Re: API review: "repository relative URL" or "relative repository URL"

[Ivan Zhakov <iv...@visualsvn.com>]

On Wed, Mar 20, 2013 at 5:43 PM, C. Michael Pilato <cm...@collab.net> wrote:
> On 03/20/2013 08:21 AM, C. Michael Pilato wrote:
>> On 03/20/2013 12:23 AM, Branko Čibej wrote:
>>> On 19.03.2013 21:38, Julian Foad wrote:
>>>> C. Michael Pilato wrote:
>>>>
>>>>> svn_path.h added two new APIs since 1.7.x, svn_path_is_repos_relative_url()
>>>>> and svn_path_resolve_repos_relative_url().  These functions have in common
>>>>> this nomenclature of a "repository relative URL".  But elsewhere, I see
>>>>> references to a "relative repository URL".  Do we have a preferred syntax
>>>>> for those ^/-prefixed things?  The book refers to them only as "relative
>>>>> URLs" at this point, but if we decide one way or another I'm happy to
>>>>> reinforce our decision in the svnbook text.
>>>> I had a hand in their creation, and I call them "repository-relative URLs".
>>>
>>> I tend to agree with this term, because "relative URL" already has a
>>> widely accepted meaning that is noticeably different from our usage.
>>
>> Okay, that all works for me.  The hyphenated form helps bind the "relative"
>> to the "repository" as opposed to the "URL", which is precisely the
>> ambiguity I was hoping to clear up.
>>
>
> Related follow-up:  should we change the output of 'svn info' so that
> instead of:
>
>    Relative URL: ^/trunk/en/README
>
> it prints:
>
>    Repository-Relative URL: ^/trunk/en/README
>
> Or is that just obnoxious?
>
Probably we can just print:
[[
URL: ^/trunk/en/README
]]

'^' symbol already states that this URL is repository relative.


-- 
Ivan Zhakov

Re: API review: "repository relative URL" or "relative repository URL"

["C. Michael Pilato" <cm...@collab.net>]

On 03/20/2013 11:11 AM, Julian Foad wrote:
>> '^' symbol already states that this URL is repository relative.
> 
> No.  The "URL:" prefix is already used for the full URL, and we want
> unique prefixes because the 'svn info' output is (I believe) commonly
> parsed by scripts that do things like 'svn info | grep "^URL:"'.

Correct.  I frequently use 'svn info' to get a full URL which I can
copy-n-paste into my web browser or into an email with 'svn checkout'
instructions (where the ^ syntax is useless), for example.

-- 
C. Michael Pilato <cm...@collab.net>
CollabNet   <>   www.collab.net   <>   Enterprise Cloud Development

Re: API review: "repository relative URL" or "relative repository URL"

[Julian Foad <ju...@btopenworld.com>]

Ivan Zhakov wrote:

> On Wed, Mar 20, 2013 at 5:43 PM, C. Michael Pilato wrote:
>>  Related follow-up:  should we change the output of 'svn info' so 
> that
>>  instead of:
>> 
>>     Relative URL: ^/trunk/en/README
>> 
>>  it prints:
>> 
>>     Repository-Relative URL: ^/trunk/en/README
>> 
>>  Or is that just obnoxious?
>> 
> Probably we can just print:
> [[
> URL: ^/trunk/en/README
> ]]
> 
> '^' symbol already states that this URL is repository relative.

No.  The "URL:" prefix is already used for the full URL, and we want unique prefixes because the 'svn info' output is (I believe) commonly parsed by scripts that do things like 'svn info | grep "^URL:"'.

- Julian