You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@bloodhound.apache.org by Ryan Ollos <ry...@wandisco.com> on 2014/01/28 00:02:46 UTC

Re: [Apache Bloodhound] #388: Test environment isolation across product boundaries - after #355

On Mon, Jan 27, 2014 at 10:43 AM, Apache Bloodhound <
dev@bloodhound.apache.org> wrote:

> #388: Test environment isolation across product boundaries - after #355
> ---------------------------+--------------------------------------------
>   Reporter:  olemis        |      Owner:  rjollos
>       Type:  task          |     Status:  review
>   Priority:  critical      |  Milestone:  Release 8
>  Component:  multiproduct  |    Version:
> Resolution:                |   Keywords:  product environment testing QA
> ---------------------------+--------------------------------------------
>
> Comment (by olemis):
>
>  Replying to [comment:8 rjollos]:
>  [...]
>  >
>  > A related issue is that, since `process_request` is checking for
>  `TICKET_CREATE` permission, the user must have `TICKET_CREATE` for their
>  current scope in order to use the QCT.
>  >
>
>  this is by design , if creating a ticket via QCT user must be granted with
>  TICKET_CREATE in both the active env and the target env . The former case
>  is not a big deal since
>
>  [source:trunk/bloodhound_theme/bhtheme/templates/bloodhound_theme.html@1553998
> :339-356
>  QCT form is not displayed] . Nevertheless it must still be asserted in
>  code to be consistent in case of direct requests hijacking system logic .
>

We should reconsider whether it is a good design in light of recent
changes. Given that tickets must be associated with a product, it make
little sense to require that a user has `TICKET_CREATE` at global scope in
order for the QCT form to be available at global scope. The QCT would be
more useful if a user could create a ticket in any product for which they
have TICKET_CREATE permission, regardless of the permissions they have in
the current scope (i.e. a the active environment).

Re: [Apache Bloodhound] #388: Test environment isolation across product boundaries - after #355

Posted by Olemis Lang <ol...@gmail.com>.
On Mon, Jan 27, 2014 at 6:18 PM, Joachim Dreimann <
joachim.dreimann@wandisco.com> wrote:

>
> > On 27 Jan 2014, at 23:02, Ryan Ollos <ry...@wandisco.com> wrote:
>
[...]

> >
> > We should reconsider whether it is a good design in light of recent
> > changes. Given that tickets must be associated with a product, it make
> > little sense to require that a user has `TICKET_CREATE` at global scope
> in
> > order for the QCT form to be available at global scope. The QCT would be
> > more useful if a user could create a ticket in any product for which they
> > have TICKET_CREATE permission, regardless of the permissions they have in
> > the current scope (i.e. a the active environment).
>
> I agree with Ryan. That's how I believe QCT was intended to be used from
> the beginning. It should only be hidden if the user has no ticket create
> permissions in any product.
>
>


This is not a trivial check (e.g. single SQL query ...) but it's performed
anyway to filter products list . It makes sense , nonetheless it's not
related to the goal of this ticket #388 . IMO there should be another one
for this feature .

-- 
Regards,

Olemis - @olemislc

Re: [Apache Bloodhound] #388: Test environment isolation across product boundaries - after #355

Posted by Joachim Dreimann <jo...@wandisco.com>.
> On 27 Jan 2014, at 23:02, Ryan Ollos <ry...@wandisco.com> wrote:
> 
> On Mon, Jan 27, 2014 at 10:43 AM, Apache Bloodhound <
> dev@bloodhound.apache.org> wrote:
> 
>> #388: Test environment isolation across product boundaries - after #355
>> ---------------------------+--------------------------------------------
>>  Reporter:  olemis        |      Owner:  rjollos
>>      Type:  task          |     Status:  review
>>  Priority:  critical      |  Milestone:  Release 8
>> Component:  multiproduct  |    Version:
>> Resolution:                |   Keywords:  product environment testing QA
>> ---------------------------+--------------------------------------------
>> 
>> Comment (by olemis):
>> 
>> Replying to [comment:8 rjollos]:
>> [...]
>>> 
>>> A related issue is that, since `process_request` is checking for
>> `TICKET_CREATE` permission, the user must have `TICKET_CREATE` for their
>> current scope in order to use the QCT.
>> 
>> this is by design , if creating a ticket via QCT user must be granted with
>> TICKET_CREATE in both the active env and the target env . The former case
>> is not a big deal since
>> 
>> [source:trunk/bloodhound_theme/bhtheme/templates/bloodhound_theme.html@1553998
>> :339-356
>> QCT form is not displayed] . Nevertheless it must still be asserted in
>> code to be consistent in case of direct requests hijacking system logic .
> 
> We should reconsider whether it is a good design in light of recent
> changes. Given that tickets must be associated with a product, it make
> little sense to require that a user has `TICKET_CREATE` at global scope in
> order for the QCT form to be available at global scope. The QCT would be
> more useful if a user could create a ticket in any product for which they
> have TICKET_CREATE permission, regardless of the permissions they have in
> the current scope (i.e. a the active environment).

I agree with Ryan. That's how I believe QCT was intended to be used from the beginning. It should only be hidden if the user has no ticket create permissions in any product.

- Joe

Re: [Apache Bloodhound] #388: Test environment isolation across product boundaries - after #355

Posted by Olemis Lang <ol...@gmail.com>.
On Mon, Jan 27, 2014 at 6:02 PM, Ryan Ollos <ry...@wandisco.com> wrote:

> On Mon, Jan 27, 2014 at 10:43 AM, Apache Bloodhound <
> dev@bloodhound.apache.org> wrote:
>
> [...]

> >
> > Comment (by olemis):
> >
> >  Replying to [comment:8 rjollos]:
> >  [...]
> >  >
> >  > A related issue is that, since `process_request` is checking for
> >  `TICKET_CREATE` permission, the user must have `TICKET_CREATE` for their
> >  current scope in order to use the QCT.
> >  >
> >
> >  this is by design , if creating a ticket via QCT user must be granted
> with
> >  TICKET_CREATE in both the active env and the target env . The former
> case
> >  is not a big deal since
> >
> >
>  [source:trunk/bloodhound_theme/bhtheme/templates/bloodhound_theme.html@1553998
> > :339-356
> >  QCT form is not displayed] . Nevertheless it must still be asserted in
> >  code to be consistent in case of direct requests hijacking system logic
> .
> >
>
> We should reconsider whether it is a good design in light of recent
> changes. Given that tickets must be associated with a product, it make
> little sense to require that a user has `TICKET_CREATE` at global scope in
> order for the QCT form to be available at global scope. The QCT would be
> more useful if a user could create a ticket in any product for which they
> have TICKET_CREATE permission, regardless of the permissions they have in
> the current scope (i.e. a the active environment).
>

I'm not sure about this , but honestly 0+ .
My interpretation is TICKET_CREATE revoked => do not create tickets in this
product .
AFAICT your interpretation is also valid

OTOH this has ramifications to the ticket I'm working on now #602 aimed at
adding product <select/> input control in /newticket . Again I still think
the form should only be visible if TICKET_CREATE has been granted to the
user , but again I'm 0+ on getting this done otherwise .

-- 
Regards,

Olemis - @olemislc

Apache(tm) Bloodhound contributor
http://issues.apache.org/bloodhound
http://blood-hound.net

Blog ES: http://simelo-es.blogspot.com/
Blog EN: http://simelo-en.blogspot.com/

Featured article: