From temp3 at agosis.com  Thu Nov  4 19:44:20 2004
From: temp3 at agosis.com (Anthony Bautista Jr.)
Date: Thu, 4 Nov 2004 11:44:20 -0800 
Subject: [regex-coach] highlight suggestion
Message-ID: <00B53B0FE0C4D511A2F80004AC5B743504B48E@TPI_SERVER>

Greetings,
 
I am new to this mailing list & tool & have to see that it definitely has
potential to be a very helpful tool / coach.
 
In going through some basic examples, I have observed the following issue:
 
For...
 
Regular Expression:
a+
Target String:
Mary had a little lamb.
And everywhere that Mary
went, the lamb was sure
to go.
 
... and while 'g' checkbox is checked, only the first 'a' character is
highlighted yellow in the 'Target String' box.  It would be nice &
consistent with the 'Match from 1 to 2', 'Replace' tab for all the 'a'
characters to highlight whilst the 'g' (replace globally) checkbox is
checked.  My thoughts are that the 'g' checkbox should be a global setting
that will affect the 'Replace' tab & 'Split' tab.  That is, if the 'g' is
unchecked, then only the first match is replaced and split (and highlighted,
my main concern) consistently, but if checked, then all matches would be
hit.
 
FYI - I'd like to share my first success, an expression that can be used to
partially hide credit card numbers transmission in clear text over the net
(like for order / invoice confirmations etc...):
 
([^{1}])([0-9]{10}) or [^{1}][0-9]{10}
 
Anyhow, thanks in advance & I look forward to feedback re: my suggestion
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/regex-coach/attachments/20041104/f6cde7f4/attachment.html>

From veit.guna at gmx.de  Sun Nov  7 14:36:00 2004
From: veit.guna at gmx.de (Veit Guna)
Date: Sun, 07 Nov 2004 15:36:00 +0100
Subject: [regex-coach] [:alnum:]
Message-ID: <418E32D0.5000007@gmx.de>

Hi there.

First of all great job you've done there with RegEx Coach :)!

While I was testing some regular expressions for logcheck (a program to 
check logfiles e.g. syslog etc. :)), I noticed, that the expression 
[:alnum:] for alphanumeric characters (used with egrep) isn't recognized 
correctly. I think it's because CL-PPCRE is used, is that correct? Will 
there be alnum support in some way in the furure?

B.R.
Veit Guna





From edi at agharta.de  Tue Nov  9 09:46:01 2004
From: edi at agharta.de (Edi Weitz)
Date: Tue, 09 Nov 2004 10:46:01 +0100
Subject: [regex-coach] [:alnum:]
In-Reply-To: <418E32D0.5000007@gmx.de> (Veit Guna's message of "Sun, 07 Nov
	2004 15:36:00 +0100")
References: <418E32D0.5000007@gmx.de>
Message-ID: <uu0rzqrs6.fsf@agharta.de>

On Sun, 07 Nov 2004 15:36:00 +0100, Veit Guna <veit.guna at gmx.de> wrote:

> First of all great job you've done there with RegEx Coach :)!

Thanks... :)

> While I was testing some regular expressions for logcheck (a program
> to check logfiles e.g. syslog etc. :)), I noticed, that the
> expression [:alnum:] for alphanumeric characters (used with egrep)
> isn't recognized correctly. I think it's because CL-PPCRE is used,
> is that correct?

Yes.

> Will there be alnum support in some way in the furure?

Yes, I think so. It's not actually hard to do that (adding Unicode
character classes) but some diligence and time is needed which I don't
have at the moment. I'll probably add this to CL-PPCRE as soon as I
need it for some project. Once CL-PPCRE has it the Regex Coach will
get it as well.

Cheers,
Edi.




From edi at agharta.de  Tue Nov  9 09:47:11 2004
From: edi at agharta.de (Edi Weitz)
Date: Tue, 09 Nov 2004 10:47:11 +0100
Subject: [regex-coach] [:alnum:]
In-Reply-To: <uu0rzqrs6.fsf@agharta.de> (Edi Weitz's message of "Tue, 09 Nov
	2004 10:46:01 +0100")
References: <418E32D0.5000007@gmx.de> <uu0rzqrs6.fsf@agharta.de>
Message-ID: <umzxrqrq8.fsf@agharta.de>

On Tue, 09 Nov 2004 10:46:01 +0100, Edi Weitz <edi at agharta.de> wrote:

> Yes, I think so. It's not actually hard to do that (adding Unicode
> character classes) but some diligence and time is needed which I
> don't have at the moment. I'll probably add this to CL-PPCRE as soon
> as I need it for some project. Once CL-PPCRE has it the Regex Coach
> will get it as well.

I forgot: Clean patches to CL-PPCRE to implement this are welcome, of
course... :)




From edi at agharta.de  Tue Nov  9 09:49:42 2004
From: edi at agharta.de (Edi Weitz)
Date: Tue, 09 Nov 2004 10:49:42 +0100
Subject: [regex-coach] highlight suggestion
In-Reply-To: <00B53B0FE0C4D511A2F80004AC5B743504B48E@TPI_SERVER> (Anthony
	Bautista, Jr.'s message of "Thu, 4 Nov 2004 11:44:20 -0800")
References: <00B53B0FE0C4D511A2F80004AC5B743504B48E@TPI_SERVER>
Message-ID: <uis8fqrm1.fsf@agharta.de>

On Thu, 4 Nov 2004 11:44:20 -0800, "Anthony Bautista Jr." <temp3 at agosis.com> wrote:

> ... and while 'g' checkbox is checked, only the first 'a' character
> is highlighted yellow in the 'Target String' box.  It would be nice
> & consistent with the 'Match from 1 to 2', 'Replace' tab for all the
> 'a' characters to highlight whilst the 'g' (replace globally)
> checkbox is checked.  My thoughts are that the 'g' checkbox should
> be a global setting that will affect the 'Replace' tab & 'Split'
> tab.  That is, if the 'g' is unchecked, then only the first match is
> replaced and split (and highlighted, my main concern) consistently,
> but if checked, then all matches would be hit.

I agree with you that using the 'g' checkbox to toggle highlighting is
a good idea and I will probably implement this in a future
release. Thanks for the suggestion.

However, I don't think that this checkbox should change the behaviour
of 'split' - it's not consistent with Perl. If you only want the first
match to split use the 'limit' facility.

Cheers,
Edi.




From dennis at netstrata.com  Tue Nov  9 15:45:26 2004
From: dennis at netstrata.com (Dennis Williamson)
Date: Tue, 09 Nov 2004 09:45:26 -0600
Subject: [regex-coach] manual "recalc"
Message-ID: <6.1.2.0.2.20041109094315.0726ad20@mail.netstrata.com>

It would be nice to have a manual "recalc" mode so text and regexes could 
be entered without being evaluated until a button is clicked or "recalc" 
mode is turned back on.

Dennis







From edi at agharta.de  Tue Nov  9 16:24:09 2004
From: edi at agharta.de (Edi Weitz)
Date: Tue, 09 Nov 2004 17:24:09 +0100
Subject: [regex-coach] manual "recalc"
In-Reply-To: <6.1.2.0.2.20041109094315.0726ad20@mail.netstrata.com> (Dennis
	Williamson's message of "Tue, 09 Nov 2004 09:45:26 -0600")
References: <6.1.2.0.2.20041109094315.0726ad20@mail.netstrata.com>
Message-ID: <uu0rzat3q.fsf@agharta.de>

On Tue, 09 Nov 2004 09:45:26 -0600, Dennis Williamson <dennis at netstrata.com> wrote:

> It would be nice to have a manual "recalc" mode so text and regexes
> could be entered without being evaluated until a button is clicked
> or "recalc" mode is turned back on.

Well, for me one of the main features is that all state is instantly
updated and you don't have to "recalc" manually. Why would you want
that?

Cheers,
Edi.




From dennis at netstrata.com  Tue Nov  9 17:43:29 2004
From: dennis at netstrata.com (Dennis Williamson)
Date: Tue, 09 Nov 2004 11:43:29 -0600
Subject: [regex-coach] manual "recalc"
In-Reply-To: <uu0rzat3q.fsf@agharta.de>
References: <6.1.2.0.2.20041109094315.0726ad20@mail.netstrata.com>
	<uu0rzat3q.fsf@agharta.de>
Message-ID: <6.1.2.0.2.20041109112955.072aa550@mail.netstrata.com>

For some reason, my system bogs at times and it takes a while between 
keypresses waiting for Regex Coach to update. This problem is, of course, 
unrelated to RC, but it would make it easier sometimes if I could turn off 
the instant update even though I like it a lot, too.

I just tried it again. This time it was ultra-quick. I guess I need to 
diagnose the problem in my system.

Thanks for your time and consideration.

Dennis

At Tuesday 11/9/2004 10:24 AM, you wrote:
>On Tue, 09 Nov 2004 09:45:26 -0600, Dennis Williamson 
><dennis at netstrata.com> wrote:
>
> > It would be nice to have a manual "recalc" mode so text and regexes
> > could be entered without being evaluated until a button is clicked
> > or "recalc" mode is turned back on.
>
>Well, for me one of the main features is that all state is instantly
>updated and you don't have to "recalc" manually. Why would you want
>that?
>
>Cheers,
>Edi.







From rz10 at cornell.edu  Fri Nov 12 17:17:28 2004
From: rz10 at cornell.edu (Ray Zimmerman)
Date: Fri, 12 Nov 2004 12:17:28 -0500
Subject: [regex-coach] Mac OS X version bugs
Message-ID: <BA823BD2-34CE-11D9-9D0E-000393DC65EA@cornell.edu>

Looks like a great tool once the bugs are out ... I tried a simple case 
in the Mac OS X version ...

Reg-Ex:

	text\d

Target String:

	text0123text

It just says ...

	Unknown token FAKE-ENTRY in parse-tree

Also, it appears the cursor keys do not work for editing the reg-ex and 
target strings.

  - Ray




From edi at agharta.de  Sat Nov 13 00:17:31 2004
From: edi at agharta.de (Edi Weitz)
Date: Sat, 13 Nov 2004 01:17:31 +0100
Subject: [regex-coach] Mac OS X version bugs
In-Reply-To: <BA823BD2-34CE-11D9-9D0E-000393DC65EA@cornell.edu> (Ray
	Zimmerman's message of "Fri, 12 Nov 2004 12:17:28 -0500")
References: <BA823BD2-34CE-11D9-9D0E-000393DC65EA@cornell.edu>
Message-ID: <u4qjumwkk.fsf@agharta.de>

Hi!

My current problem with the Mac version is the lack of time. There's
an issue with getting certain threads in sync and I've not
investigated this fully. It doesn't help that I don't have a Mac
myself and have to lend my wife's iBook for that... :)

On Fri, 12 Nov 2004 12:17:28 -0500, Ray Zimmerman <rz10 at cornell.edu> wrote:

> Looks like a great tool once the bugs are out ... I tried a simple
> case in the Mac OS X version ...
>
> Reg-Ex:
>
> 	text\d
>
> Target String:
>
> 	text0123text
>
> It just says ...
>
> 	Unknown token FAKE-ENTRY in parse-tree

Yes, this is a consequence of the threading issues I mentioned
above. The program inserts certain "fake" entries into the parse tree
generated by the regex engine but they should have been removed at
this point. Unfortunately they aren't.

> Also, it appears the cursor keys do not work for editing the reg-ex
> and target strings.

Yes, I've noticed that, too. I guess this is only a minor problem that
should be easy to fix. But - see above - I haven't touched the Mac
version since I released the first beta. I hope I can look at it in
the next week.

Sorry,
Edi.




From spam at batz.org  Mon Nov 22 20:46:33 2004
From: spam at batz.org (spam at batz.org)
Date: Mon, 22 Nov 2004 21:46:33 +0100
Subject: [regex-coach] Regex Coach on Google
Message-ID: <E1CWL56-0000in-00@common-lisp.net>

hello dudes

if anyone cares
Regex Coach is No.1 for the keyword "regex" on google.com
http://www.google.com/search?hl=en&q=regex - if anyone wants to see it
congratulations edi 

greetings tim





From edi at agharta.de  Mon Nov 22 20:57:52 2004
From: edi at agharta.de (Edi Weitz)
Date: Mon, 22 Nov 2004 21:57:52 +0100
Subject: [regex-coach] Regex Coach on Google
In-Reply-To: <E1CWL56-0000in-00@common-lisp.net> (spam@batz.org's message of
	"Mon, 22 Nov 2004 21:46:33 +0100")
References: <E1CWL56-0000in-00@common-lisp.net>
Message-ID: <u653xppnz.fsf@agharta.de>

On Mon, 22 Nov 2004 21:46:33 +0100, "spam at batz.org" <spam at batz.org> wrote:

> Regex Coach is No.1 for the keyword "regex" on google.com
> http://www.google.com/search?hl=en&q=regex - if anyone wants to see
> it congratulations edi

Hey, cool... :)

Thanks,
Edi.




From snow_squall at hotmail.com  Wed Nov 24 19:05:39 2004
From: snow_squall at hotmail.com (Snow Squall)
Date: Wed, 24 Nov 2004 19:05:39 +0000
Subject: [regex-coach] Two Problems
Message-ID: <BAY101-F29F4C90B0ECFCB99FF727E83B80@phx.gbl>

Hello, first time poster...

I have two issues that i can't seem to track down....

1.  I'm looking for a rule that will eliminate the following.  I'm looking 
for all of my web pages that have some snippit of code before the 
<!DOCTYPE... my <!DOCTYPE should start the HTML on my web pages...  I've 
seen individuals sneak the following code in:

<!-- saved from -->
<!DOCTYPE HTML Public ..ect...

So is there a regex construct that will fail if any characters are found 
before <!DOCTYPE  ???

2. Secondly, looking to find ONLY the .PDF's inside a test.com domain.  I 
wish to match the pattern http://www.test.com/snow/squall/index.pdf .  I 
know to start my regex as http://www\.test\.com but how do i ignore all the 
directory stuff and key in on the .pdf extension.

Thanks





From edi at agharta.de  Wed Nov 24 19:47:12 2004
From: edi at agharta.de (Edi Weitz)
Date: Wed, 24 Nov 2004 20:47:12 +0100
Subject: [regex-coach] Two Problems
In-Reply-To: <BAY101-F29F4C90B0ECFCB99FF727E83B80@phx.gbl> (Snow Squall's
	message of "Wed, 24 Nov 2004 19:05:39 +0000")
References: <BAY101-F29F4C90B0ECFCB99FF727E83B80@phx.gbl>
Message-ID: <upt232fnj.fsf@agharta.de>

On Wed, 24 Nov 2004 19:05:39 +0000, "Snow Squall" <snow_squall at hotmail.com> wrote:

> I have two issues that i can't seem to track down....

These are really not on topic for this mailing list because it's about
the program Regex Coach, not about learning regular expression basics.

> 1.  I'm looking for a rule that will eliminate the following.  I'm
> looking for all of my web pages that have some snippit of code
> before the <!DOCTYPE... my <!DOCTYPE should start the HTML on my web
> pages...  I've seen individuals sneak the following code in:
>
> <!-- saved from -->
> <!DOCTYPE HTML Public ..ect...
>
> So is there a regex construct that will fail if any characters are
> found before <!DOCTYPE  ???

^<!DOCTYPE

This assumes that you've read the whole HTML page into one string

> 2. Secondly, looking to find ONLY the .PDF's inside a test.com
> domain.  I wish to match the pattern
> http://www.test.com/snow/squall/index.pdf .  I know to start my
> regex as http://www\.test\.com but how do i ignore all the directory
> stuff and key in on the .pdf extension.

http://www\.test\.com/.*\.pdf

Cheers,
Edi.