questions as a diff

[Benoit Boissinot]

On Sun, Mar 01, 2009 at 04:42:14PM +0200, timeless wrote:
> could people please comment about the items tagged w/ XXX?
> 
>      hg email -o               # send all patches not in default
>      hg email -o DEST          # send all patches not in DEST
>      hg email -o -r 3000       # send all ancestors of 3000 not in default
>      hg email -o -r 3000 DEST  # send all ancestors of 3000 not in DEST
> +XXX i'm told this is supposed to match hg help push -r
> +-r --rev        a specific revision up to which you would like to push
> +   If -r is used, the named changeset and all its ancestors will
> +   to the remote repository.
> +However, the text doesn't, isn't consistent, etc., if it means the same thing,
> +please be kind to your localizers and make it say precisely the same thing as
> +best as possible, and if you have trouble spell it out "this matches the
> +behavior of ...".

You're right when -r is used with -o, it should have the same behaviour
as push.

>  HGENCODING::
> -    This overrides the default locale setting detected by Mercurial.
> +    This supercedes the encoding derived from the locale setting
> +    detected by Mercurial.
> +XXX the original text pretended ENCODING and LOCALE were equivalent.
> +I'm assuming that LOCALE includes translation (and as such was a superset)
> +If I'm wrong. Fix the message some other way or explain what it actually
> +meant and I'll figure out something else.

I think HGENCODING was done before we had locale support, hence the
mixup. I don't think HGENCODING has any effect on LOCALE.

>  HGMERGE::
>      An executable to use for resolving merge conflicts. The program
>      will be executed with three arguments: local file, remote file,
>      ancestor file.
> 
>      (deprecated, use .hgrc)
> +XXX isn't there a command line option for this?

you can modify hgrc variables with --config ui.merge=XXX

>      - author: String. The unmodified author of the changeset.
>      - branches: String. The name of the branch on which the changeset
>            was committed. Will be empty if the branch name was default.
> +XXX what if there's more than one?
> +XXX if not, isn't this horribly misnamed? if so, the help should note that

Someone should test ;)

>            the timezone: "Mon Sep 04 15:13:13 2006 0700".
>      - domain: Any text. Finds the first string that looks like an email
>            address, and extracts just the domain component.
>            Example: 'User <user at example.com>' becomes 'example.com'.
>      - email: Any text. Extracts the first string that looks like an email
>            address. Example: 'User <user at example.com>' becomes
>            'user at example.com'.
>      - escape: Any text. Replaces the special XML/XHTML characters "&",
>            "<" and ">" with XML entities.
>      - fill68: Any text. Wraps the text to fit in 68 columns.
> +XXX what does it do if there are 70 -'s in a row?
> +how does it choose where to wrap? surprise me?

good remark, it tries to be smart and to insert \n instead of spaces.
if it's not possible it might break a word, (or the 70 -'s).

Should we use textwrap.fill() instead ? from quick testing with timeit
it isn't slower.

>      - fill76: Any text. Wraps the text to fit in 76 columns.
> +XXX 80?
>      - firstline: Any text. Returns the first line of text.
> +XXX is this operating on lists or lines, or lists with entries w/ multilines?

it is operating with a string (not a list)

>      - hgdate: Date. Returns the date as a pair of numbers:
>            "1157407993 25200" (Unix timestamp, timezone offset).
>      - isodate: Date. Returns the date in ISO 8601 format.
>      - obfuscate: Any text. Returns the input text rendered as a sequence
>            of XML entities.
>      - person: Any text. Returns the text before an email address.
> +XXX example
foobar <x at y.com> -> foobar
x at y.com -> x

>        local/filesystem/path (or file://local/filesystem/path)
> +XXX local=localhost? (there's a lot of confusion about how to
> properly write file://localhost/path/)

I don't understand, I don't think there is an host part in file:// urls.

>        http://[user[:pass]@]host[:port]/[path]
>        https://[user[:pass]@]host[:port]/[path]
>        ssh://[user[:pass]@]host[:port]/[path]
> 
>      Paths in the local filesystem can either point to Mercurial
>      repositories or to bundle files (as created by 'hg bundle' or
>      'hg incoming --bundle').
> +XXX paths outside the local file system can't point to bundles?

Probably not, or only http paths then (that would be a side-effect of
using urlopen.

>      - Mercurial doesn't use its own compression via SSH; the right thing
>        to do is to configure it in your ~/.ssh/config, e.g.:
>          Host *.mylocalnetwork.example.com
>            Compression no
> +XXX why are you turning off local compression here w/o explanation?

I don't understand.

regards,

Benoit

-- 
:wq