How to split on content-length?

How to split on content-length?

[Tom Koelman]

Hi,

I am using regular splitting to sort my mail in a lot of groups. Now I
would like to add a group containing all the "big" mails, with big
being defined as, say, bigger than 140k (Swen anyone?). I guess I
should write a function now to do that, because I don't think I can
use a regular expression to express that I want to split if
X-Content-Length: is followed by a number bigger dan 143360, but I
don't understand the doc. It says:

      The second element can also be a function.  In that case, it
      will be called narrowed to the headers with the first element of
      the rule as the argument.  It should return a non-`nil' value if
      it thinks that the mail belongs in that group.

Does this mean the second argument will contain the headers?

Tom

Re: How to split on content-length?

[Hanak David]

On Tue, 23 Sep 2003, Tom Koelman wrote:

>       The second element can also be a function.  In that case, it
>       will be called narrowed to the headers with the first element of
>       the rule as the argument.  It should return a non-`nil' value if
>       it thinks that the mail belongs in that group.
>
> Does this mean the second argument will contain the headers?

No.  There is only one argument.  However, it will be called with the
currently active buffer containing the mail, narrowed to the header parts.
To narrow to the body part, call widen followed by gnus-narrow-to-body.

David

Re: How to split on content-length?

[Jesper Harder]

Tom Koelman <tkoelman@xs4all.nl> writes:

> I am using regular splitting to sort my mail in a lot of groups. Now I
> would like to add a group containing all the "big" mails, with big
> being defined as, say, bigger than 140k (Swen anyone?). I guess I
> should write a function now to do that, because I don't think I can
> use a regular expression to express that I want to split if
> X-Content-Length: is followed by a number bigger dan 143360

You /could/ use a regexp, e.g.

    [1-9][4-9][0-9]\\{4,\\}

which matches any number greater than 139999.

Re: How to split on content-length?

[Tom Koelman]

Jesper Harder <harder@myrealbox.com> writes:

> Tom Koelman <tkoelman@xs4all.nl> writes:
>
>> use a regular expression to express that I want to split if
>> X-Content-Length: is followed by a number bigger dan 143360
>
> You /could/ use a regexp, e.g.
>
>     [1-9][4-9][0-9]\\{4,\\}
>
> which matches any number greater than 139999.

I don't think this will match anything for wich the last 5 digits are
a number between 00000 and 39999. But you gave me an idea and
I modified it to

\\(1[4-9]\\|[2-9][0-9]\\)[0-9]\\{4,\\}

which still has a problem with numbers containing more than 6 digits
of which the first is a 1 and the second smaller than 4. Adding
another bit like so

\\(1[4-9]\\|[2-9][0-9]\\|[1-9][0-9][0-9]\\)[0-9]\\{4,\\}

does the trick.

Thanks,
Tom

Re: How to split on content-length?

[Ted Zlatanov]

On Wed, 24 Sep 2003, tkoelman@xs4all.nl wrote:

> I don't think this will match anything for wich the last 5 digits
> are a number between 00000 and 39999. But you gave me an idea and I
> modified it to
> 
> \\(1[4-9]\\|[2-9][0-9]\\)[0-9]\\{4,\\}
> 
> which still has a problem with numbers containing more than 6 digits
> of which the first is a 1 and the second smaller than 4. Adding
> another bit like so
> 
> \\(1[4-9]\\|[2-9][0-9]\\|[1-9][0-9][0-9]\\)[0-9]\\{4,\\}
> 
> does the trick.

That seems unnecessarily acrobatic.  Why not just match the digits and
compare them numerically to the number you want?  You can use
string-to-number, something like this might work (set data to the
header of interest and the action to what you want instead of the
message):

(let ((data "19874543435")
      (limit 150000))
  (when (and (string-match "\\([0-9]+\\)" data)
	     (> (string-to-number (match-string 1 data)) limit))
    (message "saw number greater than %d" limit)))

I would be very careful about using content-length for filtering.
Maybe you should consider doing that only if the message has an
executable attachment.

Ted