Leaderboard

thats the thing I want to do, by backporting their changes to my not-rearranged tree.

Hey everyone, could we try and keep the double posting down to the minimum? If you have extra to add to your post, please use the edit button by clicking on the three dots on the right hand side of your post. And you can use the + icon next to the quote icon if you want to reply to several posts at once. It helps keep the thread a bit neater in appearance and easier to navigate. Thanks!

Welcome! It's great to have you here, enjoy your stay!

Logically I would do that in this way. First all SP3 updates, then POSReady reg file and finally all POSReady updates.

this time，succeed，thanks！

Hello, I want to recreate the Microsoft Windows Update v3/v4/v5 sites in local with all the elements. I know the Wayback Machine Downloader to do this but I can't find any pages of this site on web.archive.org. So, can you help me doing that project? Note: this is a public project, anyone can participate, but pls, don't share other Windows Update sites than v3, v4, v5 or v6.

New NewMoon 27 Build! 32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20220319-36b2354fd-xpmod.7z 32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20220319-36b2354fd-xpmod-sse.7z 32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20220319-36b2354fd-xpmod-ia32.7z 64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20220319-36b2354fd-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 1181382: move declaration into namespace to resolve conflict. r=hsinyi (b1cbc67344) - Bug 1131470 - Part 3: Implement ScreenOrientation interface. r=baku (805ffd4d18) - Bug 1131470 - Part 4: Check sandboxing flag for orientation lock. r=baku (2896c9af4c) - Bug 1131470 - Part 5: Handle changes to active orientation lock. r=smaug (2550433007) - Bug 1131470 - Part 6: ScreenOrientation tests. r=baku (53e088ab98) - Bug 1193133 - Drop the document and window references from ServiceWorker. r=bkelly (ce09f526b8) - Bug 1186307 - part 1 - Unify the StructuredCloneCallbacks in WorkerPrivate.cpp, r=smaug (a109da831d) - pointer style (c26fa45b91) - Bug 1174971 - Introduce two variants of getOwnPropertyFromTargetIfSafe. r=gabor,r=arai (4bec0627ce) - reapply Bug 1131802 part 2. Allocate functions with reserved slots for DOM Xrays so we can store the Xray wrapper reference in those slots instead of as the function parent. r=peterv (ff39e220a3) - No bug - Add a bug number to the expected failure in message-event.html. (398f3e500a) - Bug 1164834: MessageEvents coming from postMessage should not be cancelable. r=Ms2ger (380ce180bf) (38ceaed34) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1176288 - Part 4: Fix octane regression, r=bbouvier (3ae04863c9) - Bug 1199417- Scale skia paths up before hit testing to workaround coordinate resolution limits. r=mattwoodrow (b3ec1288e8) - Bug 1117338 - Blow up if we can't convert a cairo format to a gfx one r=milan (be268bf93a) - Bug 1161642 - Cross compilation fixup. (58da4778e4) - Bug 1190705 - Ensure that canvas 2d matrix transforms are finite. r=Bas, r=jrmuizel (ec3a6a510e) - Bug 1204824. Make Matrix4x4::PreScale scale all values in the first three rows. r=bas (457cc10f80) - namespace (a341cb46be) - Bug 1162726 - Make RoundedToInt() snap the rect properly. r=roc (7fd0a19105) - Bug 1178965 - Check if Factory::GetD2D1Device fails in SourceSurfaceD2D1. r=nical (cad95d7703) - Bug 1163735 - Ensure that we don't hand out a destroyed TabChild from WorkerPrivate::InterfaceRequestor (r=khuey) (270cf6ae33) - Bug 1198451 - Disambiguate mozilla::dom::Function from mozilla::Function. r=froydnj (d08622cbd4)- reorder (eea96d09ee) - reorder (eea96d09ee) - Bug 1147990 - websocket test suite - patch 2 - webSocket.url should be the spec of nsIURI, r=smaug (a983655514) - Bug 1147990 - websocket test suite - patch 1 - URL with #ref should not be accepted, r=smaug (7af5e9528d) - Bug 1147990 - websocket test suite - patch 3 - Not all the ports are accepted, r=smaug (a34d69e29b) - Bug 1147990 - websocket test suite - patch 4 - web-platform-test suite fixed, a=tomcat CLOSED TREE (f2966f52d0) - Bug 1147990 - websocket test suite - patch 5 - web-platform-test suite - js error fixed, r=jgraham (1b52d4853e) (0ca97f19b) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1179242 - Avoid unnecessary moves to dead stack slots, tweak regalloc heuristics for bundles required to be in a specific register, r=sunfish. (aa0174e817) - No bug - Fix bug1106719.js jit-test to allow overrecursion on a CLOSED TREE. r=me (5cc971608d) - Bug 1186424: Rename FunctionBuilder into FunctionValidator; r=luke (8925748a65) - Bug 1192401 - Do not capture SavedFrame stacks before Object.prototype has been initialized. r=shu (e26c24664b) - Bug 1194838 - Make Cell a GCType for the hazard analysis, r=terrence (92f429798f) - Bug 1194829 - Rename sweep to sweepChildren because shu said so, r=shu (0a80d997ae) - Bug 1183487 - Fix sorting in check_macroassembler_style.py. r=nbp (659d3b5dd3) - pointer and spacing style (9e2b14652c) - Bug 1186693 - Add exhaustive matching to mozilla::Variant; r=Waldo (ec1defe3c4) - Bug 1186650 - Do not mark mozilla::Variant copy and move constructors as explicit. r=Waldo (76bb7e8051) - Bug 1195297 - Fix OOM handling in ExecutableAllocator r=terrence (f9b9727ac8) - Bug 1195208 - Fix ArrayBuffer.transfer isNeutered check. r=luke (d45393ca24) - Bug 836658 - Use -arch:IA32 on VS2012 to avoid requiring SSE2. r=glandium Bug 1193467: Allow SSE2 on clang-cl builds. r=glandium (7c6fd7a785) - Bug 1183195 - Fix assertion at the start of InnerViewTable::sweepAfterMinorGC() r=sfink (df395675e3) - Bug 1185959 - Properly handle static constructor method in classes. (r=Waldo) (8a6919155f) - Bug 1180054 - Fix bogus assert. (r=jandem) (94b9988568) - Bug 1197377 - Read callgraph.txt line by line to avoid string size limits, r=terrence (6b2d110d65) - Bug 1195588: IonMonkey - Actually convert number to double during recovering ToDouble, r=nbp (de43a2e4b9) (a06231e9d) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1186307 - StructuredCloneHelper in workers.postMessage(), r=smaug (0d3c817afa) - Bug 1183824 - Remove NS_ENSURE_TRUE(mCallback, NS_ERROR_NOT_INITIALIZED) warning from nsFrameMessageManager::DispatchAsyncMessageInternal. r=smaug (64dfd50f7f) - Bug 1167189: Use a size annotation on the OOM abort. r=bholley (5e7fb331a0) - missing crash reporter (47f183fab4) - Bug 1137151: Marked destructor of |nsClipboard| as protected, r=mwu (5655bf8862) - Bug 1194497 - Convert a few remaining PRUnichar to char16_t. r=roc (1554ffcec2) - Bug 952456 - Part 1: Implement gonk/nsClipboard for rich text and raw image. r=fabrice (419381f7af) - Bug 1198795 - ipc/StructuredCloneUtils should be merged with StructuredCloneHelper, r=smaug (02183262bd) - Bug 1201806 - part 1 - Remove OwningSerializedStructuredCloneBuffer and use StructuredCloneIPCHelper everywhere in IPC code, r=smaug (6dc69151b4) - Bug 1201806 - part 2 - nsStructuredCloneContainer should use StructuredCloneIPCHelper, r=smaug (b6ebaa8e95) - Bug 1202028 - Implement range based for loops for SavedFrame stacks. r=terrence (57c198910c) - missing bit of Bug 1106264 (fa67f8ab56) - Bug 1168806: Configurable I/O thread for socket IPC classes, r=kmachulis (213a8fb2b9) - Bug 1159209: Remove template parameters from |DataSocketIO|, r=kmachulis (0d658c5cf8) - Bug 1168806: Configurable consumer thread for socket IPC classes, r=kmachulis (ee054dd6b0) - Bug 1167207: Add |BluetoothDaemonConnectionConsumer|, r=shuang (b8b89290df) - Bug 1154281: Remove |UnixSocketConsumer| and its helpers, r=kmachulis (d40430b10f) - Bug 1166638: Add |StreamSocketConsumer|, r=kmachulis (59b5c1abf3) - Bug 1166638: Add |ListenSocketConsumer|, r=kmachulis (d119bcaa2a) - Bug 1168806: Use 'consumer thread' in socket interface, r=kmachulis (65d232d21c) - Bug 1170466: Inherit |ConnectionOrientedSocketIO| from |UnixSocketWatcher|, r=kmachulis (b91d0c624a) - Bug 1170466: Share socket I/O methods in |ConnectionOrientedSocketIO|, r=kmachulis (108fb6696e) - Bug 1170466: Implement accept and connect in |ConnectionOrientedSocketIO|, r=kmachulis (590b3b4617) - Bug 1172479: Replace |nsIThread| by |MessageLoop| in socket I/O code, r=kmachulis (618073dcef) - Bug 1203426 - Rename StructuredCloneIPCHelper to StructuredCloneData, r=smaug (dc31124113) - Bug 1167392 - Change nsFrameMessageManager::mParentManager to a smart pointer. r=ehsan (98d60c0398) - Bug 1186843 - Always cache delayed scripts in message manager (b0ab501959) - Bug 1189550 - In parent process, cpmm.initialProcessData should be same as ppmm.initialProcessData (r=smaug) (7a8ab62432) - Bug 1196057 - Use channel->Open2() in dom/base/nsFrameMessageManager.cpp (r=sicking) (a032d483d6) - missing parts of Bug 1132072 - Tab switch refactoring (r=mconley) (e6fbbb722a) - Bug 1175736 - Implement the iframe referrer attribute. r=ckerschb, r=bz (a5fc4e683b) - Bug 1171017: Rename |BluetoothDaemonConnectionConsumer| to |DaemonSocketConsumer|, r=shuang (0af5957609) - Bug 1170993: Manage socket buffers in sub-classes of |UnixSocketBuffer|, r=kmachulis (960de8be45) - Bug 1171017: Rename |BluetoothDaemonPDUConsumer| to |DaemonSocketIOConsumer|, r=shuang (eea1681ab8) - Bug 1171100 - Fix incorrect logic of handling Bluetooth Gatt Responses in daemon backend. r=tzimmermann (c385364d1e) - Bug 1171866 - Remove ReversedUuidToString; f=tzimmermann, r=jocelyn (9b58176c39) - Bug 1171017: Rename |BluetoothSocketPDU| to |DaemonSocketPDU|, r=shuang (ce1bbd9d4a) - Bug 1171017: Rename |BluetoothDaemonConnectionIO| to |DaemonSocketIO|, r=shuang (00f2a38220) - Bug 1171017: Rename |BluetoothDaemonConnection| to |DaemonSocket|, r=shuang (b3c3522719) - Bug 1171017: Move classes from ipc/bluetooth to ipc/hal, r=shuang (047714aee4) - Bug 1171994: Add |RilSocket| and |RilSocketConsumer|, r=htsai (276211bbb7) - Bug 1171994: Forward received RIL socket I/O via |WorkerCrossThreadDispatcher|, r=htsai (7825d1c837) - Bug 1185478: Add leak checks to socket I/O classes (40e2c01ce4) - Bug 1185478: Remove strong references from socket I/O class to socket classes (6fa94f34ff) - Bug 1163912: Set |Type| property of BluetoothDevice a initial value during LeScan; f=elin, r=jocelyn (bda66689ed) - Bug 1166215 - Cast 'BluetoothTypeOfDevice' to uint32_t before the BT signal is distributed to BluetoothAdapter. r=joliu (d38ad4e55f) - Bug 1170086 - Fix the parameter type of DispatchReplyError() in StartLeScan result handler. r=joliu (a24e23e02f) - Bug 1161956 - Fix crashes while using GATT client functions. r=jocelyn (cebfb8c700) - Bug 1171868 - Remove GeneratePathFromGattId(aId, aPath, aUuidStr); r=jocelyn (0eac291b3c) - Bug 1161945 - Fix empty characteristics and descriptors after searching services. f=elin, r=jocelyn (a0d816fee4) - Bug 1164387 - Use |btgatt_client_interface_t.get_device_type| to get the device type. r=jocelyn (c7b5c97a00) - Bug 1173139 - Reorder global creation on workers, r=khuey. (acb79ae54e) - Bug 1170071 - Part 1: Revise BluetoothGattInterface, result handler, and interface helper for daemon support of gatt server. r=shuang (0ad0c9244e) - Bug 1170071 - Part 2: Add helpers for Bluetooth daemon GATT server support. r=shuang (46a4062579) - Bug 1170071 - Part 3: Add gatt server support to GATT module for Bluetooth daemon. r=shuang (71d80aabea) - Bug 1170071 - Part 4: Add Gatt server interfaces for Bluetooth daemon. r=shuang (a556aaf048) - Bug 1187230: Convert Bluetooth to use daemon runnables, r=shuang (9be6501225) - Bug 1204382: Add #include for nsIInputStream.h to MediaParent.cpp, to fix a build error in --disable-webrtc builds. r=jib (8b20e934) (36b2354fd)

hate to admit it ... my box of chocolates contains (more than) one of those Whom was it that said life is a bowl of cherries?

Found @ https://www.freecodecamp.org/news/should-we-demand-the-latest-browser-version-d5c72f8c9ffb

I suppose I wondered if Roytam1 can continue making browsers from this base. Just curious.

Welcome, and I have the same interests as you

Welcome to MSFN - see ya around the forums ;)

Thanks, I'm trying it out locally, and yeah, I think the ?? and ?. stuff is really getting me, StackOverflow is still broken, but otherwise seems to be working at least as well as before. About those last two things that "can't" be polyfilled.... they CAN be transpiled! Take a look: https://javascript.info/polyfills Transpilers A transpiler is a special piece of software that translates source code to another source code. It can parse (“read and understand”) modern code and rewrite it using older syntax constructs, so that it’ll also work in outdated engines. E.g. JavaScript before year 2020 didn’t have the “nullish coalescing operator” ??. So, if a visitor uses an outdated browser, it may fail to understand the code like height = height ?? 100. A transpiler would analyze our code and rewrite height ?? 100 into (height !== undefined && height !== null) ? height : 100. So I think what we really need to do is have this extension auto-transpile this bleeding edge JS into "vanilla JS" that "any" normal browser can run. This, in theory, should solve not just the issue but probably a large number. My understanding is that transpiling something is something the *web developers* are supposed to do before packing their JS into the actual website source, so unlike a polyfill which runs in the browser itself. That said, I'm sure it's possible, but not sure how slow it would be if feasible. Thoughts, anyone?

@roytam1 : I've left NM28 behind me several months ago, now exclusively (from the browsers you are compiling) using Serpent 52.9.0 (32-bit) in this Vista SP2 x86 laptop; what is the future for Serpent 52 with the advent of GRE ? Could platform enhancements implemented in GRE be still backported to St52's UXP? And I don't know whether NM28 users are aware, but GRE fully removes the Pale Moon GUID (also used in NM27/NM28) and a restructure of MCP's add-ons site is currently in progress, will you also adopt this change for future NM28 versions? https://www.palemoon.org/releasenotes.shtml

Some times ago I made an iso image containing all updates of Office 2007 and 2010 by using WSUS Offline Update from www.wsusoffline.net. So I do not have to download all updates via MU if I want to install one of them.

I'm using a v69 (late 2018) and it works for 90% or so of everything I've thrown at it. Regarding the other 10%, and to be as blunt as possible, "nobody cares". I mean that towards the web masters running those 10% of web sites - "none of them care". As much as we (myself included) "live in the past", we are unrealistic to expect web masters to remain in the past. Web masters know what browsers are visiting their web site. They know that for every 10,000 visits, they have 1 using a four-year old web browser or a twenty-year old operating system. LIFE GOES ON if they "lose" that ONE visitor. Plain and simple.

I used Office 2010 on Vista by manually installing updates from WHDownloader, after that MU worked just like normal.

I guess I'll just uninstall Office 2010 and install Office 2007, who knows how long MU is gonna stay there and check for updates.

All these problems with Microsoft Office, especially endless searches of Microsoft Update, incompatible updates and so on, made me stop installing Microsoft Office packages on Windows XP a long time ago. I use Softmaker Office 2016 Portable with all available features. It's totally sufficient for me. If I had to use Microsoft Office in Windows XP, I would install Office 2007. Older software is often a better choice in an old operating system.

There are a huge number of post-SP2 updates for Office 2010. It will take a very long time for Microsoft Update to find them and install them. Leave it overnight and see how far it gets.

Not even that - but only the *latest* versions of those browsers! (Technically, I'm using "Chrome", but a 3 year old version). Maybe this will be a wakeup call to the WWW: stop using useless JS libraries you don't need and write all your damn code yourself. If you can't understand it, and you don't know what it does, and you don't know why you need, then don't use it!!! This has never failed to serve me well as a webmaster. And of course, none of my websites randomly break. jquery is pretty much the only JS library I ever use, and even then, only on pages that actually "need" JS. Everything else on top of that, I write if I need it. And if I don't need JS, I don't use it. It's not "simplistic", it's called "resilient progressive web development".

As long as their sites work in the latest versions of Edge and Google Chrome (and Safari for Macs), that's all the authors of a lot of sites care about now.

File complaints with all these websites. They need to know that blindly using breakage-causing JS code is NOT okay! I complained to GitHub and they reverted enough breakage that it now works in Chromium 70 again though New Moon 28 is still broken. Though it's a losing battle for sure... as of a few days ago, GitHub file uploads are now once again broken... it's like whack-a-mole, you polyfill a "bug" in their JS, and soon enough, a new one crops up again that breaks the whole damn site again. What we really need is some of extension to replace all the JS on websites with the JavaScript they used two years ago, pulling from the Internet Archive or something. That might actually work - scrape the JS for sites that are broken now, using old JS, and see if they start working again... Maybe we can get a list of changes going and replace the diffs we see.