Path: csiph.com!xmission!news.snarked.org!border2.nntp.dca1.giganews.com!nntp.giganews.com!news.iecc.com!.POSTED.news.iecc.com!nerds-end
From: Clint O <clint.olsen@gmail.com>
Newsgroups: comp.compilers
Subject: Re: Regular expression string searching & matching
Date: Thu, 22 Mar 2018 17:46:03 GMT
Organization: Newshosting.com - Highest quality at a great price! www.newshosting.com
Lines: 49
Sender: news@iecc.com
Approved: comp.compilers@iecc.com
Message-ID: <18-03-089@comp.compilers>
References: <18-03-016@comp.compilers> <18-03-032@comp.compilers> <18-03-034@comp.compilers> <18-03-035@comp.compilers> <18-03-041@comp.compilers> <18-03-045@comp.compilers> <18-03-054@comp.compilers> <18-03-087@comp.compilers>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Injection-Info: gal.iecc.com; posting-host="news.iecc.com:2001:470:1f07:1126:0:676f:7373:6970"; logging-data="95950"; mail-complaints-to="abuse@iecc.com"
Keywords: lex, DFA
Posted-Date: 22 Mar 2018 20:43:53 EDT
X-submission-address: compilers@iecc.com
X-moderator-address: compilers-request@iecc.com
X-FAQ-and-archives: http://compilers.iecc.com
Xref: csiph.com comp.compilers:2021

On 2018-03-20, Clint O <clint.olsen@gmail.com> wrote:
> [ reposted to try to make the special characters look right ]
>
> q0: /·[*]·([^*] | [*]+·[^/])*·[*]+·/
>     [/] q2
>     ['\x00'-.0-ÿ] q1
> q1: ∅
>     ['\x00'-ÿ] q1
> q2: [*]·([^*] | [*]+·[^/])*·[*]+·/
>     [*] q3
>     ['\x00'-)+-ÿ] q1
> q3: ([^*] | [*]+·[^/])*·[*]+·/
>     [*] q4
>     ['\x00'-)+-ÿ] q3
> q4: ([*]*·[^/]·([^*] | [*]+·[^/])*·[*]+ | [*]*)·/
>     [*] q6
>     ['\x00'-)+-.0-ÿ] q3
>     [/] q5
> q5: ε
>     ['\x00'-ÿ] q1
> q6: (([*]*·[^/] | ε)·([^*] | [*]+·[^/])*·[*]+ | [*]*)·/
>     [*] q8
>     ['\x00'-)+-.0-ÿ] q3
>     [/] q7
> q7: ([^*] | [*]+·[^/])*·[*]+·/ | ε
>     [*] q4
>     ['\x00'-)+-ÿ] q3
> q8: ((([*]*·[^/] | ε)·([^*] | [*]+·[^/])* | [*]*·[^/]·([^*] | [*]+·[^/])*)·[*]+ | [*]*)·/
>     [*] q8
>     ['\x00'-)+-.0-ÿ] q3
>     [/] q7

Thanks John for reposting this. It looks much better now.

In summary:

q5,q7 are accepting states since they contain epsilon. q2 represents the
error state.

The key to success with this algorithm is recognizing previously calculated
derivatives/expressions. When you no longer calculate unique derivatives,
the DFA construction terminates. As you can see the expressions can get
hairy pretty quickly. I don't know if you can glean much from the
successive expressions generated. It's akin to the method of walking the
parse tree but it bypasses the NFA construction entirely.

Thanks,

-Clint