twint = broker after change of policy of twitter.com
Twitter.com has changed its policy and only supports a limited set of browsers (https://help.twitter.com/en/using-twitter/twitter-supported-browsers).
This yields the following output/error for the python Requests library when requesting the twitter Token:
b'<!DOCTYPE html>\n<html dir="ltr" lang="en">\n<head>\n <meta charset="utf-8" />\n <meta name="viewport"\n content="width=device-width,initial-scale=1,maximum-scale=1,user-scalable=0,viewport-fit=cover" />\n\n <style>\n body {\n -ms-overflow-style: scrollbar;\n overflow-y: scroll;\n overscroll-behavior-y: none;\n }\n\n .errorContainer {\n background-color: #FFF;\n color: #0F1419;\n max-width: 600px;\n margin: 0 auto;\n padding: 10%;\n font-family: Helvetica, sans-serif;\n font-size: 16px;\n }\n\n .errorButton {\n margin: 3em 0;\n }\n\n .errorButton a {\n background: #1DA1F2;\n border-radius: 2.5em;\n color: white;\n padding: 1em 2em;\n text-decoration: none;\n }\n\n .errorButton a:hover,\n .errorButton a:focus {\n background: rgb(26, 145, 218);\n }\n\n .errorFooter {\n color: #657786;\n font-size: 80%;\n line-height: 1.5;\n padding: 1em 0;\n }\n\n .errorFooter a,\n .errorFooter a:visited {\n color: #657786;\n text-decoration: none;\n padding-right: 1em;\n }\n\n .errorFooter a:hover,\n .errorFooter a:active {\n text-decoration: underline;\n }\n </style>\n</head>\n<body>\n <div class="errorContainer">\n <img width="46" height="38"\n srcset="https://abs.twimg.com/errors/logo46x38.png 1x, https://abs.twimg.com/errors/[email protected] 2x"\n src="https://abs.twimg.com/errors/logo46x38.png" alt="Twitter" />\n <h1>**This browser is no longer supported**.</h1>\n <p>\n Please switch to a supported browser to continue using twitter.com. You can see a list of supported browsers in our Help Center.\n </p>\n <p class="errorButton"><a href="https://help.twitter.com/using-twitter/twitter-supported-browsers">Help Center</a>\n </p>\n <p class="errorFooter">\n <a href="https://twitter.com/tos">Terms of Service</a>\n <a href="https://twitter.com/privacy">Privacy Policy</a>\n <a href="https://support.twitter.com/articles/20170514">Cookie Policy</a>\n <a href="https://legal.twitter.com/imprint">Imprint</a>\n <a href="https://business.twitter.com/en/help/troubleshooting/how-twitter-ads-work.html">Ads info</a>\n \xc2\xa9 2020 Twitter, Inc.\n </p>\n </div>\n</body>\n</html>'
kgoderis
All 4 comments
I am also finding this.
If I do a simple lookup(user) in twint, it produces a NoneType error, and checking where this arises leads me to module user.py at function User, where it attempts u.id = inf(ur, "id"). In other words it seems to me as borne out above that Twitter is not accepting access using any of the User_agents and hence is asserting the browser is not supported, hence no data returned, hence no id field (or any other fields) scraped, and thus a null return.
So at present it looks like the user-agent list is not enabling access. This has been ongoing since yesterday 16 December. Prior to that no issue. Sounds like Twitter have changed something again.
I have tried a simplistic solution which is to use user agent Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.88 Safari/537.36 which seems to be an up to date user agent for Chrome v 87 respected by Twitter but no joy. This is probably at the limit of my technical ability in terms of work arounds. I think what is needed is something better than a simple user-agent string approach, and instead implement a 'zombie' version of a recognised browser and channel requests via that.
Vickycats
on 17 Dec 2020
Can confirm, been dead since yesterday..
iJohnMaged
on 18 Dec 2020
Mine works fine after the last patch on token.py yesterday? See Issue #1061 and PR #1075
innocentius
on 18 Dec 2020
Ah... I missed that one because there is so little activity on this repo in general. Indeed, such a patch should fix it
Sent from my iPhone
On 18 Dec 2020, at 03:11, Innocentius notifications@github.com wrote:

Mine works fine after the last patch on token.py yesterday?—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or unsubscribe.
kgoderis
on 18 Dec 2020
Related issues
edsu
·
3Comments
dmuth
·
4Comments
adnanmig
·
3Comments
Nestor75
·
3Comments
Nestor75
·
4Comments
Most helpful comment
I am also finding this.
If I do a simple lookup(user) in twint, it produces a NoneType error, and checking where this arises leads me to module user.py at function User, where it attempts u.id = inf(ur, "id"). In other words it seems to me as borne out above that Twitter is not accepting access using any of the User_agents and hence is asserting the browser is not supported, hence no data returned, hence no id field (or any other fields) scraped, and thus a null return.
So at present it looks like the user-agent list is not enabling access. This has been ongoing since yesterday 16 December. Prior to that no issue. Sounds like Twitter have changed something again.
I have tried a simplistic solution which is to use user agent Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.88 Safari/537.36 which seems to be an up to date user agent for Chrome v 87 respected by Twitter but no joy. This is probably at the limit of my technical ability in terms of work arounds. I think what is needed is something better than a simple user-agent string approach, and instead implement a 'zombie' version of a recognised browser and channel requests via that.