Leaderboard

StartAllBack 3.9.16

New build of Serpent/UXP for XP! Test binary: Win32 https://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20251011-3219d2d-uxp-b709881389-xpmod.7z Win64 https://o.rthost.win/basilisk/basilisk52-g4.8.win64-git-20251011-3219d2d-uxp-b709881389-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-20251011-3219d2d-uxp-b709881389-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-20251011-d849524bd-uxp-b709881389-xpmod.7z Win32 IA32 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20251011-d849524bd-uxp-b709881389-xpmod-ia32.7z Win32 SSE https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20251011-d849524bd-uxp-b709881389-xpmod-sse.7z Win64 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win64-git-20251011-d849524bd-uxp-b709881389-xpmod.7z Win7+ x64 AVX2 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win64-git-20251011-d849524bd-uxp-b709881389-w7plus-avx2.7z No official UXP changes picked since my last build. No official Pale-Moon changes picked since my last build. No official Basilisk changes picked since my last build. My changes picked since my last build: - mailnews: fix OAuth2Providers path (16dc4c9168) - Add support for HTMLSlotElement.assignedElements() (#2) (169c949e49) - Merge pull request #37 from ClassicNick/Merge-20251005 (f24e68da7b) - style: color-mix: when either color resolves to NS_COLOR_CURRENTCOLOR, it will cause color value reevaluation and cause stack overflow. hack it to use parent color instead. (702d61e68e) - caps: hack - check `mCSP` RefCount to avoid circular calls (b709881389) 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.

New build of Serpent/UXP for XP! Test binary: Win32 https://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20250927-3219d2d-uxp-50b6047c84-xpmod.7z Win64 https://o.rthost.win/basilisk/basilisk52-g4.8.win64-git-20250927-3219d2d-uxp-50b6047c84-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-20250927-3219d2d-uxp-50b6047c84-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-20250927-d849524bd-uxp-50b6047c84-xpmod.7z Win32 IA32 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20250927-d849524bd-uxp-50b6047c84-xpmod-ia32.7z Win32 SSE https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20250927-d849524bd-uxp-50b6047c84-xpmod-sse.7z Win64 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win64-git-20250927-d849524bd-uxp-50b6047c84-xpmod.7z Win7+ x64 AVX2 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win64-git-20250927-d849524bd-uxp-50b6047c84-w7plus-avx2.7z Official UXP changes picked since my last build: - No issue - Clarify Azure Back-end info. (c7e103b06e) - Revert "[gfx] Guard against possible race via gfxFontEntry::GetFontTable." (060e2965b8) - [gfx] Clarify comments. (6e19656e11) - Issue #2852 - Parse full cache-control headers. (16164b3d38) - [gfx] Use std::size rather than hardcoding an array size (3a121f34c8) - Issue #2852 - Follow-up: Add comment and fix formatting. (1bcdad3ef1) - Issue #2790 - Follow-up: Ensure both text and background are set. (c9e08cb2f5) - Revert "Issue #2258 - Part 4: Clean up unused pointers." (7b4bb08257) - Revert "Issue #2258 - Part 3: Allow sniffing with XCTO:nosniff + empty MIME type." (2d4d48db72) - Revert "Issue #2258 - Part 2: Move XCTO:nosniff check into sniffers." (73c9c1dd73) - Revert "Issue #2258 - Part 1: Support XCTO:nosniff when navigating." (0d00c84033) - Issue #2258 - Make sure we favor sniffing when the content-type header is empty. (659a3c0fcc) 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.

change user agent compatibility to firefox compatibility in preferences dialog.

https://github.com/yt-dlp/yt-dlp/issues/14456#issuecomment-3339167951: --extractor-args "youtube:player-client=default,-tv_simply"

I made an update to yt-dlp for Vista, while also applying the diff's we discussed (_video.py changed a bit, due to an error). Furthermore, I found a PyInstaller 6.15.0 for Vista here (whl) and used that with success. (didn't work on XP...)

This may be of some use for Win XP users in today's world with Win XP being phased out more everyday ... this service is also listed to work with Windows 98 SE / ME also. I used it about 6 years ago myself and now am being forced to use it again in Oct 2025 to be able to get my Win XP computer on the internet with a VPN. The paid service I was using has finally cut off all Win XP connections. They warned two years ago it was coming and the service still works for my Windows 7 computer but I still mostly use my Win XP computer for everyday connecting. As I said above ... I used this free VPN about 6 years ago and I have hooked the XP computer up to the service again. I will say that they have really improved things and the service is great in 2025. Perhaps some of you know about it already but if not ... then you can check it out if you need a VPN for Win XP. https://www.vpngate.net/en/ https://www.vpngate.net/en/download.aspx VPN Gate - Public Free VPN Cloud by Univ of Tsukuba, Japan Welcome to VPN Gate. (Launched on March 8, 2013.) - You can get through your government's firewall to browse restricted websites. (e.g. YouTube.) - You can disguise your IP address to hide your identity while surfing the Internet. - You can protect yourself by utilizing the strong encryption while using public Wi-Fi. More Details... - TunnelCrack protection implemented in SoftEther VPN Client (August 31, 2023) Supports Windows, Mac, iPhone, iPad and Android. Compatible OS: Windows, CPU: Intel x86 and x64 (Windows 98 SE / ME / 2000 SP4 / XP SP2, SP3 / Server 2003 SP2 / Vista SP1, SP2 / Server 2008 SP1, SP2 / Hyper-V Server 2008 / 7 SP1 / Server 2008 R2 SP1 / Hyper-V Server 2008 R2 / 8 / 10 / Server 2012 / Hyper-V Server 2012) -------------------------------------------------------------------------------------------------- It looks complicated but it really isn't. Spend some time reading everything and you should be good to go. I'm impressed from using it 6 years ago to how it has improved. One word about connecting, if you need an US connection for banking, Tubi or whatever ... they are not always there, just at various times. You will figure all this out as you get familiar with the service. ...

I'm sorry. I have no idea... ... At least I got this figured out, as it was seriously bugging me ; the "what" is hidden inside a changelog: https://github.com/adang1345/PythonVista/blob/master/CHANGELOG.md Your v2025.09.23.052315 yt-dlp_x86_win7.exe binary has been compiled with a flavour of 3.13.7 which, at that time, supported only Win7+ ; Vista support for 3.13.7 came on Sep 25th, 2 days after v2025.09.23.052315 was built, but in time for the v2025.09.27.071342 to be built with a Vista-enabled 3.13.7 interpreter (this assumes you manually updated to the Vista-compat flavour of 3.13.7) ... What is still unclear to me is: a) were both versions built on the same OS (Win7 SP1 ?) ? b) were both versions packaged with "official" PyInstaller-6.16.0 ? v2025.09.27.071342: [debug] yt-dlp version nicolaasjan/yt-dlp@2025.09.27.071342 (win7_x86_exe*) [debug] Python 3.13.7 (CPython AMD64 32bit) - Windows-Vista-6.0.6003-SP2 (OpenSSL 3.0.16 11 Feb 2025) and v2025.10.05.061237 [debug] yt-dlp version nicolaasjan/yt-dlp@2025.10.05.061237 (win7_x86_exe*) [debug] Python 3.13.7 (CPython AMD64 32bit) - Windows-Vista-6.0.6003-SP2 (OpenSSL 3.0.16 11 Feb 2025) have been both compiled on a 64-bit OS, with the 32-bit build of Vista-compatible 3.13.7; why, then, is the next v2025.10.13.80600, built on a Vista-compatible 3.14.0, broken on actual Vista SP2 32-bit? I have a hunch, based on adang1345's notes: https://github.com/adang1345/PythonVista/blob/master/README.md So, the same 3.14.0 CPython will behave differently when run on a Win7 machine to when run on a Vista machine; likewise for the PyInstaller-packaged resultant yt-dlp_x86.exe binaries; but this theory doesn't explain builds v2025.09.27.071342+v2025.10.05.061237 (which run OK on Vista), unless they were also compiled on a Vista SP2 64-bit VM ... FWIW, adang1345 has made a new release of 3.13.9 less than an hour ago, https://github.com/adang1345/PythonVista/tree/master/3.13.9 and issued a re-release of 3.14.0 along with the above: https://github.com/adang1345/PythonVista/tree/master/3.14.0 His changelog only mentions the addition of 3.13.9, though ... This one launches OK: [debug] yt-dlp version local@2025.10.13 [eafedc218] (win_x86_exe) [debug] Python 3.14.0 (CPython AMD64 32bit) - Windows-Vista-6.0.6003-SP2 (OpenSSL 3.0.18 30 Sep 2025) Was a Vista SP2 64-bit VM used in this case? And which PyInst version? This one launches fine : yt-dlp_x86_win7.exe -v [debug] Command-line config: ['-v'] [debug] Encodings: locale cp1253, fs utf-8, pref cp1253, out cp1253 (No VT), error cp1253 (No VT), screen cp1253 (No VT) [debug] yt-dlp version nicolaasjan/yt-dlp@2025.10.13.080600 (win7_x86_exe*) [debug] Python 3.14.0 (CPython x86 32bit) - Windows-Vista-6.0.6003-SP2 (OpenSSL 3.0.18 30 Sep 2025) [debug] exe versions: none [debug] Optional libraries: Cryptodome-3.23.0, brotli-1.1.0, certifi-2025.10.05, mutagen-1.47.0, requests-2.32.5, sqlite3-3.50.4, urllib3-2.5.0, websockets-15.0.1 [debug] Proxy map: {} [debug] Request Handlers: urllib, requests, websockets [debug] Plugin directories: none [debug] Loaded 1834 extractors Usage: yt-dlp_x86_win7.exe [OPTIONS] URL [URL...] yt-dlp_x86_win7.exe: error: You must provide at least one URL. Type yt-dlp --help to see a list of all options. Less puzzled now , but still with unanswered questions ... Time to call it a night, regards...

It does. Thank you! I can confirm, the binary works on Win11 Pro (on my brand new Framework Desktop).

Updated yt-dlp_Py3.14.0_x86_Vista. (with IOS patch ) @Reino; Please test if this still suits your needs.

New build of BOC/UXP for XP! Test binary: MailNews Win32 https://o.rthost.win/boc-uxp/mailnews.win32-20251011-40a79c75-uxp-b709881389-xpmod.7z BNavigator Win32 https://o.rthost.win/boc-uxp/bnavigator.win32-20251011-40a79c75-uxp-b709881389-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.win32-20251011-id-656ea98-uxp-b709881389-xpmod.7z IceApe-UXP(suite) https://o.rthost.win/hbl-uxp/iceape.win32-20251011-id-656ea98-ia-c642e3c-uxp-b709881389-xpmod.7z source repo (excluding UXP): https://github.com/roytam1/icedove-uxp/tree/winbuild https://github.com/roytam1/iceape-uxp/tree/winbuild

Fortunately, my main system (Linux Mint) didn't have any issue compiling Brotli "automagically" when installing the dependencies for my newly compiled Python 3.14 (altinstall; that is, it is installed without interfering with system Python). Compiler is shipped with the distro. No problem, I was suspecting that already. It is.

Thanks ; they note: Hopefully, the brotli py3.14 wheel (file brotli-1.1.0-cp314-cp314-win32.whl) you kindly shared remains Vista-compatible; though, TBH, I'm not aware of any tests (within the context of yt-dlp or otherwise) I can submit it to ... Other than that, thank you for the details of the compilation procedure ; I agree it's not for the faint-hearted (or those lacking access to recent OS+H/W); when I, regrettably, find that PyPI misses wheels for a specific "module+module version+CPython+architecture" combination, I usually issue: python -m pip install module==specific_version and then let pip do its job in invoking setuptools+wheel to compile from module source; in the meantime, I cross my fingers for things to go as expected , which isn't always the case ; if the compilation succeeds, then pip will store in its cache a copy of the generated wheel, for easy re-usage; the local path to that wheel is being displayed in the command prompt window, so I always grab a copy from there ; of course, this is for pure Python modules, for ones with C/C++ extensions, a compiler must be made available to CPython (as you detailed yourself); CPython > 3.4 requires VS2015+, not compatible with my current OS ... Huh??? The above was meant to be written as: ... are compiled with adang1345's py3.13 (branch) implementation so that was a typo/oversight on my part ; apologies... Ofc I'm aware you've been using 3.13.7 for the win7 (+Vista) yt-dlp variants, that's why I wrote an update was due to recently released 3.13.8 ... ... That's your call, surely, but I'm more of a conservative person myself , who doesn't believe "the latest is the greatest"; however, if you're absolutely certain the yt-dlp code is fully compatible with recently released py3.14 (and the CPython itself doesn't come with any major bug), then go for it ; personally, I'd stick with the minimum CPython version mandated by yt-dlp, soon to be py3.10 (but it's also nice I can experiment with higher versions here, all thanks to adang1345 ) ... Best regards.

I used this: https://github.com/mstorsjo/llvm-mingw The 32bit build was compiled with Python 3.14.0 (x86) from adang1345 on Windows 10 IoT Enterprise LTSC. The 64bit build was compiled with Python 3.14.0 (x86-64) from adang1345 on Windows 7. To let Python use this compiler, place a file named `distutils.cfg` (see below) in `.\Lib\distutils` in Python install directory (create folder `distutils`). [build] compiler = mingw32 Build Brotli: Download source tarball and extract. In source directory run: python setup.py build_ext --compiler=mingw32 install Build wheel: python setup.py build_ext --compiler=mingw32 bdist_wheel In the directory where the built wheel is, run: pip install --force-reinstall brotli-1.1.0-cp314-cp314-win32.whl Huh??? .\yt-dlp_win7.exe -v [debug] Command-line config: ['-v'] [debug] Encodings: locale cp1252, fs utf-8, pref cp1252, out cp1252 (No VT), error cp1252 (No VT), screen cp1252 (No VT) [debug] yt-dlp version nicolaasjan/yt-dlp@2025.10.05.061237 (win7_exe*) [debug] Python 3.13.7 (CPython AMD64 64bit) - Windows-7-6.1.7601-SP1 (OpenSSL 3.0.16 11 Feb 2025) [debug] exe versions: ffmpeg git-2025-10-04-e05f8ac-ffmpeg-windows-build-helpers (fdk,setts), ffprobe git-2025-10-04-e05f8ac-ffmpeg-windows-build-helpers, phantomjs 2.5.0 [debug] Optional libraries: Cryptodome-3.23.0, brotli-1.1.0, certifi-2025.08.03, curl_cffi-0.13.0, mutagen-1.47.0, requests-2.32.5, sqlite3-3.50.4, urllib3-2.5.0, websockets-15.0.1 [debug] Proxy map: {} [debug] Request Handlers: urllib, requests, websockets, curl_cffi [debug] Extractor Plugins: NSigDeno (YoutubeIE) [debug] Plugin directories: C:\Users\Nico\AppData\Roaming\yt-dlp\plugins\bgutil-ytdlp-pot-provider\yt_dlp_plugins, C:\Users\Nico\AppData\Roaming\yt-dlp\plugins\yt-dlp-YTNSigDeno\yt_dlp_plugins [debug] Loaded 1833 extractors And the next Win7 builds will be Py3.14 based.

Hi ; well, cmalex's py3.11.4 assemblies (one compatible with just XP and one with Vista+) aren't currently being hosted on GitHub (AFAIAA), but they can be ; OTOH, adang1345's py3.13.7 assembly is already hosted on GitHub, and so is 3dyd's custom PyInstaller build used to make the Win7/Vista yt-dlp distributions; it may be possible, at least in theory, to author a modded GA workflow file to point to the custom implementations of CPython+PyInst necessary to package the Win7/Vista builds; e.g., upstream are using their own, custom, build of PyInstaller to create the "official" yt-dlp releases; perhaps a consultation with your "AI friend(s)" (or a question on SO, etc.) could prove beneficial in fully automating even the (currently) manually created release assets ; just a thought from a dedicated (hardly intelligent ) friend of yours... Best wishes !

IMO the internet got 'bad' once 'normal' people found it. But I think that is more nostalgia for the most part. As for when that happened, I'd say (at least in the US) that happened once Classmates came around. Everyone who was in high school or college went onto there to keep touch afterwards. That site got stepped on by Facebook which was originally designed for college students, but became so popular that old people joined it to keep in touch with those who were on it. For the current generation that is moving into the world, Facebook is seen as a site for old people which is a quote I've heard from multiple people of the younger generations, an insight I am afforded since I 'moonlight' at an arcade. This is only my perspective on things, but normal/old people from 20 years ago were basically computer illiterate, and astonishingly, people of that same stereotype in the current day are still mostly computer illiterate despite having grown up in with computers in a fully connected world. Combine that with the mentality that corporations have of treating adults (employees) like children and you end up with web technologies designed to protect the computer illiterate but in a fashion doesn't feel like the correct way to do it. I find it amusing that the link to this article on the Guardian brings me to a page where half of it is covered by a blue privacy/cookie notification. But overall I think that there are certainly many good points about how the internet has evolved and it is actually better in many ways now than it used to be. It is certainly more useful, something I didn't really grasp because I never had a mobile phone that was new enough to do anything cool until just recently. I think it will get better and we will figure out how to evolve it to deal with the current issues it has. There are more eyes on it now than ever before and that means more ideas of how to do things will occur. But that doesn't mean that there won't be some new dumb trend that everyone adopts that we'll have to deal with. We may end up looking back and laugh about how we were all bent out of shape about captchas and cookie banners in the face of whatever annoying thing we'll be having to deal with.

It 's working here on New Moon.

Updated yt-dlp_Py3.13.7_x86_Vista. (with IOS patch )

... Hi ; most unfortunately, this isn't how things work ; Google , the owners of YouTube, have unleashed a war of annihilation against every kind of YT downloaders and they're NOT excluding/pardoning any of them based on the OS they're compatible with ; and, trust me, every breaking change Google implement (on purpose) isn't meant to leave "older" versions of downloaders unscathed ... XP-compatibility in Q4 of 2025 is also a concern for many Windows apps, not only Video Downloaders... I tend to stay clear of GUI media downloaders, so, alas, if you're in search for one of them, I'm afraid I won't be the person to recommend one to you... My own experience is that most of the GUI YT downloaders are frontends of the well-known CLI (Python) apps youtube-dl and its fork, yt-dlp; the GUI apps either use the standalone youtube-dl.exe or yt-dlp.exe binaries under the hood or use the respective libs inside their source code; as a consequence, whenever Google break youtube-dl/yt-dlp, the apps become broken, too ... Recently, Google implemented a breaking change of a great magnitude, which should explain your observation that yt-dlp and youtube-dl were impacted/broken, too, a stop-gap/interim solution has been implemented for both, but they were actually "stitched back together with duct tape", meaning they can break again at any time... More permanent solutions for YouTube downloading in those apps will leave, probably, any WinOS < Win10 uncovered (too early to tell with absolute certainty, maybe Win7 will be spared, but things are very grim for XP/Vista ) ; at the time of this writing, working versions can be found at: https://github.com/ytdl-org/ytdl-nightly/releases/download/2025.09.29/youtube-dl.exe (official nightly version) https://github.com/nicolaasjan/youtube-dl/releases/download/2025.09.29/youtube-dl.exe (fork of the above with a few custom modifications) https://github.com/nicolaasjan/yt-dlp/releases/download/2025.09.27.071342/yt-dlp_x86_winXP.exe If you're not familiar with those two CLI apps, am afraid this isn't the place to ask for help; however, ample HowTos can be found online with a simple web search, both apps have detailed wiki sections (assuming one already knows to invoke apps from the Windows command prompt) ... For developments related to both youtube-dl+yt-dlp, keep an eye on the dedicated MSFN thread: https://msfn.org/board/topic/184368-who-here-has-a-youtube-dl-compile-for-winxp Best regards ...

... Probably not their work , per se (seeing that they have the Polish flag before their username); published builds there are the impressive work of a Japanese person, whose username (in Japanese) is "あべちん" (that Google translates to "Abechin") ; in fact, I had contacted him some years ago about some issues in his FFmpeg builds, at a time when machine translation wasn't perfect (and AI totally absent); his blog's comment form, e.g. http://blog.k-tai-douga.com/article/191451101.html#comment (Leave a comment) wouldn't accept just Latin characters (I posted in English), so I had to include some Kanji characters, too, for my message to go through ... We did communicate in the end, the problems were mitigated in a future FFmpeg release...

Confirmed. Thanks.

Thank you @nicolaasjan for your yt-dlp builds. I have made a Windows XP compatible build of FFmpeg that targets a Pentium 4. I relied heavily on rdp and @Reino build scripts so thanks also to them. ffmpeg-8.1-630-635cb4543f-win32-static-xpmod-P4.7z libfdk-aac-2.0.3-158-d8e6b1a-win32-xpmod-P4.7z

alright, found problematic code and fixed, will be available in next build. https://github.com/roytam1/UXP/commit/16dc4c916886bfb6947726f0e8a38a9e474f7644

That's why I decided to make diffs again. 😐 I will apply the custom changes only to the "Vista" variant offered here. It will work on higher versions of Windows as well, so @Reino can test it. Updated yt-dlp_Py3.13.7_x86_Vista.

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-20250927-fb9f6eff8-xpmod.7z Win64 https://o.rthost.win/basilisk/basilisk55-win64-git-20250927-fb9f6eff8-xpmod.7z repo: https://github.com/roytam1/basilisk55 Repo changes: - ported from UXP: No issue - Clarify Azure Back-end info. (c7e103b0) (a9de514bd) - import from UXP: Revert "[gfx] Guard against possible race via gfxFontEntry::GetFontTable." (060e2965) (068d2d8b2) - import from UXP: [gfx] Clarify comments. (6e19656e) (840d26a08) - import from UXP: Issue #2852 - Parse full cache-control headers. (16164b3d3) (88a1c0050) - import from UXP: [gfx] Use std::size rather than hardcoding an array size (3a121f34) (fc33f4e70) - import from UXP: Issue #2852 - Follow-up: Add comment and fix formatting. (1bcdad3e) (c24c415a9) - import from UXP: Issue #2790 - Follow-up: Ensure both text and background are set. (c9e08cb2f) (39e14308a) - import from UXP: Revert "Issue #2258 - Part 4: Clean up unused pointers." (7b4bb082) (99f5c8f0f) - import from UXP: Revert "Issue #2258 - Part 3: Allow sniffing with XCTO:nosniff + empty MIME type." (2d4d48db) (a38d80555) - import from UXP: Revert "Issue #2258 - Part 2: Move XCTO:nosniff check into sniffers." (73c9c1dd) (2f8889ad7) - ported from UXP: Revert "Issue #2258 - Part 1: Support XCTO:nosniff when navigating." (0d00c840) (434f021fb) - import from UXP: Issue #2258 - Make sure we favor sniffing when the content-type header is empty. (659a3c0f) (fb9f6eff8)

New build of BOC/UXP for XP! Test binary: MailNews Win32 https://o.rthost.win/boc-uxp/mailnews.win32-20250927-40a79c75-uxp-50b6047c84-xpmod.7z BNavigator Win32 https://o.rthost.win/boc-uxp/bnavigator.win32-20250927-40a79c75-uxp-50b6047c84-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.win32-20250927-id-656ea98-uxp-50b6047c84-xpmod.7z IceApe-UXP(suite) https://o.rthost.win/hbl-uxp/iceape.win32-20250927-id-656ea98-ia-c642e3c-uxp-50b6047c84-xpmod.7z source repo (excluding UXP): https://github.com/roytam1/icedove-uxp/tree/winbuild https://github.com/roytam1/iceape-uxp/tree/winbuild

... I'm sorry I have to say this , but once the hack currently implemented by upstream stops working (and it will stop ), for XP/Vista[/Win7] users, for which the scheduled workaround, deno, isn't compatible, "Apple-device" (iOS+web_safari) type of YouTube (HLS) streams will be the ONLY accessible options, at least until Google restricts them further with a POT implementation; the log below was generated with yesterday's Vista build of yours , it provides a glimpse of how things would look when "the hack" has been broken and the deno-enabling PR merged: yt-dlp_x86 --extractor-args "youtube:player_client=ios,web_safari" -vF "O5TS2CbBhKo" [debug] Command-line config: ['--extractor-args', 'youtube:player_client=ios,web_safari', '-vF', 'O5TS2CbBhKo'] [debug] Encodings: locale cp1253, fs utf-8, pref cp1253, out utf-8 (No VT), error utf-8 (No VT), screen utf-8 (No VT) [debug] yt-dlp version local@2025.09.25 [4429fd045] (win_x86_exe) [debug] Python 3.11.9 (CPython x86 32bit) - Windows-Vista-6.0.6003-SP2 (OpenSSL 3.5.2 5 Aug 2025) [debug] exe versions: none [debug] Optional libraries: Cryptodome-3.23.0, brotli-1.1.0, certifi-2025.08.03, mutagen-1.47.0, requests-2.32.5, sqlite3-3.45.1, urllib3-2.5.0, websockets-15.0.1 [debug] Proxy map: {} [debug] Request Handlers: urllib, requests, websockets [debug] Plugin directories: none [debug] Loaded 1833 extractors [debug] [youtube] [pot] PO Token Providers: none [debug] [youtube] [pot] PO Token Cache Providers: memory [debug] [youtube] [pot] PO Token Cache Spec Providers: webpo [youtube] Extracting URL: O5TS2CbBhKo [youtube] O5TS2CbBhKo: Downloading webpage [debug] [youtube] Forcing "main" player JS variant for player 0004de42 original url = /s/player/a61444a1/player_ias.vflset/en_US/base.js [youtube] O5TS2CbBhKo: Downloading ios player API JSON [youtube] O5TS2CbBhKo: Downloading web safari player API JSON [debug] [youtube] O5TS2CbBhKo: ios client https formats require a GVS PO Token which was not provided. They will be skipped as they may yield HTTP Error 403. You can manually pass a GVS PO Token for this client with --extractor-args "youtube:po_token=ios.gvs+XXX". For more information, refer to https://github.com/yt-dlp/yt-dlp/wiki/PO-Token-Guide . To enable these broken formats anyway, pass --extractor-args "youtube:formats=missing_pot" [debug] [youtube] Extracting signature function 0004de42-main-107 [debug] Loading youtube-sigfuncs.0004de42-main-107 from cache [debug] Loading youtube-nsig.0004de42-main from cache [debug] [youtube] Decrypted nsig Nc-7Qm-M-NmbBoH => S5jQA8mpxn7KDw WARNING: [youtube] O5TS2CbBhKo: Some web_safari client https formats have been skipped as they are missing a url. YouTube is forcing SABR streaming for this client. See https://github.com/yt-dlp/yt-dlp/issues/12482 for more details [youtube] O5TS2CbBhKo: Downloading m3u8 information [youtube] O5TS2CbBhKo: Downloading m3u8 information [debug] Sort order given by extractor: quality, res, fps, hdr:12, source, vcodec, channels, acodec, lang, proto [debug] Formats sorted by: hasvid, ie_pref, quality, res, fps, hdr:12(7), source, vcodec, channels, acodec, lang, proto, size, br, asr, vext, aext, hasaud, id [info] Available formats for O5TS2CbBhKo: ID EXT RESOLUTION FPS CH | FILESIZE TBR PROTO | VCODEC VBR ACODEC ASR MORE INFO ---------------------------------------------------------------------------------------------------------------------- sb3 mhtml 48x27 0 | mhtml | images storyboard sb2 mhtml 80x45 1 | mhtml | images storyboard sb1 mhtml 160x90 1 | mhtml | images storyboard sb0 mhtml 320x180 1 | mhtml | images storyboard 233 mp4 audio only | m3u8 | audio only unknown Untested, Default, low, IOS 234 mp4 audio only | m3u8 | audio only unknown Untested, Default, high, IOS 602 mp4 256x144 12 | ~ 2.55MiB 89k m3u8 | vp09.00.10.08 89k video only Untested, IOS 91 mp4 256x144 24 | ~ 5.55MiB 193k m3u8 | avc1.4D400C mp4a.40.5 WEB-S 269 mp4 256x144 24 | ~ 5.55MiB 193k m3u8 | avc1.4D400C 193k video only Untested, IOS 603 mp4 256x144 24 | ~ 5.06MiB 176k m3u8 | vp09.00.11.08 176k video only Untested, IOS 92 mp4 426x240 24 | ~ 10.22MiB 356k m3u8 | avc1.4D4015 mp4a.40.5 WEB-S 229 mp4 426x240 24 | ~ 10.22MiB 356k m3u8 | avc1.4D4015 356k video only Untested, IOS 604 mp4 426x240 24 | ~ 8.23MiB 286k m3u8 | vp09.00.20.08 286k video only Untested, IOS 93 mp4 640x360 24 | ~ 23.51MiB 818k m3u8 | avc1.4D401E mp4a.40.2 WEB-S 230 mp4 640x360 24 | ~ 23.51MiB 818k m3u8 | avc1.4D401E 818k video only Untested, IOS 18 mp4 640x360 24 2 | ? 11.63MiB 405k https | avc1.42001E mp4a.40.2 22k 360p, WEB-S 605 mp4 640x360 24 | ~ 14.13MiB 492k m3u8 | vp09.00.21.08 492k video only Untested, IOS 94 mp4 854x480 24 | ~ 41.82MiB 1456k m3u8 | avc1.4D401E mp4a.40.2 WEB-S 231 mp4 854x480 24 | ~ 41.82MiB 1456k m3u8 | avc1.4D401E 1456k video only Untested, IOS 606 mp4 854x480 24 | ~ 22.32MiB 777k m3u8 | vp09.00.30.08 777k video only Untested, IOS 95 mp4 1280x720 24 | ~ 72.92MiB 2538k m3u8 | avc1.64001F mp4a.40.2 WEB-S 232 mp4 1280x720 24 | ~ 72.92MiB 2538k m3u8 | avc1.64001F 2538k video only Untested, IOS 609 mp4 1280x720 24 | ~ 33.23MiB 1157k m3u8 | vp09.00.31.08 1157k video only Untested, IOS 96 mp4 1920x1080 24 | ~138.27MiB 4813k m3u8 | avc1.640028 mp4a.40.2 WEB-S 270 mp4 1920x1080 24 | ~138.27MiB 4813k m3u8 | avc1.640028 4813k video only Untested, IOS 614 mp4 1920x1080 24 | ~ 50.15MiB 1746k m3u8 | vp09.00.40.08 1746k video only Untested, IOS NB: I'm not even sure 4K video streams are being made available for either iOS or Safari ; opus audio is also unavailable; and the web_safari HLS streams are NOT parsed by default, they have to be explicitly requested via the '--extractor-args "youtube:player_client=web_safari"' cmdline flag (or equivalent config setting); personally, I'd hope the POT-less iOS streams remain accessible in the XP/Vista yt-dlp variants, though I realise this wish/expectation could be, simply, "a lot to ask for..." Best wishes!

Another update (sorry...) to yt-dlp for Vista, now made with the new and shiny Python 3.13.7 from adang1345. Link.

Updated yt-dlp for Vista (OpenSSL 3.5.2) here.

I wasn't misleading anyone. I said "pluck some 100 Euro notes", not pluck ONE 100 Euro note," or even "pluck a few 100 Euro notes!" We're all well aware that the cost of a modern PC is more than a few 100-Euro notes. I figured the reader would know how many notes would have to be plucked! But quibbles over adjectives aside, I think we actually agree! My whole point was that there's no such thing as a money tree! Most of us actually have to work quite hard to earn enough money to buy a modern PC; thus it was quite arrogant for someone to respond to a complaint about a Web site being deliberately blocked from working on an older PC with "just go out and buy a new one," as if that were a trivial thing for anyone to do!

98, SE, and ME do have VirtualAllocEx, only 95 doesn't. Use ImportPatcher or Dependency Walker to find all missing functions. If there are only a few, I can write a wrapper.

Hello Leokids123. You asked: "Can you disable your adblocker and enable Javascript?" Answer: Javascript is enabled in St52 when visiting Paypal. And this is the result when trying to sign in my Paypal account with ST52 and - hosts file adblocking disabled - DNS server adblocking disabled It is as NotHereToPlayGames said: "the SECURITY CHALLENGE CANNOT BE LOADED."

I did just discover that the paypal slide mouse screen in my older version of Serpent is DEPENDENT UPON the ISP that I use. I canceled my previous ISP but it remains active until the end of my billing cycle (I only have two more days) - I get no slide mouse screen in Serpent when connected via my previous ISP (cable). I *DO* get the slide mouse in Serpent (unsolvable - checkbox turns green but page does NOT advance) in Serpent when connected via my new ISP (fiber optic).

Intel finally released an official 24H2 compatible graphics drivers for 7th-10th Gen processors (probably also compatible with 25H2) https://www.intel.com/content/www/us/en/download/776137/intel-7th-10th-gen-processor-graphics-windows.html?wapkw=driver

Can you test this Win7 build: yt-dlp_x86_win7.7z ? Built with adang1345's Py3.14 with custom PyInstaller 6.15 on Windows 10.

I think that must be because I built directly after the standalone one, i.e. without deleting the build directory and start over. After I built the `onedir` version today from a fresh source directory, I got: [debug] Command-line config: ['-v'] [debug] Encodings: locale cp1252, fs utf-8, pref cp1252, out cp1252 (No VT), error cp1252 (No VT), screen cp1252 (No VT) [debug] yt-dlp version nicolaasjan/yt-dlp@2025.10.13.080600 (win_x86_dir) [debug] Python 3.11.4 (CPython x86 32bit) - Windows-XP-5.1.2600-SP3 (OpenSSL 3.5.2 5 Aug 2025) [debug] exe versions: ffmpeg N-121254-g635cb4543f (fdk,setts), ffprobe N-121254-g635cb4543f, phantomjs 1.9.8 [debug] Optional libraries: Cryptodome-3.23.0, brotli-1.1.0, certifi-2025.10.05, mutagen-1.47.0, requests-2.32.5, sqlite3-3.50.4, urllib3-2.5.0, websockets-15.0.1 [debug] Proxy map: {} [debug] Request Handlers: urllib, requests, websockets [debug] Plugin directories: none [debug] Loaded 1834 extractors

Agreed. I remember when "annoying" used to be (depracated) MARQUEE to scroll text. EVERY web site you went to had these d@mn things. Way back then, the "not old people" learned how to combat the annoyance. And it's still us that do that combat nowadays. ie, I didn't even realize the web site had a "blue privacy/cookie notification". This is all I've ever needed for those -- https://greasyfork.org/en/scripts/432050-anti-cookies-consent

For some reason the StartAllBack website states the latest version is still 3.9.15, but this installer is in fact for 3.9.16. Also, the latest version isn't coming down via checking for updates through the app.

Hmm... Must have made a mistake. Try this.

I never claimed that it is! NOR did the person that originally cited the "100", it was just a passing PHRASE *not* intended to be taken LITERALLY. YOU CAN AFFORD A COMPUTER OR A MOBILE DEVICE. I have 1-million-percent confidence in that statement. YOU CAN AFFORD ONE! I did not say that a computer costs 1% of your annual income or that it costs 10% of your annual income. The bottom-line REALITY is that EVERYBODY reading this CAN afford a computer or a mobile device. And I'd all but "bet" on it that EVERYBODY reading this has MORE THAN ONE (ie, one computer plus one mobile, or one computer plus two mobile, or two computers plus one mobile, or two computers plus two mobiles). You requested another member to "Please don't mislead people". SO WHY ARE YOU DOING THE SAME? You are ACTING LIKE you have to SELL A LUNG OR A KIDNEY to "buy a computer". And I am telling you, THAT IS MISLEADING PEOPLE.

I'd estimate that upwards of 75% of new CSS and Javascript features fall into that category. I wish I could say "unbelievable," but the arrogance of some folks is totally believable. Just go out to your money tree and pluck off some 100 Euro (or whatever your local currency is) notes, and hand them to your local PC dealer. Doesn't everyone have a money tree? And even if you're willing to spend the money, does anyone have any idea how much work goes into making your PC, well, personal? No, they want you to just throw that old PC in the rubbish and start customizing your new PC from scratch - with a new OS version that removes support for much of the personalization that Win 7 lets us do. You will have the Win 10/11 "look" whether you like it or not! Because Micro$oft says so!

Okay, well, I can report that the Restore Session page **IS** working in the most-recent version of Serpent v52. I terminated via Task Manager ten consecutive times and ALL TEN presented a populated windows/tabs table. In the past, two or three of those ten would *NOT* present the table, it would just open/restore my last session **WITHOUT** presenting the table - this is not a "functioning" feature. But I can not replicate this behavior with most-recent version, ALL TEN terminations function properly.

Try these experimental filters in Ublock Origin (I don't know if they are working with uBO Legacy...). From: https://old.reddit.com/r/uBlockOrigin/comments/1nf7heb/youtube_ads_detection_breakages_2025_megathread_2/ It worked for me in current Firefox. </Offtopic>

Hi @NotHereToPlayGames, at first my compliment for your hard work here! You made it possible to have a more recent Chrome browser in Windows XP without annoying Chinese language strings. Thanks for that! Now I have a question relating to the RAM consumption of your 360Chrome editions. I use Windows XP on a very old computer with only 1.5 GB RAM. Is there an option to reduce RAM consumption significantly? Maybe a kind of single process mode as in Mypal 68? AFAIK people are advised to use 360Chrome version 11 on older systems. I tried your 360Chrome version 11 and 13.5, but in both cases the RAM consumption is very high starting the browser with an empty tab only. Any recommendations? Cheers, AstroSkipper

Here is a tip how to get completely rid of an installed addon in Mypal 68: 1. You have to find out the extension ID of that addon. Click Help -> Troubleshooting Information. In section Extensions you can copy the related ID. 2. Under Tools -> Add-ons remove the problematic addon. 3. Close browser. 4. Go to your profile folder and open the file extensions.json in a good editor like Notepad++ which supports highlighting of brackets. 5. Search for the extension ID and delete its complete string from {"id": to "location":"app-profile"} curly brackets inclusive. Each extension begins and ends with these curly brackets. Between two different extensions may be only one comma. 6. Now delete in subfolder extensions related xpi file and its staged folder if existent. 7. Now start your browser, open Add-ons and you'll see the problematic addon has gone completely. Kind regards, AstroSkipper

You mean while calling up MU in Windows XP you get the error code 80072F78. Is that correct?

You're welcome! Nice to hear that you liked it.

@Dave-H Fine. I've added my HTTPSProxy's config.ini so you can use that or compare it to yours. https://www.mediafire.com/file/6emtdvx2vmw4iz8/HTTPSProxy_-_config.ini/file

Solve captcha to buy a loaf of bread , thanks but no thanks . I ignore and boycott such "shops". Do you know we train AI for them for free (as slaves) ?

here's a good read for ya , from the respected member @sdfox7 https://msfn.org/board/topic/176837-disable-web-fonts-in-google-chromemozilla-firefox/