IndexError: list index out of range

Matt Mackall mpm at selenic.com
Sun Aug 17 20:43:20 CDT 2008


On Mon, 2008-08-18 at 00:39 +0000, TK Soh wrote:

> OT: I read on the blog that the OP has decided to switch to git
> because of this, and other, problems in Mercurial. Maybe something
> does require fixing here beyond this bug.

Python applications throw exceptions when they hit bugs. C applications
either mysteriously dump core or, more often, pretend nothing happened.
So when someone complains "I got a backtrace in Python app and am
switching to a C app that seems more robust", I'm not gonna get worked
up about it.

Complaining about Mq and then switching to Git is also a little
ridiculous. From his "Rosetta Stone":

hg qpush ?
hg qpop ? 
hg qimport -r tip ? 
hg qnew -f some.patch ?

That's because there are no equivalents in Git unless you use a
third-party extension. And extensions both there and here are something
you use at your own risk.

He also complains about Mq + rename tracking (first spotted and fixed
between 1.0.1 and 1.0.2) when Git has no rename tracking at all.

There's an extension to record hunks, it's called record.

The hgweb problem is just bizarre. Mercurial's fairly popular main repo
is served off a virtual server on a loaded box in my basement using slow
CGI over DSL, and the load from Mercurial is negligible. He's probably
got Pygments turned on by the silly Debian packaging.

I will admit that there are a bunch of annoying Mq bugs that need
fixing. Perhaps we should unbundle it until they get sorted out.
 
But whatever, best of luck with Git. I've never used it without getting
a headache.

-- 
Mathematics is the supreme nostalgia of our time.



More information about the Mercurial mailing list