Re: [Gnu-arch-users] Preventing matches in regular expressions

[Aaron Bentley]

Tom Lord wrote:

> So then this command:
>
>     > $fai categorize --local --unrecognized fox.zip
>
> means something like:
>
>     "Let the rules I've specified so far be called X.
>  
>      The new rules are: 
> 	If the file is fox.zip it's unrecognized,
>         otherwise, things work just like under rules X."
>
> Is that really the idea?

Yes.  Worse than that, I'd like to have specific-name rules have a 
higher precedence than file-class rules.

> I don't know of an easy way to do that without modifying tla.

> The problem here is that the per-file activities of the `inventory'
> loop are performance-critical.   It's already bad that we cal
> `regexec' several times.   One really doesn't want to make that loop
> any more costly.

I'm in full agreement there.  If I had to choose, I'd choose more speed 
over more flexability.

> In rx, you can write:
>
>     R1[[:cut 1:]]|R2[[:cut 2:]]|R3[[:cut 3:]]| ...
>
> and a single call to regexec will tell you which regexp matched.
> (Rx `regexec' returns an extra value -- the "state label" which
> can be set by using the `cut' operator.)

That bears a striking resemblance to the inventory optimization work I 
did with the cut operator.  This might work nicely also:

R1[[:cut 3:]]|R2[[:cut 1:]]|R3[:cut 3:]

As far as I can tell, reusing cut numbers is legal, and we could then 
assign 1 to junk, 2 to backup, 3 to precious, etc.

> So, for your fai feature, you could:
>
>   Propose a new alternative to =tagging-method files (or new 
>   syntax to go inside of them), allowing for directives 
>   in which the ordering of directives is significant.

Okay, I'll consider that.  The current semantics are already very close 
to what I want, because .arch-inventory tends to contain specific-file 
regexes, while =tagging-method tends to contain file-class regexes.

Since I'm trying to automate things, I've been trying to look at every 
case, which is why I was looking for a way to handle cases when the 
regexes appear in the same files.

I think file-class in .arch-inventory happens often enough to worry 
about, but I'm not sure whether it justifies rewriting the inventory 
code.  I don't think there's a such thing as file-specific code in 
=tagging-method.

That is, in =tagging-method, a rule like

precious ^\.listing$

defines a class of files that all have the same name.  Whereas the same 
rule in .arch-inventory would refer to a specific file.

>   Modify the inventory code to use the rx optimization. 
>   As a side effect, you'll probably make `tla inventory' faster.

Well, first we need to fix the cut operator handling.

Aaron

--
Aaron Bentley
Director of Technology
Panometrics, Inc.