1327002 - Browser tries to load text as url if I drop it to web page, and interferes with tab history

Status: NEW

(Firefox :: Tabbed Browser, defect, P3)

Description

[arni2033]

>>>   My Info:   Win7_64, Nightly 49, 32bit, ID 20160526082509
STR_1:
1. Open http://example.org in a new tab
2. Select all text in urlbar, replace it to "https://ya.ru", press Enter
3. Click Back button in urlbar
4. Drag and drop text "asdf 1" from this tab or from external application (Notepad++)
   to the web page opened in Step 3

AR:
 Browser shows "about:blank" in urlbar. Browser shows error page "The address isn�t valid"
 Browser doesn't allow user to click Back button in urlbar to go back to http://example.org ,
 actually it just replaces current tab item with error page.
ER:  Either X1 or X1+X2 or Y
 X1) Browser should add error page as NEXT item in tab history, allowing user to click Back button
     in urlbar to go back to http://example.org  (STR_2)
 X2) Browser should display in urlbar the actual url it was trying to access
 Y)  Browser should navigate to search results in default search engine (Google) for "asdf 1" (STR_3)


STR_2:  (reference of good behavior)
1. Open http://example.org in a new tab
2. Drag and drop text "asdf 1" from this tab or from external application (Notepad++)
   to the web page opened in Step 1
   
AR:
 "about:blank" is displayed in urlbar. Browser shows error page "The address isn�t valid". Back button
 in urlbar is enabled, allowing user to click Back button in urlbar to go back to http://example.org


STR_3:  (reference of good behavior)
1. Open http://example.org in a new tab
2. Select all text in urlbar, replace it to "https://ya.ru", press Enter
3. Click Back button in urlbar
4. Drag and drop text "asdf 1" from this tab or from external application (Notepad++)
   to the TAB in tabs toolbar opened in Step 1

AR:  Browser navigates to search results in Google for "asdf 1", even if you skip Steps 2,3

Comment 1

[Cosmin Muntean [:cmuntean], Ecosystem QA]

User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:50.0) Gecko/20100101 Firefox/50.0
Firefox: 50.1.0, Build ID: 20161208153507

I have managed to reproduce this issue on the latest Firefox (50.1.0) release and latest Nightly (53.0a1) build.

Comment 2

[:Gijs (out until Jan 5th; he/him)]

Can you check if this is a regression (e.g. can you reproduce in Firefox 4, or later versions) ?

Comment 3

[Cosmin Muntean [:cmuntean], Ecosystem QA]

I have tested this issue on Firefox 4, but at step 4, when I drag and drop the "asdf 1" text in the page I get a "The URL is not valid and cannot be loaded." error.

Also, I have tested this on Nightly 15.0a1 and the issue is not reproducible. On this version I am able to click the "Back" button and then I am redirected to the new tab page, not to the "example.com" how Arni described in expected results.

I have performed a regression considering this is a good behavior. Here are the results:

Last good revision: 02b26fb307b4 (2012-07-10)
First bad revision: 53860f11100c (2012-07-11)
Pushlog: http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=02b26fb307b4&tochange=53860f11100c

Gijs, should we consider this a regression? I am asking you this because I have performed the regression on the assumption that the above described behavior was the correct one.

Comment 4

[:Gijs (out until Jan 5th; he/him)]

(In reply to Cosmin Muntean [:CosminMCG] from comment #3)
> I have tested this issue on Firefox 4, but at step 4, when I drag and drop
> the "asdf 1" text in the page I get a "The URL is not valid and cannot be
> loaded." error.
> 
> Also, I have tested this on Nightly 15.0a1 and the issue is not
> reproducible. On this version I am able to click the "Back" button and then
> I am redirected to the new tab page, not to the "example.com" how Arni
> described in expected results.
> 
> I have performed a regression considering this is a good behavior. Here are
> the results:
> 
> Last good revision: 02b26fb307b4 (2012-07-10)
> First bad revision: 53860f11100c (2012-07-11)
> Pushlog:
> http://hg.mozilla.org/mozilla-central/
> pushloghtml?fromchange=02b26fb307b4&tochange=53860f11100c
> 
> Gijs, should we consider this a regression? I am asking you this because I
> have performed the regression on the assumption that the above described
> behavior was the correct one.

I mean, it's a regression, but it's a very old regression, it looks like... maybe from when we stopped using alert messages for these errors instead of the network error pages. Going back to about:newtab is wrong, we should be going to example.com . The fact that it changed in the window you found looks like it'd be because of bug 724239.

I don't think this needs to be very important as it is, as given that we don't load a search page, it's not very useful functionality as-is and so I don't think people will rely on it doing something sensible.