Re: [Nmh-workers] Thoughts: header/address parsing

[Lyndon Nerenberg]

On Aug 3, 2014, at 6:30 PM, Ken Hornstein <address@hidden> wrote:

> Yeah.  We'd store the original contents of the headers, of course, but they
> wouldn't go through any additional parsing until they were accessed.

Ah. I thought you were talking about a lazy evaluation parser.

I don't think you are going to find any measurable savings in deferring much, 
if any, of the header parsing, even into complimentary formats.  The 
measurements I did in the Cyrus IMAP server's MIME parser showed that that part 
of the code was in the noise compared to the I/O overhead.

We were optimizing for response time across hundreds of transactions per 
second.  I can't imagine how anyone using MH from the command line would ever 
notice this sort of tune-up.

I'm all in favour of cleaning up the code, but let's not drive the process from 
a false premise.

--lyndon

signature.asc
Description: Message signed with OpenPGP using GPGMail