[darcs-users] darcs patch: Ignore junk more conservatively. (and 1 more)

David Roundy droundy at darcs.net
Wed Oct 8 17:10:06 UTC 2008


On Wed, Oct 08, 2008 at 02:31:57PM +0000, Eric Kow wrote:
> For discussion, and maybe stable.
> 
> 1. Zooko wants darcs patches to tell darcs how many ignore-this
>    lines to ignore.  I understand his proposal (at last!), but
>    feel it is a bit too complicated.  (Sure, the parsing is
>    very very easy, but why do it when we can just change the
>    junk names to something users are less likely to use?)

I don't like this idea.

> 2. a. darcs 2.1.0pre3 ignores /every/ junky line
>    b. darcs unstable generates junky lines at the beginning
>       of every patch log
>    c. Zooko proposes (as a short-term alternative to #1)
>       that we ignore junky lines at the /end/ of a patch log
> 
>    This patch is a compromise between b and c.  To get (c)
>    we could easily flip the (dropWhile is_ignored) into
>    (takeWhile is_not_ignored), which could be fine too but
>    entails changing darcs-unstable.

I don't like these ideas.

> 3. * Max suggested that we call the line 'Patch-salt'
>    * David pointed out that 'Ignore-this' is more general
>    * Tommy suggested 'Ignore-this-patch-salt'
>    * Nathan proposed 'Hidden-metadata'
> 
>    I vote for the Tommy route, which seems to meet all
>    criteria and also reduces the risk of ignoring too much.

Nor do I like these.

What I don't know (and maybe it's hidden somewhere in those emails
that I only skimmed) is what problem there is.  Why is it considered
important for users to be able to write "Ignore-this: foobar" in their
patches and have darcs display this to people who pull from them?

David

> Wed Oct  8 14:54:17 BST 2008  Eric Kow <eric.kow at gmail.com>
>   * Ignore junk more conservatively.
>   Instead of ignoring every line that looks like junk, only ignore
>   initial lines that do so.
> 
> Wed Oct  8 14:57:52 BST 2008  Eric Kow <eric.kow at gmail.com>
>   * Make patch junk name more specific.
>   We want to reduce the chances of somebody's patch comment being
>   accidentally ignored by darcs.
>   Idea by Max Battcher and Tommy Pettersson.


More information about the darcs-users mailing list