Closed Bug 9202 Opened 26 years ago Closed 25 years ago

When replying, remove properly formed signatures

Categories

(MailNews Core :: MIME, enhancement, P3)

Product:
MailNews Core
Component:
MIME
Type:
enhancement

Tracking

(Not tracked)

Status:
VERIFIED FIXED

People

(Reporter: phil, Assigned: BenB)

References

(Blocks 1 open bug)

Details

(Whiteboard: Fixed. Patch in bug 31906.) 

It would be nice if we were smart enough so that when quoting a message, we
would remove a properly formed signature, which would reduce noise in quoting. I
think we've talked to the editor folks about this -- cc'ing sfraser.

Reasonable criteria for this seem like:
- signature must begin with "-- " at the beginning of a line
- signature must be within N lines of end of msg (N == 4 or 5?)
- result of signature snipping must not be an empty document
- signature snipping should be disable-able using a hidden JS pref
*** Bug 3747 has been marked as a duplicate of this bug. ***
Status: NEW → ASSIGNED
Target Milestone: M9
This might not be too hard to do, but with the use of more and more rich text
email products, I don't think this is a big a "bang for the buck" as a few
years ago.

- rhp
This is a big bang for buck for news users.
Target Milestone: M9 → M10
Necko is going to chew up the rest of M9.

- rhp
Target Milestone: M10 → M14
RFE. M14.
Blocks: gnksa
Max. lines of sig: 4
Target Milestone: M14 → M20
Would be a nice net contribution.

- rhp
I just submitted a couple of programs to Sven Guckes sigsupport page, which

pushes the total number of programs listed that will remove the sig when replying

to an even 3 dozen.  So here's the URL: <http://www.math.fu-berlin.de/~guckes/

afw/#sigsupport>
QA Contact: lchiang → pmock
Changing qa assigned to pmock@netscape.com.
*** Bug 29120 has been marked as a duplicate of this bug. ***
*** Bug 30206 has been marked as a duplicate of this bug. ***
I fixed this. It will propably land together with bug #31906. REASSIGNing to
myself. Rich, please redo, if you disagree.
Assignee: rhp → mozilla
Status: ASSIGNED → NEW
Component: Composition → MIME
accept
Status: NEW → ASSIGNED
> I fixed this.

Well, not completely. I forgot format=flowed, and I can't get a clue about this
code; it doesn't want to work.
Also,

Rich,
is it save to match "-- \n" for "a line consisting only of "-- ""? Worksforme on
Linux, but I'm not sure about Win32.
Woohoo! Thanks for looking into this Ben. Now I might actually start using a
signature :-)

If this is in the content tree, dom linebreaks are \n, regardless of which
platform we're running on, and \r is illegal.

If you're doing your matches outside the content tree, then the question is
whether someone has already substituted platform line endings for dom line
endings.  This is likely -- the output system will make that substitution.  So
in that case, you might want to build up a match pattern consisting of "-- " +
NS_LINEBREAK (from nsCRT.h) and then match on that.
I haven't checked how the signature removal works, but remember that if a message contains more than one '-- ', the *last* one is the signature delimiter:

A "personal signature" is a short closing text automatically added to the end of articles by posting agents, identifying the poster and giving his network addresses, etc. If a poster or posting agent does append such a signature to an article, it MUST be preceded with a delimiter line containing (only) two hyphens (ASCII 45) followed by one SP (ASCII 32). The signature is considered to extend from the last occurrence of that delimiter up to the end of the article (or up to the end of the part in the case of a multipart MIME body). Followup agents, when incorporating quoted text from a precursor, SHOULD NOT include the signature in the quotation. Posting agents SHOULD discourage (at least with a warning) signatures of excessive length (4 lines is a commonly accepted limit).

Perhaps this is how it *is* implemented now, but I just wanted to make sure you knew about it.
huftis,
- I see no way, how I could implement this. libmime is line-oriented.
- What is the source of your quote? It is not in GNKSA 2.0, including its quote
of "son-of-1036" (I don't know, which # this rfc has).
- I see no sense in it. If a user adds a sig and a mailinglist bot does the
same, I want both sigs to be removed.

Re format=flowed:
Bad news: There *is* a sig recognition in the flowed code, but it doesn't seem
to work. As I said, I feel unable to debug it. I can try it again, but don't see
much chance of success. I filed bug 34455 about it, but as long as Daniel
Bratell doesn't volunteer to fix it, I can't fix this (9202) bug for flowed msgs
:-(.
Depends on: 34455
As far as I can tell there is no "majority" on which sigdash to use (first or 

last).  And in practice, it's only an issue when a second sigdash has been 

added by a moderator or injector (i.e. news.newusers.questions or 

RemarQ) in which case you want to use the first anyway.



And although people should keep their sigs reasonably short, this isn't 

the place to do anything about that.



I.e. don't bother looking for a different way of handling it.  Stopping at the 

first is quite acceptable.
I assume we wouldn't be stopping at the first that's inside a quotation, but
only at the first that's at the "outer level", correct? 
akk, yes.
Rich, what line endings (see above) do I get in mimetpla?
Target Milestone: M20 → M16
Hi Ben,
mimetpla.cpp will typically give you CRLF output for line endings.

- rhp
The problems with format=flowed didn't matter at all currently, because
currently, the class for flowed is never used for quoting (filed bug 39370). The
normal plain text converter is used, which can remove the sig. I could have
saved myself *so* many hours (days?), if I noticed that earlier :-(. Removing
dependency on bug 34455.

Anyway, it works now!

There is still the vertical whitespace between the normal text and the former
sig, but I'll propably remove this for another (to-be-filed) bug anyway.
No longer depends on: 34455
Whiteboard: Fixed. Patch in bug 31906.
Great! Thanks Ben!

M17 :(
Depends on: 31906
Target Milestone: M16 → M17
Fix checked in by rhp.
Status: ASSIGNED → RESOLVED
Closed: 25 years ago
Resolution: --- → FIXED
Verified as fixed win32, linux, and macos under "Plain Text" using the following 
builds:
 Win32 commercial seamonkey build 2000-062509-m17 installed on P500 Win98
 Linux commercial seamonkey build 2000-062509-m17 installed on P200 RedHat 6.1
 MacOS commercial seamonkey build 2000-062508-m17 installed on G3/400 OS 9.04

Using a plain text compose window, it correctly removes the signature.  Using a 
html compose window, it does not correctly removes the signature.  It actually 
puts the signature in the middle of the quoted signature.
I will log a separate bug on this HTML quoting issue.
Status: RESOLVED → VERIFIED
Bug 43829 about sigs in HTML.
Where is the js pref to disable sig stripping?

Or did this slip through between the original description and the implementation?
Product: MailNews → Core
(In reply to comment #30)
> Where is the js pref to disable sig stripping?
> 
> Or did this slip through between the original description and the implementation?

Bump?  Still no idea where I can disable this using a js pref.
(In reply to comment #31)
> Bump?  Still no idea where I can disable this using a js pref.

Who said it was possible to disable it with a pref?
no-one :) It was just in the initial "requirements", and I'd like it :)
Product: Core → MailNews Core
You need to log in before you can comment on or make changes to this bug.