<html>
<head>
<meta http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<meta name=Generator content="Microsoft Word 11 (filtered)">
<style>
<!--
/* Font Definitions */
@font-face
{font-family:Tahoma;
panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:10.0pt;
font-family:Arial;}
a:link, span.MsoHyperlink
{color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{color:blue;
text-decoration:underline;}
span.Typed
{font-family:"Courier New";}
span.tty
{font-family:"Courier New";}
span.Name
{font-style:italic;}
span.Variable
{font-family:"Times New Roman";
font-style:italic;}
span.EmailStyle22
{font-family:Arial;
color:navy;}
@page Section1
{size:8.5in 11.0in;
margin:1.0in 1.25in 1.0in 1.25in;}
div.Section1
{page:Section1;}
-->
</style>
</head>
<body lang=EN-US link=blue vlink=blue>
<div class=Section1>
<p class=MsoNormal><span style='color:navy'>How do you create a module on the
fly? Wouldn't you have to write it to disk and then compile and load it? Or
is there another way?</span></p>
<p class=MsoNormal><span style='color:navy'> </span></p>
<p class=MsoNormal><span style='color:navy'>Cheers,</span></p>
<p class=MsoNormal><span style='color:navy'> </span></p>
<p class=MsoNormal><span style='color:navy'>David</span></p>
<p class=MsoNormal><span style='color:navy'> </span></p>
<div>
<div class=MsoNormal align=center style='text-align:center'>
<hr size=2 width="100%" align=center tabindex=-1>
</div>
<p class=MsoNormal><b><span style='font-family:Tahoma'>From:</span></b><span
style='font-family:Tahoma'> erlang-questions-bounces@erlang.org
[mailto:erlang-questions-bounces@erlang.org] <b>On Behalf Of </b>Robert Virding<br>
<b>Sent:</b> Wednesday, October 03, 2007 16:32<br>
<b>To:</b> Yariv Sadan<br>
<b>Cc:</b> erlang-questions@erlang.org; G Bulmer<br>
<b>Subject:</b> Re: [erlang-questions] Regular expression library (was Not
anErlang fan)</span></p>
</div>
<p class=MsoNormal> </p>
<p class=MsoNormal style='margin-bottom:12.0pt'>That's not too difficult to do
at all. The speed you get would depend on which regexp family you do it for.<br>
<br>
Robert</p>
<div>
<p class=MsoNormal><span class=gmailquote>On 27/09/2007, <b>Yariv Sadan</b>
<<a href="mailto:yarivsadan@gmail.com">yarivsadan@gmail.com</a>> wrote:</span></p>
<p class=MsoNormal>I was wondering if it would be possible/beneficial to write
a regexp <br>
compiler that takes a regexp string and compiles it into an Erlang<br>
module.<br>
<br>
You would use it as follows:<br>
<br>
regexp2:compile("f*o+o?", myregexp, [native]),<br>
true = myregexp:matches("foo"). <br>
<br>
This could be a very fast regexp engine.<br>
<br>
What do you think?<br>
<br>
Yariv<br>
<br>
On 9/26/07, Robert Virding <<a href="mailto:rvirding@gmail.com">rvirding@gmail.com</a>>
wrote:<br>
> The regular expression module which is included in the distribution is a <br>
> simple rather naïve one written in Erlang which uses a back-tracking<br>
> algorithm directly on the AST. It has all the problems Russ Cox mentioned
in<br>
> his paper. I wrote it long ago when no one was interested in parsing
regexps <br>
> in Erlang. :-)<br>
><br>
> I am rewriting it at the moment, still in Erlang but using the same<br>
> algorithm, or similar to the one, which Russ describes. It is about the
same<br>
> speed as the existing one for simpler regexps but it can handle all the <br>
> pathological cases as it should. It will handle POSIX syntax and semantics<br>
> (but no back-references*). There will also be a version which will
implement<br>
> a subset of PERL regexps with PERL semantics. Both will work on either
lists <br>
> or binaries.<br>
><br>
> I hadn't seen the lambda-the-ultimate discussion you mentioned and will<br>
> check it out.<br>
><br>
> There are a number of regexp packages for Erlang which use PCRE or other <br>
> libraries which I don't have references to. Look in the "Not an
Erlang fan"<br>
> thread here which gives some references to them.<br>
><br>
> An interesting alternative would be to write the regexp parser/compiler in
<br>
> Erlang and have the basic regexp matcher in the emulator. This actually<br>
> wouldn't be so big, I think. Get the best of both worlds. An I used to<br>
> complain (still do in fact) that the emulator is to big. :-) <br>
><br>
> Robert<br>
><br>
> * Actually there are two types of POSIX regexps in the standard, one which<br>
> has back-references and one which doesn't.<br>
><br>
><br>
><br>
> On 25/09/2007, G Bulmer < <a href="mailto:gbulmer@gmail.com">gbulmer@gmail.com</a>>
wrote:<br>
> > After reading the original thread, I wondered what implementation of<br>
> > regular expression matching is used by Erlang?<br>
> > <br>
> > I ask because this paper points out that many implementations have<br>
> > bad pathological behaviour:<br>
> > <a href="http://swtch.com/~rsc/regexp/regexp1.html">http://swtch.com/~rsc/regexp/regexp1.html
</a><br>
> > An observation they make is:<br>
> > "Regular expression matching can be simple and fast, using
finite<br>
> > automata-based techniques that have been known for decades. In<br>
> > contrast, Perl, PCRE, Python, Ruby, Java, and many other languages <br>
> > have regular expression implementations based on recursive<br>
> > backtracking that are simple but can be excruciatingly slow."<br>
> ><br>
> > This is followed by a discussion at <a
href="http://lambda-the-ultimate.org/">http://lambda-the-ultimate.org/</a><br>
> > node/2064<br>
> > Which seems to come to a working implementation.<br>
> > Usefully, they note that Thompson's original patent has expired, so<br>
> > there is no obstacle to using the original, and apparently superior, <br>
> > algorithm<br>
> ><br>
> > I looked at 'man regex' on my Mac, and it describes the POSIX<br>
> > implementation of regex. It does contain the note "Originally
written<br>
> > by Henry Spencer", which is suggested as the basis/culprit of
the <br>
> > pathological regular expression implementations (I should add, as<br>
> > Henry Spencer is one of my heroes, that he is also credited with TCL<br>
> > 8 regex, which works very well according to the graphs in the paper).
<br>
> ><br>
> > I have tried grep'ing through *.c in otp_src for regex and regcomp,<br>
> > and came up with nothing!<br>
> ><br>
> > Does anyone know what is used in Erlang? It may be a relatively small
<br>
> > amount of work to incorporate a new regex to provide more stable<br>
> > behaviour.<br>
> ><br>
> > GB<br>
> ><br>
> > On 25 Sep 2007, at 10:00,<br>
> <a href="mailto:erlang-questions-request@erlang.org">erlang-questions-request@erlang.org</a>
wrote:<br>
> ><br>
> > > Date: Mon, 24 Sep 2007 09:20:08 -0700 (PDT)<br>
> > > From: Thomas Lindgren < <a
href="mailto:thomasl_erlang@yahoo.com">thomasl_erlang@yahoo.com </a>><br>
> > > Subject: Re: [erlang-questions] Not an Erlang fan<br>
> > > To: "Erlang-Questions \(E-mail\)" < <a
href="mailto:erlang-questions@erlang.org">erlang-questions@erlang.org</a>> <br>
> > > Message-ID:<br>
> <<a href="mailto:761781.89097.qm@web38802.mail.mud.yahoo.com">761781.89097.qm@web38802.mail.mud.yahoo.com</a>
><br>
> > > Content-Type: text/plain; charset=iso-8859-1<br>
> > ><br>
> > ><br>
> > > -<br>
> > > Note that this does not include the actual regexp<br>
> > > search of each line ... The code to do that could be<br>
> > > an FSM + dictionary insert instead of the simple <br>
> > > check-and-accumulate for newline. Extra cost? Unclear.<br>
> > ><br>
> > > One could merge that code into the scanning the binary<br>
> > > but this should ideally be done by a regexp compiler, <br>
> > > shouldn't it? Not done at this time AFAIK.<br>
> > ><br>
> > > Best,<br>
> > > Thomas<br>
> ><br>
> > > Date: Mon, 24 Sep 2007 19:55:12 +0200<br>
> > > From: Claes Wikstrom < <a href="mailto:klacke@hyber.org">klacke@hyber.org</a>><br>
> > > Subject: Re: [erlang-questions] Not an Erlang fan<br>
> > > To: Bob Ippolito < <a href="mailto:bob@redivi.com">bob@redivi.com</a>>
<br>
> > > Cc: "Erlang-Questions \(E-mail\)" < <a
href="mailto:erlang-questions@erlang.org">erlang-questions@erlang.org</a>><br>
> > > Message-ID: < <a href="mailto:46F7FA00.8070507@hyber.org">46F7FA00.8070507@hyber.org</a>><br>
> > > Content-Type: text/plain; charset=ISO-8859-1; format=flowed<br>
> > ><br>
> > > Bob Ippolito wrote:<br>
> > ><br>
> > >> On 9/24/07, Patrick Logan < <a
href="mailto:patrickdlogan@gmail.com">patrickdlogan@gmail.com</a>> wrote:<br>
> > >><br>
> > >>>>>><br>
> <a href="http://www.tbray.org/ongoing/When/200x/2007/09/22/Erlang">http://www.tbray.org/ongoing/When/200x/2007/09/22/Erlang
</a><br>
> > >>>>>><br>
> > >>>>>> Tim Bray might raise some valid points here,
even if he's<br>
> > >>>>>> slightly<br>
> > >>>>>> biased by his background. <br>
> > >>>>>><br>
> > ><br>
> > > No, the only fast way today to process a large file line/by/line
is to<br>
> > ><br>
> > > 1. file:open(Filename, [read, raw])<br>
> > > 2. In a loop {ok, Bin} = file:read(Fd, BufSize),<br>
> > > 3. Use a binary regex matcher such as<br>
> > > <a
href="http://yaws.hyber.org/download/posregex-1.0.tgz">http://yaws.hyber.org/download/posregex-1.0.tgz
</a><br>
> > ><br>
> > > (I don't know the state of the regex lib in OTP today, last time<br>
> > > I looked it sucked bigtime though)<br>
> > ><br>
> > > /klacke<br>
> > > <br>
> > _______________________________________________<br>
> > erlang-questions mailing list<br>
> > <a href="mailto:erlang-questions@erlang.org">erlang-questions@erlang.org</a><br>
> > <a href="http://www.erlang.org/mailman/listinfo/erlang-questions">http://www.erlang.org/mailman/listinfo/erlang-questions</a><br>
> ><br>
><br>
><br>
> _______________________________________________<br>
> erlang-questions mailing list<br>
> <a href="mailto:erlang-questions@erlang.org">erlang-questions@erlang.org</a><br>
> <a href="http://www.erlang.org/mailman/listinfo/erlang-questions">http://www.erlang.org/mailman/listinfo/erlang-questions</a><br>
></p>
</div>
<p class=MsoNormal> </p>
</div>
</body>
</html>