Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

New Twitter, new problem #75

Closed
crockpoot opened this issue Jul 25, 2019 · 10 comments

Comments

@crockpoot
Copy link

commented Jul 25, 2019

Just noticed that Twitter has a new UI. I don't hate it but what I do hate is that I received none of my usual Twitter feeds today. I assume the two are connected.

@ciderpunx

This comment has been minimized.

Copy link
Owner

commented Jul 25, 2019

Yes, I'm afraid it looks like they broke things for us again.

I'll need to refactor the code, which isn't likely before the weekend.

For now the workaround is to use the mobile version of the code. Change the /twitter_user_to_rss in your URL to mobile_twitter_to_rss and you should be OK for now.

I've not rewritten search yet, so that'll remain broken for now. Sorry about that!

I'll update the front page of the site to use the mobile code.

ciderpunx pushed a commit that referenced this issue Jul 25, 2019

root
Default ot mobile search
Twitter updated their default UI, so we update to using the mobile
version of twitrss.me for users rather than the desktop version.

Search will remain broken for now until I can fix it all up.

cf. #75
@micressor

This comment has been minimized.

Copy link

commented Jul 25, 2019

@ciderpunx It looks like apache.conf is not updated for the /mobile_twitter_to_rss?

@Omnialias

This comment has been minimized.

Copy link

commented Jul 25, 2019

Something that tripped me up a little that is worth mentioning here: the syntax of the value of the dc:creator element when using mobile_twitter_to_rss.pl is different and includes the "friendly name" on Twitter, not just the username.

@dperelman

This comment has been minimized.

Copy link
Contributor

commented Jul 28, 2019

I was able to fix it on my end by just enabling cookie support when screenscraping: bd3e47c . I'll make a PR in a moment.

@fabrei

This comment has been minimized.

Copy link

commented Jul 31, 2019

For now the workaround is to use the mobile version of the code. Change the /twitter_user_to_rss in your URL to mobile_twitter_to_rss and you should be OK for now.

thanks for your suggestion to change it to /mobile_twitter_to_rss ..but it only works once for a little while..it worked today in the morning with one twitter account. afterwards, it didn't work with another. i got this message: ERR: Can’t screenscrape Twitter Redirecting you back to TwitRSS.me in a few seconds. You might have spelled the username wrong or something. but after more or less exactly 5 hours it worked with another account. is there a kind of time frame to wait until it works again?

@1lomeno3

This comment has been minimized.

Copy link

commented Aug 5, 2019

Any update on this? The workaround does not work for me :(

@dperelman

This comment has been minimized.

Copy link
Contributor

commented Aug 5, 2019

@1lomeno3 By "the workaround" are you referring to /mobile_twitter_to_rss, my PR (#76), or both? My branch has some of my own changes which I believe are unrelated, but it's currently working for me.

@Omnialias

This comment has been minimized.

Copy link

commented Aug 5, 2019

I made the changes that you added in your PR @dperelman and am getting the same problems that @fabrei has. Specifically, it seems like cookie caching might be getting in the way of something? I have found that I have to browse to the RSS feed, refresh it a few times, then it'll grab the RSS. This works for a day or so, then breaks and I have to go and refresh that RSS feed again.

@dperelman

This comment has been minimized.

Copy link
Contributor

commented Aug 5, 2019

Oh. I see. The way I'm using it (calling it directly from Liferea, not hosting it on a web server), I only ever make one request per invocation of the script, but it actually handles requests in a loop. My PR initializes the cookie jar outside of the loop. If you instead move

$browser->cookie_jar( {} );

to immediately before

my $response = $browser->get($url);

then it should clear cookies before every request, which might get better behavior. I'll update my PR.

@ciderpunx ciderpunx closed this in #76 Aug 6, 2019

@ciderpunx

This comment has been minimized.

Copy link
Owner

commented Aug 6, 2019

So I believe that #76 and #77 combined resolve the issue for now.

My apologies for being so slow to respond, I've been AFK for a while for various reasons and not keeping as much of an eye on TwitRSS as I should have.

In the long run, I want to do a proper refactor with the idea that the parser/fetchers can be modules used by the cgi process rather than part of the controller script itself.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
7 participants
You can’t perform that action at this time.