1107864 - select tag doesn't update it's status on refresh

Status: RESOLVED INVALID

(Core :: DOM: Core & HTML, defect)

Description

[bert.bruynooghe]

User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:33.0) Gecko/20100101 Firefox/33.0
Build ID: 20141030112145

Steps to reproduce:

simple html being served:
<select>
 <option>1</option>
 <option selected>2</option>
</select>

https://dl.dropboxusercontent.com/u/16379342/test.html

Open the html in browser=> 2 is selected
change option to 1 in browser
refresh (not shift refresh)


Actual results:

get 304 (expected) and status of input is still option 1 (unexpected)


Expected results:

should have option 2 again, as that is what in the served html.

Comment 1

[Loic]

I'm not able to reproduce it with FF34 on Win 7, I got the option "2" after refreshing the page.

Could you force the update to FF34 in the options and test again.
If that fails, could you test with a clean profile (https://support.mozilla.org/en-US/kb/profile-manager-create-and-remove-firefox-profiles).

Comment 2

[bert.bruynooghe]

Sorry, it is not possible for me right now to update to FF34, as I currently need to write a workaround for these problems, and I need some means of testing those. Same problem occurs on other user's machines, so I doubt that creating a clean profile would solve the issue on FF33.

Comment 3

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

I see the behavior described (on Linux).  But I thought it was generally considered a feature that we preserve form state across reloads.  (Though I admit I generally don't want that behavior.)

Comment 4

[bert.bruynooghe]

Still seeing the same behaviour on FF34 on OSX. 
I agree that would maybe be a good feature to preserve state across reloads, but then it must be taken further. The issue I had was that there were input fields hidden/unnamed based on the value of the select (done by javascript), but that state was definitely reset. I had to add some code to have the page initialize the inputs upon reload, which then again caused FUOC.

Comment 5

[Boris Zbarsky [:bzbarsky]]

Value-restoration behavior for form controls on reload is purposeful, yes...