Re: [cxf] branch master updated: CXF-7525...

2017-12-29 Thread Sergey Beryozkin

Hi Andriy

Indeed, makes sense to keep it and consider extending CXF UserModel over 
time to be a bit richer so that it can accommodate 
Swagger/OpenApi/something else/ models...


Cheers, Sergey
On 29/12/17 17:13, Andriy Redko wrote:

Hi Sergey,

Got it, thanks a lot for explaining the context. So I guess, we could
keep it, right? There is a diff from older Swagger specs, but more or less,
the parser does the job. What do you think?

Best Regards,
 Andriy Redko

SB> Hi Andriy

SB> The idea there was to support the dynamic creation of CXF JAX-RS
SB> endpoints from a given Swagger doc, i.e, support Swagger-first dev
SB> without the code-generation, by converting Swagger docs into CXF
SB> specific UserResource(s) which in turn can be used to initialize the
SB> endpoint...
SB> I've found UserResource can not entirely match swagger model, the
SB> original idea behind UserResource/etc is doc-ed here:
SB> 
http://cxf.apache.org/docs/jax-rs-advanced-features.html#JAX-RSAdvancedFeatures-RESTfulserviceswithoutannotations

SB> with the model/etc classes expected to be avail on the classpath

SB> Sergey



SB> On 29/12/17 16:35, Andriy Redko wrote:

Hi Sergey,



I am not sure actually it is needed, but I am trying to match our Swagger2 
implementation
where we have/are using Swagger2ParseUtils. We have test cases for that as well 
as it is being
used in system tests for parsing the spec and checking expectations. Does it 
make sense?
Thanks.



Best Regards,
  Andriy Redko



SB> Hi Andriy



SB> Just curious, why would this new module will need OpenApiParseUtils ?



SB> Thanks, Sergey





Re: [cxf] branch master updated: CXF-7525...

2017-12-29 Thread Andriy Redko
Hi Sergey,

Got it, thanks a lot for explaining the context. So I guess, we could
keep it, right? There is a diff from older Swagger specs, but more or less,
the parser does the job. What do you think?

Best Regards,
Andriy Redko

SB> Hi Andriy

SB> The idea there was to support the dynamic creation of CXF JAX-RS 
SB> endpoints from a given Swagger doc, i.e, support Swagger-first dev 
SB> without the code-generation, by converting Swagger docs into CXF 
SB> specific UserResource(s) which in turn can be used to initialize the 
SB> endpoint...
SB> I've found UserResource can not entirely match swagger model, the 
SB> original idea behind UserResource/etc is doc-ed here:
SB> 
http://cxf.apache.org/docs/jax-rs-advanced-features.html#JAX-RSAdvancedFeatures-RESTfulserviceswithoutannotations

SB> with the model/etc classes expected to be avail on the classpath

SB> Sergey



SB> On 29/12/17 16:35, Andriy Redko wrote:
>> Hi Sergey,

>> I am not sure actually it is needed, but I am trying to match our Swagger2 
>> implementation
>> where we have/are using Swagger2ParseUtils. We have test cases for that as 
>> well as it is being
>> used in system tests for parsing the spec and checking expectations. Does it 
>> make sense?
>> Thanks.

>> Best Regards,
>>  Andriy Redko

>> SB> Hi Andriy

>> SB> Just curious, why would this new module will need OpenApiParseUtils ?

>> SB> Thanks, Sergey




Re: [cxf] branch master updated: CXF-7525...

2017-12-29 Thread Sergey Beryozkin

Hi Andriy

The idea there was to support the dynamic creation of CXF JAX-RS 
endpoints from a given Swagger doc, i.e, support Swagger-first dev 
without the code-generation, by converting Swagger docs into CXF 
specific UserResource(s) which in turn can be used to initialize the 
endpoint...
I've found UserResource can not entirely match swagger model, the 
original idea behind UserResource/etc is doc-ed here:

http://cxf.apache.org/docs/jax-rs-advanced-features.html#JAX-RSAdvancedFeatures-RESTfulserviceswithoutannotations

with the model/etc classes expected to be avail on the classpath

Sergey



On 29/12/17 16:35, Andriy Redko wrote:

Hi Sergey,

I am not sure actually it is needed, but I am trying to match our Swagger2 
implementation
where we have/are using Swagger2ParseUtils. We have test cases for that as well 
as it is being
used in system tests for parsing the spec and checking expectations. Does it 
make sense?
Thanks.

Best Regards,
 Andriy Redko

SB> Hi Andriy

SB> Just curious, why would this new module will need OpenApiParseUtils ?

SB> Thanks, Sergey



Re: [cxf] branch master updated: CXF-7525...

2017-12-29 Thread Andriy Redko
Hi Sergey,

I am not sure actually it is needed, but I am trying to match our Swagger2 
implementation
where we have/are using Swagger2ParseUtils. We have test cases for that as well 
as it is being
used in system tests for parsing the spec and checking expectations. Does it 
make sense? 
Thanks.

Best Regards,
Andriy Redko

SB> Hi Andriy

SB> Just curious, why would this new module will need OpenApiParseUtils ?

SB> Thanks, Sergey



Re: [cxf] branch master updated: CXF-7525...

2017-12-29 Thread Sergey Beryozkin
Though may be indeed makes sense to support the basic creation of 
endpoints via this route...


Sergey
On 29/12/17 16:24, Sergey Beryozkin wrote:

Hi Andriy

Just curious, why would this new module will need OpenApiParseUtils ?

Thanks, Sergey


Re: [cxf] branch master updated: CXF-7525...

2017-12-29 Thread Sergey Beryozkin

Hi Andriy

Just curious, why would this new module will need OpenApiParseUtils ?

Thanks, Sergey