1219518 - Blank page when returning from Play Store Activity from Ritual redirect, unlike Chrome

Status: RESOLVED INCOMPLETE

(Firefox for Android Graveyard :: General, defect, P5)

Description

[Michael Comella (:mcomella) [NI reported issues only: ex-Mozilla]]

Attachment: Blank redirect page source (unformatted, sorry)

In bug 1218086, I solved an issue with intent handling on the following link:
  http://invite.ritual.co/GAVIN4555

However, after I fixed it, the behavior diverged from chrome, as per bug 1218086 comment 4:

Worth noting that they do a redirect which acts differently in Firefox and Chrome after this patch. In Chrome, the play store will open and when you press back, it'll take you to Bugzilla. In Firefox, the play store will open and when you press back, it'll take you to their blank redirect page: https://bnc.lt/m/qgZOBscy2m

Comment 1

[:Margaret Leibovic]

miketaylr, do you think this is a website issue, or something we're doing wrong on our end?

Comment 2

[Michael Comella (:mcomella) [NI reported issues only: ex-Mozilla]]

I'll try to capture the redirect pages to determine what's really going on here.

Snorp mentioned Chrome checks for handlers on every link press whereas we wait for the mimeType and so this could be causing the behavior. He mentions we should manually go back via the back stack or something else I don't remember.

Comment 3

[Mike Taylor [:miketaylr]]

Testing the redirect-chain via curl with a Fennec ua:

http://invite.ritual.co/GAVIN4555 -> 301
https://bnc.lt/m/qgZOBscy2m -> 307
intent://open#Intent;scheme=ritualco;package=co.ritual.app;S.browser_fallback_url=market%3A%2F%2Fdetails%3Fid%3Dco.ritual.app%26referrer%3Dlink_click_id%253D188364718606453296?referrer=link_click_id%3D188364718606453296;end

(I guess the bnc.lt URL is from the https://branch.io/ service)

I'm gonna leave ni? to come back and look once Bug 1218086 lands in Nightly (I have no idea how to pull down and build a reviewboard patch w/ my github/gecko-dev setup).

Comment 4

[Mike Taylor [:miketaylr]]

(oops, re-setting ni?)

Comment 5

[Mike Taylor [:miketaylr]]

Looking at this again, I wonder if it's not a difference between how we and Chrome Mobile handle history/back buttons. Is it possible Chrome tries to do something smart like hop back in the stack if the previous entry is a 301 w/ Content-Length: 0? 

I can't really see anything obvious to suggest the site is treating us differently than Chrome -- this is happening a the HTTP level (AFAICT) and the responses are the same.

Comment 6

[Mike Taylor [:miketaylr]]

I just tested on Opera Mobile (both Blink and old Presto), and they have the same behavior as us. So it seems like Chrome Mobile is just doing something nicer for their users here.

Comment 7

[Michael Comella (:mcomella) [NI reported issues only: ex-Mozilla]]

Realistically, I'm not going to get to this so I'm unassigning.

By the way, thanks for your help, Mike. :)

Comment 8

[Mike Taylor [:miketaylr]]

(seems like another instance of this bug here: https://github.com/webcompat/web-bugs/issues/5499)

Comment 9

[Firefox Bug Husbandry Bot]

Re-triaging per https://bugzilla.mozilla.org/show_bug.cgi?id=1473195

Needinfo :susheel if you think this bug should be re-triaged.

Comment 10

[BMO Automation]

We have completed our launch of our new Firefox on Android. The development of the new versions use GitHub for issue tracking. If the bug report still reproduces in a current version of [Firefox on Android nightly](https://play.google.com/store/apps/details?id=org.mozilla.fenix) an issue can be reported at the [Fenix GitHub project](https://github.com/mozilla-mobile/fenix/). If you want to discuss your report please use [Mozilla's chat](https://wiki.mozilla.org/Matrix#Connect_to_Matrix) server https://chat.mozilla.org and join the [#fenix](https://chat.mozilla.org/#/room/#fenix:mozilla.org) channel.