From: James Gray <redmine@...>
Date: 2010-11-03T10:49:32+09:00
Subject: [ruby-core:33028] [Ruby 1.9-Feature#4017] [PATCH] CSV parsing speedup

Issue #4017 has been updated by James Gray.


I'm for adding the read_limit option (I prefer that name to io_read_limit).

However, I'm not sure how I feel about defaulting it to some number.  That's going to mean that some documents CSV used to correctly parse could now fail.  While a limit is a good idea in practice, I don't feel comfortable making that choice for the programmer.  I think you need to opt-in to a sensible limit for your purposes and, until you do, we assume you want the best parsing we can provide.

So, my main question is, can we drop the default limit and keep the gains?

I would definitely before applying a patch like this if we:

* Rename the limit to just read_limit
* Default to using no limit

I hope that makes sense.
----------------------------------------
http://redmine.ruby-lang.org/issues/show/4017

----------------------------------------
http://redmine.ruby-lang.org