Support for Youtube SSL/https - Userscripts.org

Signup
Login

3ICE's Youtube Link Cleaner / Optimizer / Shortener / Tracking Remover

By 3ICE — Last update Jul 24, 2011 — Installed 12,436 times.

Support for Youtube SSL/https

in
7 posts, 3 voices

Apr 10, 2011 12:14pm ranch2thedre... User	as per: http://userscripts.org/topics/69990 could support for SSL/https be included? First I tried modifying the includes - btw, I suggest using the ones from Download YouTube Videos as MP4: // @include http://.youtube.com/watch? // @include https://.youtube.com/watch? instead of what is currently used: // @include http://www.youtube.com/* ... but this didn't make it work. Then I tried changing 2 instances of `http` in the script to `https` instead, but didn't make it work either. Example: https://www.youtube.com/watch?v=1rnfE4eo1bY&pla...

Apr 10, 2011 12:45pm 3ICE Script's Author	Thanks. Userscripts.org is giving me a 404 error no matter where I click, so I will not be able to read your message for a while. Hope it's fixed soon. --3ICE (Just getting the system to send you this message took me like 10 tries.)

Apr 10, 2011 12:57pm 3ICE Script's Author	(Userscript.org is almost back to normal now.) Hi, I have added support for https. New includes: // @include http://www.youtube.com/* // @include https://www.youtube.com/* New regexp failsafe: if(!document.location.href.match( /https?:\/\/[a-zA-Z\.]youtube\.com\//)){return;} `/https?:/` matches "http:" and "https:" (Regexp will have to be updated again, replacing `[a-zA-Z\.]` with `www.`.) --3ICE p.s.: And if you can show me at least one youtube link where the part in front of ".youtube.com" is not "www", I will of course replace "www" with "*". (Few years ago they did use country specific subdomains, like "uk.youtube.com", "fr.youtube.com", but that is no more.)

Apr 10, 2011 10:44pm ranch2thedre... User	Just tested new version of the script but my example url above wasn't cleaned. `http` still works though. I've tested with newest version of Greasemonkey on AMO (0.9.2), with Scriptish version 0.1b9 which is the newest ("preliminarily") reviewed Scriptish on AMO and version 1.0 of Scriptish on GitHub (have not tested Scriptish version 1.0 from AMO, but it seems like it's older than the one on GitHub so...). So as per this statement of yours: I have added support for https. Are you really sure it supports `https`? or put in another way: did your update really change anything to the better in practise/reality? EDIT: tried replacing `[a-zA-Z\.]` with `www.` as yeah if `[a-zA-Z\.]` just is kinda like a placeholder like ``, but meaning 3 characters (+`.`) then yes `www.` would seem to do it as well. Didn't make any difference - still doesn't work on `https`. EDIT2: Also tried removing the regexp altogether as you suggest in the code (can be easily seen here fx) making the script run on all websites, but didn't better the situation either. if you can show me at least one youtube link where the part in front of ".youtube.com" is not "www", I will of course replace "www" with "" Ah, you've got me (I can't show you one). I can see that. Though `` works just as great as `www.`. Or potentially it "works better"/is better because it is prepared fx if they go back and use country specific subdomains as you mention. I could understand if you were against having/adding an include like: // @include http://youtube.com/watch? I'm against that as well. No point in doing that as it takes up an extra line/include and currently YouTube doesn't use this structure (have they ever?), they have decided to additionally use `www.` and that will get auto-added if you try a url without. But actually that wasn't the main reason why I suggested the includes used by Download YouTube Videos as MP4. The main reason (apart from of course that it have an `https` include) is because it has `watch?` in it compared to yours. And I think it's better practice to be specific about the include instead of having sitewide ones. So can you show me a "polluted" YouTube link that ain't in the `youtube.com/watch?`-form? Btw did you notice the `&list=UL`-part I had added to my example url? you could add that to your list you're maintaining... Userscripts.org is giving me a 404 error (...) Yeah I had been getting LOTS of errors too earlier around when I created this topic. EDIT3: related or new script you think?: noticed when doing a YouTube search that the parameter `&aq=f` gets added to the url. I searched on Google but found no documentation on it. I investigated a little further on YouTube and it seems it might have something to do with the Autocomplete/suggest function. If you choose suggestion 1 the parameter `&aq=0` will be added to the url. So from `0-9`. I guess `f` in `&aq=f` probably stands for `false`. The `a` might stand for `autocomplete` and `q` for `query`, or what do you say? I'd be interested if you had a better suggestion. I'm curios as to the shortening and info behind the parameter. Could you add to the script the function to keep YouTube's search function from adding these "polluting" parameters? EDIT4: I realize now requesting this function to the script give me a "polluted" YouTube link that ain't in the `youtube.com/watch?`-form. It would be in the `youtube.com/results?`-form? Also I now realize from another case I've found that if using just the more narrow `youtube.com/watch?`-form as the include then it won't clean out the `&feature=fvst`-parameter in the 'Featured Videos'-video links (example (screenshot) - in this case fx the 2nd link "Beyoncé - Ego (Remix) ft. Kanye West" will link to `http://www.youtube.com/watch?v=iW5EzxFR4SM&feature=fvst`). I've collected, a list of searches that may give you this gray 'Featured Videos' block, by using YouTube's related search suggestions (what I call the bar with 9 suggestions in the previous screenshot. It can also have 10 though). Playlist links won't be cleaned either if you have the narrow include: `http://www.youtube.com/watch?v=kJFscLngknY&playnext=1&list=PL3BFF255D66A5F7C6`. Maybe this script is too rigid/tough in this playlist example though. Wouldn't it be better if the image to the left of the playlist and the title link of the playlist linked to the playlist like: `http://www.youtube.com/playlist?p=3BFF255D66A5F7C6` or just kept unchanged (`http://www.youtube.com/watch?v=kJFscLngknY&playnext=1&list=PL3BFF255D66A5F7C6`) and then the 3 (or 2 or 1. I think 3 is max and also most dominant) links to the right of the image would be cleaned to be just the links of the videos. Don't know if that's possible though. EDIT5: just stumbled upon another way for playlist url: `http://www.youtube.com/view_play_list?p=3BFF255D66A5F7C6` - I think I like the first and shorter one better though. + doh not to forget the frontpage which gives you: &feature=grec_index &feature=topvideos_CATEGORY &feature=aso &feature=fvhl if you use the narrow include. Related to this are 2 things: 1) `http://www.youtube.com/user/beyonceVEVO?feature=chclk` won't be cleaned out - neither at the search page or when following the link. Neither will `http://www.youtube.com/user/aliciakeys?blend=1&ob=5` - tried looking for those two parameters, only found this which didn't make me any wiser. I've seen `blend=` so far with these `1,4,6,7,8,9,10,23,24`. I've seen `ob=` with `4,5` so far. As to - us users - what is it good for? Absolutely nothin'! it seems. Neither `http://www.youtube.com/user/RihannaVEVO?feature=artistob` is cleaned. And on this page neither is `http://www.youtube.com/watch?v=HAfFfqiYLp0&feature=channel_video_title` - it is though if you follow the link, but still... 2) the "destiny's child" related search suggestion link has the url: `http://www.youtube.com/results?suggested_categories=10&search_query=destiny's+child&lclk=destiny's_child/d/st/0` - I think it should really just be like `http://www.youtube.com/results?search_query=destiny's+child`. More: `http://www.youtube.com/artist?a=GxdCwVVULXewLzmGzHQKeUHShvQYOHQH&feature=artistob` isn't cleaned. There's also `http://www.youtube.com/shows` and it doesn't clean the links totally: `http://www.youtube.com/show/isitagoodideatomicrowavethis?feature=recent_shows`. Also there's the ones of the form `youtube.com/show?p=` like `http://www.youtube.com/show?p=AbxogaeOUEs`, but they seems to get all cleaned - so does `http://www.youtube.com/show/gma`. And the last thing I've found for now: YouTube - Google Instant with Bob Dylan has the link `http://www.youtube.com/redirect?q=http://gawker.com/5633106/the-google-ad-steve-jobs-will-hate&event=as_seen_on_redirect&usg=JzZaIKzlt1BmDxi-jkaZtKAFkNg=` when it should really just be `http://gawker.com/5633106/the-google-ad-steve-jobs-will-hate`. Similarly YouTube - Introducing Google Instant has the link `http://www.youtube.com/redirect?q=http://www.google.com/instant&event=as_seen_on_redirect&usg=ACYuEp5tRd6Spv1_bn6kbNL0UrE=` when it should really just be `http://www.google.com/instant`. In the same way YouTube - Kanye West - All Of The Lights ft. Rihanna, Kid Cudi has the link `http://www.youtube.com/redirect?q=http://www.omg-facts.com/page/1&event=as_seen_on_redirect&usg=jPhZ7jtGeXIdiXdpkUN3G-P9jfY=` that should just be `http://www.omg-facts.com/page/1`, BUT in addition this one also has the link `http://www.youtube.com/artist?a=GxdCwVVULXd4aUDsb_CKDIl61iOFWAlF&feature=watch_metadata` which should just have been `http://www.youtube.com/artist?a=GxdCwVVULXd4aUDsb_CKDIl61iOFWAlF`. phew ''O.O'''! damn YouTube...!

Apr 22, 2011 11:39am 3ICE Script's Author	The scope of my userscript is limited to cleaning video links that begin with "http://www.youtube.com/watch?v=" and only does this on pages that start with "http://www.youtube.com/". If I tried to keep up with all the user pages, playlists, search results, and redirects, then my script would turn into a bloat. Auto-playlists are destroyed by my script, but I'm okay with that. I have always used the "view_play_list" method you stumbled upon just now. If I add the "/watch?" part to includes, then the script would not run on youtube.com frontpage for example. I found the reason why my update didn't work. I forgot to test the isYouTubeLink function. What I missed was the `or (s.substr(0,29)=="https://www.youtube.com/watch?")` line's second number. It should have been 30 instead of 29 because of the added "s" character: `or (s.substr(0,30)=="https://www.youtube.com/watch?")` As for all your other notes, they should definitely be turned into a new script. But someone else is going to have to write it. --3ICE

May 15, 2011 10:17am derula Scriptwright	Well... no. No, it still won't work for HTTPS. Why? Yes, isYouTubeLink will correctly return true if on a YouTube site. However, cleanURL won't work because the position of the ? in the link is hardcoded. That's what regular expressions are for... I'll have a go at improving the script if you don't mind? Edit: Wow, I have a weird problem here. The onload handler never fires! I haven't even changed anything about that code. Any ideas? Edit2: Fixed that by changing the false in b.addEventListener("load", CleanURLs, false); to true. Anyway, all the links ranch2thedressing mentioned are now cleaned correctly. Please tell me, 3ICE, if I can release that! I'll leave your credit in it, of course.

Jul 24, 2011 5:21pm 3ICE Script's Author	Permission to fork my script granted to derula (in a private message). EDIT: Will link to your version on the main page when it's uploaded. Please send me a PM with the script ID once you published it. EDIT2: Thanks for reporting the hardcoded "?" issue in cleanURL, I fixed it. I am publishing version 3.333 in the next hour. Why a whole hour? Because I know I will get carried away while testing :) I also copied your fix for the addEventListener()s. I don't know why they worked before, but they stopped working for me too just now so the fix was very useful. Thank you.

Voices