|
| Re: do we need these requirement options [message #521887 is a reply to message #521868] |
Fri, 19 March 2010 07:44   |
|
On 03/19/2010 03:52 AM, Henrik Lindberg wrote:
> Hi,
> I think we may need to handle the following requirement options in the
> model:
> - greedy
> - optional
> - multiple
>
> At least when using a build unit to publish to p2 as it would otherwise
> not be possible to specify those.
>
> What do you think?
>
Greedy, definitely. And I think optional and multiple should be replace by integer min/max to match the (new) p2
requirement.
- thomas
|
|
|
| Re: do we need these requirement options [message #521911 is a reply to message #521887] |
Fri, 19 March 2010 04:36 |
 Henrik Lindberg
Messages: 2509
Registered: July 2009 |
Senior Member |
|
|
ok,
feel free to suggest some syntax for these, and thinking about p2
requirement syntax as well.
Currently, the syntax is:
"requires" ((NameSpace | "unit") '/')? name ('/' VersionRange) ';'
I can imagine adding keywords for greedy, max, min and optional and
declare these as modifiers (i.e. like private/public, final etc. to form
i.e. something like:
requires greedy max 3 optional ((NameSpace | "unit") '/')? name ('/'
VersionRange) ';'
but... feel free to suggest something.
- henrik
(and variations on that theme for multiple requirements).
On 3/19/10 8:44 AM, Thomas Hallgren wrote:
> On 03/19/2010 03:52 AM, Henrik Lindberg wrote:
>> Hi,
>> I think we may need to handle the following requirement options in the
>> model:
>> - greedy
>> - optional
>> - multiple
>>
>> At least when using a build unit to publish to p2 as it would otherwise
>> not be possible to specify those.
>>
>> What do you think?
>>
> Greedy, definitely. And I think optional and multiple should be replace
> by integer min/max to match the (new) p2 requirement.
>
> - thomas
|
|
|
|
| Re: do we need these requirement options [message #603431 is a reply to message #521868] |
Fri, 19 March 2010 07:44 |
|
On 03/19/2010 03:52 AM, Henrik Lindberg wrote:
> Hi,
> I think we may need to handle the following requirement options in the
> model:
> - greedy
> - optional
> - multiple
>
> At least when using a build unit to publish to p2 as it would otherwise
> not be possible to specify those.
>
> What do you think?
>
Greedy, definitely. And I think optional and multiple should be replace by integer min/max to match the (new) p2
requirement.
- thomas
|
|
|
| Re: do we need these requirement options [message #603435 is a reply to message #521887] |
Fri, 19 March 2010 09:20 |
Henrik Lindberg
Messages: 2509
Registered: July 2009 |
Senior Member |
|
|
ok,
feel free to suggest some syntax for these, and thinking about p2
requirement syntax as well.
Currently, the syntax is:
"requires" ((NameSpace | "unit") '/')? name ('/' VersionRange) ';'
I can imagine adding keywords for greedy, max, min and optional and
declare these as modifiers (i.e. like private/public, final etc. to form
i.e. something like:
requires greedy max 3 optional ((NameSpace | "unit") '/')? name ('/'
VersionRange) ';'
but... feel free to suggest something.
- henrik
(and variations on that theme for multiple requirements).
On 3/19/10 8:44 AM, Thomas Hallgren wrote:
> On 03/19/2010 03:52 AM, Henrik Lindberg wrote:
>> Hi,
>> I think we may need to handle the following requirement options in the
>> model:
>> - greedy
>> - optional
>> - multiple
>>
>> At least when using a build unit to publish to p2 as it would otherwise
>> not be possible to specify those.
>>
>> What do you think?
>>
> Greedy, definitely. And I think optional and multiple should be replace
> by integer min/max to match the (new) p2 requirement.
>
> - thomas
|
|
|
|
Powered by
FUDForum. Page generated in 0.02517 seconds
] 
Search
Help
Register
Login
Home
