Re: Internal Representation of Strings

"Ira Baxter" <idbaxter@semdesigns.com>
Sat, 21 Feb 2009 10:35:27 -0600

          From comp.compilers

Related articles
[14 earlier articles]
Re: Internal Representation of Strings bartc@freeuk.com (Bartc) (2009-02-18)
Re: Internal Representation of Strings tony@my.net (Tony) (2009-02-18)
Re: Internal Representation of Strings tony@my.net (Tony) (2009-02-18)
Re: Internal Representation of Strings cr88192@hotmail.com (cr88192) (2009-02-19)
Re: Internal Representation of Strings cr88192@hotmail.com (cr88192) (2009-02-21)
Re: Internal Representation of Strings tony@my.net (Tony) (2009-02-21)
Re: Internal Representation of Strings idbaxter@semdesigns.com (Ira Baxter) (2009-02-21)
Re: Internal Representation of Strings cr88192@hotmail.com (cr88192) (2009-02-22)
Re: Internal Representation of Strings DrDiettrich1@aol.com (Hans-Peter Diettrich) (2009-02-22)
Re: Internal Representation of Strings DrDiettrich1@aol.com (Hans-Peter Diettrich) (2009-02-22)
Re: Internal Representation of Strings bartc@freeuk.com (Bartc) (2009-02-22)
Re: Internal Representation of Strings scooter.phd@gmail.com (Scott Michel) (2009-02-22)
Re: Internal Representation of Strings cr88192@hotmail.com (cr88192) (2009-02-23)
[13 later articles]
| List of all articles for this month |
From: "Ira Baxter" <idbaxter@semdesigns.com>
Newsgroups: comp.compilers
Date: Sat, 21 Feb 2009 10:35:27 -0600
Organization: Compilers Central
References: 09-02-051 09-02-068 09-02-078 09-02-084 09-02-090 09-02-105
Keywords: parse, practice
Posted-Date: 21 Feb 2009 18:56:02 EST

> [Let's say you have a gigantic parse tree with 10,000 nodes. That means
> you'd have 40K of length words. Who cares? -John]


Gigantic parse trees have 30 million nodes. We encounter them.
(You can argue such things are crazy. I agree. But I don't
argue with my customer when he shows up with one.)
My personal theory is that with memory touches being expensive
in terms of (hundreds of) clock cycle (and getting further away as
transistors shrink), any scheme that burns
a few machine instructions of clocks to enable statistically common
cases to be "local" is a very good tradeoff.
As an example, we store small integers in number-carrying leaf
nodes, with a hash table used to store pretty rare big integers.
--
Ira Baxter, CTO
www.semanticdesigns.com



Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.