Jump to content
MSFN is made available via donations, subscriptions and advertising revenue. The use of ad-blocking software hurts the site. Please disable ad-blocking software or set an exception for MSFN. ×


  • Content Count

  • Joined

  • Days Won

  • Donations


mixit last won the day on September 17 2019

mixit had the most liked content!

Community Reputation

96 Excellent

About mixit

Profile Information

  • OS
    XP Pro x86

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Indeed, and only the ESR versions of FF 52 have out-of-the-box support for all other NPAPI plugins. @siria 's tweak from above is needed to enable them in the "regular" 52. As for @Sfor's original question about UA spoofing, I used Custom User Agent String for a while, but ultimately got fed up with its IMHO inconvenient and error-prone UI for managing the UA strings. Since then I've been using UAControl in conjunction with User-Agent JS Fixer. which also spoofs the UA JavaSript sees. UAControl, being a XUL extension, should work well with FF 51 as well, should you need to continue using
  2. @Mathwiz Saw your Github discussion from a while back because you credited me in an edit to an earlier post, thanks. Since I didn't notice you mentioning if you're already doing this or not, let me say just in case that you should probably use the same UA you have for github.com on githubassets.com and githubusercontent.com. The second one may be less important, but I experienced an issue not too long ago with spoofing FF 60.0 only on github.com and some UI script coming from githubassets.com not being in sync with that because of a different version in my non-site-specific UA.
  3. @Aram & @Dave-H @-moz-document domain(instagram.com) { div._97aPb > div { display: block !important; } div.bsGjF > div { display: block !important; } div.bsGjF > div > div { display: block !important; } } seems to work with both single videos and carousels, or instagram.com##div._97aPb > div:style(display: block !important) instagram.com##div.bsGjF > div:style(display: block !important) instagram.com##div.bsGjF > div > div:style(display: block !important) added to uBlock's cosmetic filters. (And not to be a grouch, but when you want help
  4. @-moz-document domain(instagram.com) { div._97aPb > div { display: block !important; } } is what it should be in userContent.css. I don't use userContent.css myself and I guess I figured people would just plug the (entire) changed part into userContent.css based on the previous example. Not that \i blame you, I've made plenty of similar mistakes copying code from one kind of syntax to another. (Edit: And even in my last 2 posts here, I've somehow managed to write !'d instead of I'd and \i instead of I That's actually kind of mysterious, though I guess an irregular sleep schedule can
  5. Thanks! You totally imitated my style, too, so if !'d missed your note, I might have wondered a while about when on earth did I add that line . FWIW, there could be more stuff like that for branched browsers, my focus was always on mainline Firefox.
  6. instagram.com##div._97aPb > div:style(display: block !important) seems to work for me on ESR 52 right now (in conjunction with having 56.0 in the UA). (I'm not crazy about using the non-specific child thing in there, but that other div has no class of its own, and _97aPb doesn't seem to be used elsewhere, so hopefully this won't break anything else.)
  7. Based on my testing these fools are doing this solely based on the OS version part of the user agent string, e.g. "Mozilla/5.0 (Windows NT 6.1; rv:52.0) Gecko/20100101 Firefox/52.0" works, but "Mozilla/5.0 (Windows NT 5.1; rv:52.0) Gecko/20100101 Firefox/52.0" doesn't. Good old FF ESR 52.0 seems to work fine otherwise (I don't have an account, so can't check for logged-in features). It never ceases to amaze me how lazy web developers can be with their a**-umptions, even though there are ways to check if the actually needed functionality is present or not. It's not as if a major site like
  8. I am indeed a genius, because I already had this solution two days ago, but thought there was something missing from it, because I could only get it to work when loading a video page URL directly, and not when clicking on a video on a user's list. You know, when it pops up with apparently the same URL, but it looks like the video is put into an overlay over the list page with an X in the corner. So I spent quite some time trying to debug this, and in the end it turned out that for some weird reason Instagram was using different URLs to fetch my test videos in these two cases. The direct-to-vid
  9. Apparently they've changed something, at least on the @qvcuk page you gave as an example, the latest video https://twitter.com/qvcuk/status/1108737574363754497 doesn't work. It's probably not really a video problem, but they've changed something in the surrounding JS/CSS. At least their earlier GIF https://twitter.com/qvcuk/status/1108698147474341890 and video https://twitter.com/qvcuk/status/1108654509352378369 work fine for me.
  10. Turns out you don't have to patch anything in the browser, just change the page styling:
  11. You guys give up too easily, is this the infamous XP lair of fanatics or not!? Fortunately there is (for time being, anyway) a pretty simple fix for this "video problem". (I use quotes because it has nothing to do with video as such.) You don't need to switch or recompile browsers either - at least not right now over this particular issue. Apparently ESR 52.9 handles CSS flex layout a bit differently, so with the current Instagram code the video and its surroundding elements kindly get the height of 0 pixels... This can be fixed by modifying a surrounding <div> container to use
  12. Unless people are seeing issues, in this case the delay-load linkage to brcypt.dll.as such shouldn't be a problem, even in the original Office 2007 Compatibility Pack SP3 (KB2526297) MSO.DLL (12.0.6607.1000) has it. I thought the same thing you did at first, but going back and checking earlier versions seems reassuring.
  13. https://msfn.org/board/topic/171814-posready-2009-updates-ported-to-windows-xp-sp3-enu/?do=findComment&comment=1148337 Just look in the Brazilian folder instead of Portuguese. (Incidentally, this post was the very next one after the one you quoted )
  14. You shouldn't need to restart the stream, just hitting the left arrow key and then the right arrow key should do it - this jumps you 5 seconds back and then 5 seconds forward to where you were before (in case some other page element has focus, you may need to click on the video first). It's still an annoyance, but much less disruptive this way. I pretty much do it on autopilot at this point, after years of practice courtesy of Mozilla. This works at most other video sites as well. since most players bind these keys similarly.
  15. @Dave-H I'm still getting the 2017 version here. MS caches seem to be a crapshoot in terms of getting the latest certificate updates (for example I'm also still not getting the latest update @heinoganda notified us about). Not the first time this has happened, either. I wouldn't even be surprised if the version you downloaded manually just now was different than the one WU gets when it tries. I don't know about "exactly", but functionally, yes, for our purposes they should be the same. The automatic updater wouldn't know about your manual updates as the mechanism it uses is different (.
  • Create New...