VLC Media player." Audio seems to play, and I don't know if there is supposed to be a visualiser or not, but if there is, it isn't there. Just a black screen. When I open a video file for testing, it insists in opening the video in a command prompt window, and it is all covered by DOS characters as it plays. The equaliser works and I'm not sure about playlists, Cd's and DVD's.
    1 point
  31. If I may, the issue you just pointed out (ordinal vs. name) is only one of the causes of the so called "DLL hell", but it cannot be underestimated how a large part of it is to be attributed to sloth and "unneeded complexity" (and in some cases plain stupidity) on behalf of the actual programmers. Some reference: http://www.codeproject.com/Articles/4896/The-DLL-Hell-Problems-and-Solutions Please note how this has been removed: http://msdn.microsoft.com/en-us/library/ms811694.aspx but we can have it through the WayBack Machine: https://web.archive.org/web/20100305054645/http://msdn.microsoft.com/en-us/library/ms811694.aspx Part of it is due to "abuse of DLL's" however. What I mean is, if you have to make 2+2 you normally have a mental table for it and you know that it makes up 4. You do not search for an algorithm capable of doing addition on integers, real, fractional, and imaginary numbers with precision up to 2^31 and 25 decimal places. What a number of programmers abuse is the use of such external, often overcomplicated (because they are meant for some non-trivial scope) code when there is in the programming language already the provisions to make a smaller, simpler, more suitable and faster algorithm. The use of some "programming studios" is another culprit, when examining a program for dependencies (let's say with Dependency Walker) you will find how there are often completely unneeded dependencies to functions that also exist in another .dll which is actually needed or dependencies created "automagically" by the CASE tool that have no real meaning/use. What is IMHO absurd is that the good MS guys (who have had "full control" on them since the very beginning) in all these years did not manage to keep at least the so-called "known" DLL's and the small bunch of VCC redistributables under control and their way out (supposed "solution" in their perverted minds) is the total folly that Windows Side-by-Side (WinSxS) "assemblies" represents (including their crazy folder/file names). jaclaz
    1 point
  32. I can't speak for him but for 100% compatibility two criteria must be met - Function Names and their Ordinals. The latter case - out of order functions, is a biggy. In the past it was common for applications and even Windows to call functions in a DLL by importing them via ordinal. So if someone recompiled a DLL without changing a single line of code but re-sorting them out of original order an incompatibility is created because the 3rd ( or whatever ) function is completely different in another version of the same DLL. Having never used it myself ( so I don't know for sure ) I would expect that this topic is an important part of the KernelEx project here at MSFN. Naturally in the former case the function "Names" implies that everything else is the same within the function ( i.e., if it has same exact name then it's still ANSI/Unicode, same arguments, etc, as its same-named predecessor ), but even this might not always be the case since there is nothing stopping some programmer from changing ( for example ) from ANSI to Unicode but leaving the NAME unchanged. Likewise a function could be completely rewritten and butchered to use a completely different prototype while keeping the NAME unchanged. These are exceedingly bad ideas but cannot be ruled out. But for the sake of argument its *assumed* that same name means same function. How was this failure vector even possible? It was considered faster and more efficient to call by ordinal, in other words cutting corners. But speed and efficiency isn't everything. IMHO consistency, stability and compatibility are. But that ship has clearly sailed!
    1 point
×
×
  • Create New...