Closed Bug 302617 Opened 19 years ago Closed 17 years ago

view source is not same that the served page

Categories

(Toolkit :: View Source, defect)

Product:
Toolkit
Component:
View Source
Platform:
x86
Windows XP
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED WORKSFORME

People

(Reporter: mail, Unassigned)

References

(
URL
)

Details

(Whiteboard: DUPEME?) 

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b3) Gecko/20050712 Firefox/1.0+
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b3) Gecko/20050712 Firefox/1.0+

By server side scripting, page is served as application/xhtml+xml for browser
that support it, and as text/html for browser that don't support it.
View source represent the code served as text/html instead of the one served as
application/xhtml+xml

Reproducible: Always

Steps to Reproduce:
1. Open url www.fruibile.it
2. Right click on page and click on "View Page Source" (show the code as served
as text/html).
3. Right click on page and click "View page info" (shows that is correctly
served as application/xhtml).
4. Try to save the page with "File - Save page As" (save the page in a mixed
version of application/xhtml+xml and text/html

Actual Results:  
Show wrong code

Expected Results:  
Show right code as parsed by browser
Possibly a dupe of bug 225979?
It persist also with Deer Park Alpha 2.
I can see this in a frame that is server generated.
I reload the frame but it is still the old source.
The frame shows the correct content in the browsing window.

using
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a1) Gecko/20050821
Firefox/1.6a1
Yes, u need to reload the page before Mozilla update the right content.
If u use another browser that support application/xhtml like Opera this problem 
don't is available.
This bug occurs with the latest netscape, whether you use IE or Firefox as the
rendering engine.  It also occurs with firefox itself in the 1.0.7 I'm using,
and the dearpark.  I know that they are all basically "fixing" the code I guess
before they are being rendered, and what the "view source" is doing is
outputting the "fixed" code, and not the original.  This is bad, because you
don't know if you've made errors, and a lot of people like me use firefox for
development, and if we're using a scripting language we need to see the RAW
output, not the fixed stuff.  

Therefore a good way to fix this, may be to under the "View" menu to put, "view
as original" or something like that.  This preference either should be
automatically saved, or/and be in the options as a permanent preference.  

For now, I'm not using firefox or netscape for development anymore, because it
can lead to problems.  
(In reply to comment #0)
> User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b3)
Gecko/20050712 Firefox/1.0+
> Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b3)
Gecko/20050712 Firefox/1.0+
> 
> By server side scripting, page is served as application/xhtml+xml for browser
> that support it, and as text/html for browser that don't support it.
> View source represent the code served as text/html instead of the one served as
> application/xhtml+xml
> 
> Reproducible: Always
> 
> Steps to Reproduce:
> 1. Open url www.fruibile.it
> 2. Right click on page and click on "View Page Source" (show the code as served
> as text/html).
> 3. Right click on page and click "View page info" (shows that is correctly
> served as application/xhtml).
> 4. Try to save the page with "File - Save page As" (save the page in a mixed
> version of application/xhtml+xml and text/html
> 
> Actual Results:  
> Show wrong code
> 
> Expected Results:  
> Show right code as parsed by browser


For example go to:
    http://www.sunraystudios.com/misc/bug.html

The ACTUAL source=>
    <html>
	    <script>unclosed
    </html>

the VIEW SOURCE=>
    <html>
            <script></script>unclosed
    </html>

the File=>Save Page As results=>
    <html>
	    <script>unclosed
    </html>

Therefore the issue is that it is trying to output the DOM-compliant stuff,
instead of the actual source.
I'd say severity MAJOR as viewing source is so important.

The first idea of viewing the original source as 'view original' is wrong - there should be one view source which shows always the original source.

A button could be placed on the view source pane to format, and another to 'tidy' the html tags around, using the identical code that firefox now uses to fix the code in the current broken view source.

If someone wants html tidy to be implemented, this shoudl yet again be another option, but don't hide the current firefox code fixing, because that is also important (to show how firefox fixes and interprets given code)
I also have this issue. For me, I'm trying to view the source code of a restricted "login-required" page:

/members-only/

...but the source code is displaying the login page where the browser is directed to if there is no logged-in session:

/login.php

I cannot find a way of viewing the source of a restrited page!

Using FF 1.5.

Quinn
It's severity is MAJOR ... when I select "view source", I want to see SOURCE, not FireFox's interpretation of what the source should have been. 

Changing the menu item to say: "View Corrected Source", or adding a warning that FF has possibly changed the source, would be the fast fix, but not being able to see the original source makes FF useless as for debugging simple HTML.

This has already caused me a lot of problems: we were blaming translation software for corrupting HTML when the problem was a stray quote inside an IMG tag in the English source.  There was no visible problem in the source, when viewed in FF.  Later, someone looked at the source with MSIE and found the error.

It's severity is MAJOR ... when I select "view source", I want to see SOURCE, not FireFox's interpretation of what the source should have been. 

Changing the menu item to say: "View Corrected Source", or adding a warning that FF has possibly changed the source, would be the fast fix, but not being able to see the original source makes FF useless as for debugging simple HTML.

This has already caused me a lot of problems: we were blaming translation software for corrupting HTML when the problem was a stray quote inside an IMG tag in the English source.  There was no visible problem in the source, when viewed in FF.  Later, someone looked at the source with MSIE and found the error.
There should be 3 View / Save Page options:

(please also see link to CSS saving bugs).

Save / View source
   - Save / View the files that were sent from the server

Save / View FireFox render source
   - Save / View a source render of FireFox's original DOM build of this source

Save / View Page Snapshot
   - Save / View a source render of FireFox's CURRENT DOM snapshot, including anything that may have been altered by client side scripting, written into the page using doc.write and ajax type stuff.

The first should be what is shown. IE shows original source, but KILLS you by saving it's own version of the source.

Also, view source should NEVER EVER make a second request:
  - This is a security risk (you are sending a http request which is information from the computer)
  - This could cause problems with the current page (view source on a credit card payment page).

This is CRITICAL, BLOCKING and CONFIRMED (even though I broke etiquette by talking about a few inter-related points here).

Please update accordingly.
I haven't had this issue anymore.  If you go to the page:
http://www.sunraystudios.com/misc/bug.html

It gives you the original source, not the fixed source.  So has been fixed in 1.5 for quite some time by now, at least from what I can see.

As far as seeing the current DOM snapshot, this would be nice if it were built-in, because I use "view as rendered source" extension for Firefox when doing AJAX/DHTML stuff, but it would be more convenient if it were just a built-in feature of the view source.
I'm still having this issue with a page I wrote that has a php function preventing non-logged in users from viewing the page. Even though I am logged in, the View Source shows the code of the redirected login page. I am assuming Camino is not serving up the cookies the page needs.
Whiteboard: DUPEME?
Can you attach a test case of the page you are having the problem with or point us to a URL? I'm not seeing this in the current trunk builds or in the 2.0.0.4 release. 
Well, I had posted an example at:  http://www.sunraystudios.com/misc/bug.html
, but as I had noted, in Comment#12, the issue has been fixed for over a year.  I checked again in the most recent version just now, and the source is the same as what is the actual source, so the fix still holds.
Yeah, I looked at the page and it looked good but it wasn't entirely self-explanatory. I'll resolve this as fixed.

Thanks!
Status: UNCONFIRMED → RESOLVED
Closed: 17 years ago
Resolution: --- → FIXED
It wasn't supposed to be self-explanatory, that's why I had commented here.  This should have been set to resolved a year ago, as I noted, a year ago :>)
FIXED is reserved for bugs that were fixed by a specific, identifiable patch. Since we're not sure what code change fixed this, it should be resolved as WORKSFORME.

->WORKSFORME
Resolution: FIXED → WORKSFORME
Product: Firefox → Toolkit
You need to log in before you can comment on or make changes to this bug.