Return to the
comp.compilers page.
Search the
comp.compilers archives again.
| Related articles |
|---|
| regular expression search algorithm forsyth@minster.york.ac.uk (1993-03-18) |
| Reg. expr. character class compression cohenw@ecn.purdue.edu (1993-03-18) |
| Reg. expr. character class compression dmr@research.att.com (1993-03-21) |
| Newsgroups: | comp.compilers |
| From: | dmr@research.att.com (Dennis Ritchie) |
| Keywords: | lex, DFA |
| Organization: | Compilers Central |
| References: | 93-03-063 |
| Date: | Sun, 21 Mar 1993 06:19:52 GMT |
Keep in mind that during the next few years the potential size of the
character set and hence of a character class is likely to increase from
about 256 to about 65536. Converting the Plan 9 regular expression stuff
to Unicode wasn't trivial. (But I didn't do the work and don't know what
techniques were used.)
Another note: the regular expression algorithm described by Thompson was
patented; the patent must have issued around 1973. AT&T never prosecuted
it, and it has probably expired; I mention it as a curiosity, and as an
illustration that software patents are not especially new.
Dennis Ritchie
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.