|
From: | Rob Savoye |
Subject: | Re: [Gnash-commit] /srv/bzr/gnash/trunk r11136: migrated rtmp netstream and netconnect classes from rsavoye's local branch and fixed various test cases in misc-ming.all |
Date: | Wed, 17 Jun 2009 09:08:43 -0600 |
User-agent: | Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1b3pre) Gecko/20090513 Fedora/3.0-2.3.beta2.fc11 Thunderbird/3.0b2 |
On 06/17/09 09:00, Benjamin Wolsey wrote:
Anyway, I really fail to see the problem with starting from a design that's known to work, so that all the things that worked before still
It's called refactoring. :-) I believe the new design is better, and it certainly passes a large amount of test cases that the original code didn't. I've spent months analyzing how RTMP works in obscene detail, and I don't believe anyone else working on Gnash has.
am just worried that it will be delayed by rewriting that code so that fixing the regressions holds things up for weeks, when it's so unnecessary to rewrite it at all.
As mentioned, it was a migrating code process problem, not the new code itself. We'll track down what the problem was, backing out the change for now. Checkins should always pass "make check" before being committed. In addition, any code dealing with basic operation like NetConnection should be tested on a variety of web sites.
- rob -
[Prev in Thread] | Current Thread | [Next in Thread] |