Home » Archived » Buckminster » Generating feature version suffixes
| |
Re: Generating feature version suffixes [message #540686 is a reply to message #536930] |
Thu, 17 June 2010 01:06 |
Peter Nehrer Messages: 241 Registered: July 2009 |
Senior Member |
|
|
Henrik,
I've been trying to track this down, and it seems as though it should
already be happening. It appears independent of the qualifier replacer.
I haven't debugged it, but following the code, a feature's manifest
action, which invokes expand.feature.version from buckminster.pdetasks,
should already do the job -- use FeatureConsolidator, which in turn uses
FeatureVersionSuffixGenerator to produce the desired suffix.
Is there anything that pops into your mind as to why this wouldn't work
out of the box? Thanks again.
--Peter
On 5/31/10 10:20 AM, Henrik Lindberg wrote:
> There are qualifier replacement stategies available - they are described
> in the documentation.
>
> I am not sure the latest replacement stategies have been documented.
> Search for open documentation issues in buckminster bugzilla.
>
> - henrik
> On 5/31/10 3:58 PM, Peter Nehrer wrote:
>> Hi,
>>
>> is there a way to have Buckminster generate feature versions suffixes,
>> basically the PDE equivalent of generateFeatureVersionSuffix=true?
>>
>> I'd like the feature to get a new qualifier whenever one of its bundles
>> changes (i.e., gets a new revision-based qualifier) even when the
>> feature's own revision hasn't changed. Thanks!
>>
>> --Peter
>
|
|
|
Re: Generating feature version suffixes [message #540695 is a reply to message #540686] |
Thu, 17 June 2010 05:31 |
|
Hi Peter,
Did you specify the properties that declares what generator to use? I.e. something like:
qualifier.replacement.*=generator:lastRevision
generator.lastRevision.format=r{0,number,00000}
Regards,
Thomas Hallgren
On 06/17/2010 03:06 AM, Peter Nehrer wrote:
> Henrik,
>
> I've been trying to track this down, and it seems as though it should
> already be happening. It appears independent of the qualifier replacer.
> I haven't debugged it, but following the code, a feature's manifest
> action, which invokes expand.feature.version from buckminster.pdetasks,
> should already do the job -- use FeatureConsolidator, which in turn uses
> FeatureVersionSuffixGenerator to produce the desired suffix.
>
> Is there anything that pops into your mind as to why this wouldn't work
> out of the box? Thanks again.
>
> --Peter
>
> On 5/31/10 10:20 AM, Henrik Lindberg wrote:
>> There are qualifier replacement stategies available - they are described
>> in the documentation.
>>
>> I am not sure the latest replacement stategies have been documented.
>> Search for open documentation issues in buckminster bugzilla.
>>
>> - henrik
>> On 5/31/10 3:58 PM, Peter Nehrer wrote:
>>> Hi,
>>>
>>> is there a way to have Buckminster generate feature versions suffixes,
>>> basically the PDE equivalent of generateFeatureVersionSuffix=true?
>>>
>>> I'd like the feature to get a new qualifier whenever one of its bundles
>>> changes (i.e., gets a new revision-based qualifier) even when the
>>> feature's own revision hasn't changed. Thanks!
>>>
>>> --Peter
>>
>
|
|
|
Re: Generating feature version suffixes [message #540985 is a reply to message #540695] |
Thu, 17 June 2010 18:57 |
Peter Nehrer Messages: 241 Registered: July 2009 |
Senior Member |
|
|
Thomas,
I get the same effect with default as well as lastRevision generator.
So far I narrowed it down to this:
1. if at least one feature is included, then suffix is generated
2. if only plugins are included, then no suffix is generated
Stepping thru it a little bit, I noticed that when FeatureConsolidator
tries to get the feature's bundles (feature.getImports() in
consolidateFeatureVersion(List<ComponentIdentifier>), it gets back an
empty array. Indeed, the feature model doesn't have anything in imports;
all bundles are in plugins (so feature.getPlugins() would have worked).
Subsequently the FeatureVersionSuffixGenerator finds nothing to make a
suffix from.
Btw this probably means that the suffixes aren't getting correctly
generated if the included bundles' versions are ignored.
Is this a bug or am I missing something? Thanks.
--Peter
On 6/17/10 1:31 AM, Thomas Hallgren wrote:
> Hi Peter,
>
> Did you specify the properties that declares what generator to use? I.e.
> something like:
>
> qualifier.replacement.*=generator:lastRevision
> generator.lastRevision.format=r{0,number,00000}
>
> Regards,
> Thomas Hallgren
>
> On 06/17/2010 03:06 AM, Peter Nehrer wrote:
>> Henrik,
>>
>> I've been trying to track this down, and it seems as though it should
>> already be happening. It appears independent of the qualifier replacer.
>> I haven't debugged it, but following the code, a feature's manifest
>> action, which invokes expand.feature.version from buckminster.pdetasks,
>> should already do the job -- use FeatureConsolidator, which in turn uses
>> FeatureVersionSuffixGenerator to produce the desired suffix.
>>
>> Is there anything that pops into your mind as to why this wouldn't work
>> out of the box? Thanks again.
>>
>> --Peter
>>
>> On 5/31/10 10:20 AM, Henrik Lindberg wrote:
>>> There are qualifier replacement stategies available - they are described
>>> in the documentation.
>>>
>>> I am not sure the latest replacement stategies have been documented.
>>> Search for open documentation issues in buckminster bugzilla.
>>>
>>> - henrik
>>> On 5/31/10 3:58 PM, Peter Nehrer wrote:
>>>> Hi,
>>>>
>>>> is there a way to have Buckminster generate feature versions suffixes,
>>>> basically the PDE equivalent of generateFeatureVersionSuffix=true?
>>>>
>>>> I'd like the feature to get a new qualifier whenever one of its bundles
>>>> changes (i.e., gets a new revision-based qualifier) even when the
>>>> feature's own revision hasn't changed. Thanks!
>>>>
>>>> --Peter
>>>
>>
>
|
|
|
Goto Forum:
Current Time: Fri Mar 29 14:38:23 GMT 2024
Powered by FUDForum. Page generated in 0.05194 seconds
|