253704 - Enter key in form differs in action from using Search Button to submit

Status: RESOLVED DUPLICATE of bug 156683

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

Description

[Peter]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.6) Gecko/20040206 Firefox/0.8
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.6) Gecko/20040206 Firefox/0.8

On the 'search' tab, I type my search term in and press enter. Surprisingly, the
display switches to a different "tab" (the 'map' tab). It behaves correctly if I
click on the button. This URL also behaves correctly in Opera & IE.

Reproducible: Always
Steps to Reproduce:
1. Go to URL
2. Select "search" tab
3. Type in your search term (e.g. London)
4. Press enter

Actual Results:  
Display switched to the map tab.

Expected Results:  
Return should be the same as clicking on the "Search" button - it should bring
up a list of results.

Comment 1

[jon rekai]

No submit button was explicitly activated, so the first one on the form was
used.  That's imgSelectMap.

The others may be using the first submit button immediately following the form
control in which enter was pressed, which /is/ the search button
(PortSelectionSearch:butSearch:_ctl0).

Comment 2

[Hermann Schwab]

wfm Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.8a2) Gecko/20040714

I don´t see a Java applet on this site, only some Javascript. Please make a
clean install of a current Firefox, or the next release, and retry before
installing extensions.

Comment 3

[jon rekai]

> I don´t see a Java applet on this site, only some Javascript. Please make a

Thx, hhschwab.  There's no java applet I saw either (failed to note that with
previous comment).

Might the form have been incorrectly identified as java?

Based on the indicated actual and expected results, the 0th button was chosen as
the default, but the next button should have been (on a form with a single text
area).

Comment 4

[Hermann Schwab]

sorry for the spam, was wrong in comment #2

Also seems I´ve overlooked comment #1, otherwise I wouldn´t have commented.
Seems comment #1 explains all.

replaced Summary with hopefully a clarifying one.
original summary: enter key in java applet doesn't select default button

After looking into the code, I used LiveHTTPheaders to see what was sent.
Excerpts from log made by LHTTP 0.8: http://livehttpheaders.mozdev.org/

using Enter:
Content-Type: application/x-www-form-urlencoded
Content-Length: 3397
__EVENTTARGET=&__EVENTARGUMENT=&__IMAGEMAP_AREA=-1&__IMAGEMAP_X=-1&__IMAGEMAP_Y=-1&__IMAGEMAP_ID=-1&__VIEWSTATE=...veryLongString...&imgSelectSearch.x=51&imgSelectSearch.y=9&SENCImagePortClicked=

using the button:
Content-Type: application/x-www-form-urlencoded
Content-Length: 1847
__VIEWSTATE=...veryLongString...


re comment #3
Might the form have been incorrectly identified as java?

No I don´t think so, if an applet is encountered, the java runtime is started,
and I see a Java Icon in the tray, until I close the browser completely.
Some people are thinking Java is the shortform of Javascript, or vice versa.
Guess thats due to Netscape, they wanted to profit from the JAVA name, so Java
and Javascript are completely different things.

What would be the correct Component or Resolution for this bug?

HTML: Form Summission seems better than LAYOUT: Form Controls.

If I believe your explanation, this bug should be resolved invalid,
but other browsers are doing it the way the reporter wants.

OPERA 7.50 is working the way the reporter wants.
Netscape4.79 looks very broken, can submit, but doesn´t get the results, is
endless loading.
 

Comment 5

[jon rekai]

HTML: Form Summission sounds appropriate.

Whether sending the 0th element or the next one (for a form with a single text
field and image submits) is the current correct behavior, I'm not sure.  The
HTML spec doesn't define form submission via the enter key, but it would appear
the majority favors the expected behavior (which Dillo browser also exhibits).

(Which submit button to activate on behalf of the enter key and whether to send
its name/value appears to be an ongoing issue, e.g. bug 104211.)

Comment 6

[Boris Zbarsky [:bzbarsky]]

*** This bug has been marked as a duplicate of 156683 ***