Re: [PATCH] Improve f=f interoperability (was: [PATCH] fix

[Johannes Stezenbach <js@xxxxxxxxx>]

Hi,

On Tue, Oct 23, 2007, Rocco Rutte wrote:
>> On Thu, Oct 18, 2007, Rocco Rutte wrote:
>
>>>    original: flowed? | reply: flowed? | trim spaces
>>> 1          N                 N               N
>>> 2          N                 Y               Y
>>> 3          Y                 N               Y
>>> 4          Y                 Y               N
>
>>> 1: fixed reply to fixed mail: no problem
>>> 2: flowed reply to fixed mail: your example, problem
>>> 3: fixed reply to flowed mail: potential problem later on, esp. with 
>>> other MUAs
>>> 4: flowed reply to flowed mail: no problem
>
> Attached is a patch which is a first take at implementing the above matrix. 
> I only tested each case once but it seems to work.

I also tested each case briefly and it works for me. I'm very happy
with it.

> Now that I look at the changes, the behavior differs quite a lot compare to 
> the current one. Maybe we should only use these changes depending on 
> $flowed_compat (or something like that) defaulting to 'no' (to keep the 
> current behavior as default)?

IMHO the default behaviour should be the one which avoids
interoperability problems. I suspect most users will use
the default settings and don't want think about the details
(and they attribute problems to format=flawed, not to
implementation or configuration issues).

But an option to disable the space stripping would be good.
(E.g. if someone wants to repair the quoting manually to
create proper format=flowed replies).

> +    if (option (OPTTEXTFLOWED) && l >= 3 && ascii_strncmp (buf + l - 3, "-- 
> ", 3) != 0)
> +    {
> +      /* when generating format=flowed from format=fixed,
> +       * strip all trailing spaces except for the signature
> +       * separator to improve interoperability */

I don't get the signature separator test -- why not just
ascii_strcmp(buf, "-- ")? Wouldn't your code treat "foo-- "
as a signature separator?


Thanks,
Johannes