[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <250b5819111d60626804fda395fedfc8@smtp.hushmail.com>
Date: Sun, 10 Jun 2012 11:51:23 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Patch: allow --markov=SECTION in addition to --markov[=[MIN_LEVEL-]LEVEL[:START[:END[:[MIN_LENGHT-]LENGTH]]]]
On 06/10/2012 11:39 AM, magnum wrote:
> On 06/10/2012 11:35 AM, magnum wrote:
>> But the first case, --markov=markov should use the existing section
>> [Options:Markov] so it should be the same as just saying --markov with
>> no parameters. And it does have a statfile defined.
>
> My bad. The default is [Options:Markov] but new ones are like
> [Markov:name]. Seems to work fine (and I did not get the segfault when
> using a custom section).
This is arguable, but:
1. Would it not be more logical to rename [Options:Markov] to
[Markov:default]? In Jumbo-5 it's all under just [Options] so there is
no history of [Options:Markov] that we need to preserve.
and
2. I think it would be nice to have missing "required" arguments in a
custom section defaulting to what's in [Markov:default].
That way, I could have a section using another statfile just using this:
[Markov:rockyou]
Statsfile = $JOHN/rockyou.stats
...and all other parameters would be taken from [Markov:default]. Or
would that get messy?
magnum
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
