Re: Suggestion for Enhancement; B, H, ... and D

Sean had to ask,

| I've gotta ask:  why not have a regexp switch, such as a trailing slash
| operator?
|
| * ^Subject:.*sometext/D
|
| (or the converse: have the recipe flagged D, then use a case insensitive
| switch on the lines that should remain insensitive).

Because it's currently legal syntax for something else; therefore it would
make later binaries misread older rcfiles instead of being properly backward
compatible.

| Alternately, have a "sensitivity on" "sensitivty off" marker pair (and if
| we're going to introduce that, it'd make sense to ensure that the markers
| inherently support multiple flagging types, so that it can be easily
| extended in the future)?

Illustration, s'il vous plaît?

Your next suggestion looks good from my perspective as a user:

| The alternate BOL symbol would be doable - if that route is taken, the
| flags should have an explicit terminator:
|
| + BDjk + ^Subject:.*sometext
|
| Thus, additional flags could be introduced without necessitating a
| rethinking of the approach.  In the above example, FLAGS precede the
| second plus - if you were using a variable, it'd appear after the second
| plus, so letters wouldn't be confused.  Plus flagged recipe lines would
| ALWAYS have the plusses in pairs like that.
|
| OR, rules could be:
|
| ** BDjk * ^Subject:.*sometext
|
| The second asterisk appearing immedatley after the one at BOL would
| signify that this is the alternate type (or it could be a plus).  In this
| way, the basic "asterisk starts a condition" format would be retained,
| though it might lead to confusion with some users thinking it is a typo...

** might look like a typo there but *+ would not.  *? is already in use.

Another possibility would be two asterisks or two plus signs after the
single-condition flags, on the model of "var ??" now.  It could be

 * BDjk ** ^Subject:.*sometext
or
 * BDjk ++ ^Subject:.*sometext

The question is, other than B, D, and H, what current flags would make sense
there?  Digits, A, a, E, and e apply to all conditions if any, and can't
logically be appplied only to one; b, c, h, i, r, w, and W apply to the
action rather than to the conditions.  B and H can be applied to individual
conditions with B ??, H ??, and HB ??, so right now all this is about D and
only D.




_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail