86261 - Browser's "File -> Send Page" should use original cached source, not refetch.

Status: NEW

(SeaMonkey :: MailNews: Composition, defect)

Description

[Aleksey Nogin]

Reproducible: always

Steps to reproduce:
 1) Go to http://ntiaotiant2.ntia.doc.gov/test/mozillatest.html and submit a form
 2) Use "File -> Send Page" to e-mail the resulting page somewhere

Expected result: the contents being viewed are mailed.

Actual results: the page is refetched before it is mailed.

This is especially nasty since normally you do not get to see the refetched page
you are forwarding. If for some reason the refetched page is very different from
the original, it may be a very nasty surprise. Once I thought I was forwarding
my collegue a submission confirmation page, but in fact I forwarded an error
message saying the submission failed!

This bug also existed in Netscape 4.

P.S. Test case http://ntiaotiant2.ntia.doc.gov/test/mozillatest.html was
provided by [e-mail] for bug 55583

Comment 1

[Sebastian Leske]

The same behavious seems to occur if you save a page (File / Save As ...).
The browser will not save the page you are seeing, but refetch it.
That's bad for at least two reasons:

1) Refetching the page may be slow (esp. on dialup)
2) Sometimes the newly fetched page may be different. E.g., I do online
banking (via SSL), and I can't save any pages there (such as an account
statement) because the pages cannot be refetched (hence saving only saves
an error message).
To check this for yourself,
go to my banking site,
https://www.postbank-banking.de/

Log into the demo account described left of the login form
(Kontonummer 9 999 999 999 , PIN 11111, all without spaces)
select Funktion: Kontoauszug

In the next screen, just klick on the little green OK (bottom right).
You will now see your account statement.
Try saving it (right click / Save Frame As...). Saving will proceed without
apparent problems, but the file saved will only contain an error message.

Comment 2

[Aleksey Nogin]

Sebastian, you are talking about a separate issue - bug 84106. Both depend on
bug 40867, but they are still different bugs.

Comment 3

[scottputterman]

Jean-Francois, are you the right owner for this?  It sounds like Save Page
exhibits the same problem.

Comment 4

[Jean-Francois Ducarroz]

I don't know! all I do is to fetch the url I receive. Do we have a way to
spefify to not reload the url? 

Comment 5

[Mark Bitterling]

Will this automatically be resolved by the recently landed patch for bug 40867
or is there something additional that needs to be done to take advantage of the
new patch?

If this gets resolved in time for 1.0 that would be awesome!

Comment 6

[P Wagland]

I just did a test on this, and it appears that the resolution to bug 40867 does
not solve this problem entirely. The simplest way to test this is just to send
yourself a copy of this page. It goes to the network to get the page. It will
however send a static page without going back to the network. So that is at
least a start :-)

Comment 7

[Anant Sahai]

I can confirm this bug in 1.0rc2.

Curiously, if I hit "Save page as" from a dynamic page (in my case a credit card
payment confirmation), it saves to disk just fine. I can even do it repeatedly
as text, html, complete, etc. 

But if I hit "Send page" from the same page, it sends the wrong thing as an
attachment. Apparently it still tries to refetch it and sends the resulting
error message on as the attachment (silently).

Comment 8

[Gregg Meehan]

changing QA contact from Sheela to Gregg

Comment 9

[Jesse Ruderman]

Another testcase: http://www.random.org/cgi-bin/randseq?min=1&max=100.

This is silent dataloss.

Comment 10

[Mike Fedyk]

This is not only linux.  Please change platform to all.

Comment 11

[Brant Gurganus]

By the definitions on <http://bugzilla.mozilla.org/bug_status.html#severity> and
<http://bugzilla.mozilla.org/enter_bug.cgi?format=guided>, crashing and dataloss
bugs are of critical or possibly higher severity.  Only changing open bugs to
minimize unnecessary spam.  Keywords to trigger this would be crash, topcrash,
topcrash+, zt4newcrash, dataloss.

Comment 12

[R.K.Aa.]

*** Bug 199153 has been marked as a duplicate of this bug. ***

Comment 13

[Asa Dotzler [:asa]]

*** Bug 159106 has been marked as a duplicate of this bug. ***

Comment 14

[Daniel B.]

Note bug 288462.

Comment 15

[Daniel B.]

> If you want this to be a tracking bug, use the "depends" and "blocks" fields...
> Please don't just comment on all sorts of other bugs pointing to this one.

Another question:

Should I use "blocks", to mean that fixing all the other bugs right depends
on understanding the intended behavior and refining the architecture to 
support it?

Or should I use "depends on" to mean that all the other bugs needs to be fixed
for this one to be fully satisfied.

Thanks,
Daniel

Comment 16

[Daniel B.]

I wrote:

> Another question: ...

Oops.  Never mind (meant for a different report).  Bad Bugzilla UI.

Daniel

Comment 17

[Mark Banner (:standard8)]

Only SeaMonkey has a send page option... (TB / Firefox use send link)