Mailing-List: contact bloodhound-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: bloodhound-dev@incubator.apache.org
Received-SPF: pass (nike.apache.org: domain of olemis@gmail.com designates
 209.85.223.172 as permitted sender)
MIME-Version: 1.0
Date: Mon, 18 Feb 2013 01:10:47 -0500
Message-ID: 
 <CAGMZAuPRHQe6iU7e7uDOmEmdn2QeGjYFyeXdAtrqLWTbnqTnGQ@mail.gmail.com>
Subject: [BEP-0003] #355 - Summary of the work done towards multi-product
 support (2013-02-16)
From: Olemis Lang <olemis@gmail.com>
To: bloodhound-dev@incubator.apache.org
Content-Type: text/plain; charset=ISO-8859-1

Hi !

Starting this week and I'll dedicate some time to write a summary of
the current state of ticket #355 , which is the first milestone
towards multi-product (MP) support . In advance I considr that this
message might be important to all persons interested in the project .
Along the way I mention some important TODOs in the roadmap . So here
I go :

Up to this point there is a subset of Trac test suite that may be run
against product environments . It consists of 794 test cases (334
success , 30 failures , 430 errors) . The high number of failures is
mainly a consequence of the fact that recently I've just prioritized
TC conversion rather than making them work . The number of successful
TCs has always increased in time , as I still paid attention to track
most regressions , so there is no need to worry or over react about
that at this point .

TODO-1: Immediate conclusion => there is a lot to fix yet .

For a more detailed report and trendings please see this spreadsheet [1]_ .

What is in the test suite ?

Only unit tests making use of product environments . That leaves
functional TCs (see #387) , doctests (see #385) and a few other test
cases (see #396) out of the equation . We'll deal with those
afterwards . In terms of subsystems the situation is as follows :

  - Ticket TCs are almost done
    * TODO-2: There's an urgent need to implement something important we are
      missing, and it is product-specific ticket sequence.
      At the moment ticket ID in product scope is taken as the global
      (i.e. unique) ticket ID , which is not accurate . Most test cases
      pass but the underlying implementation is ignoring this particular
      enhancement.
    * TODO-3: The only missing piece of the puzzle are wiki syntax TCs
(see #390)
      which MUST make explicit use of the output of TODO-2
    * TODO-1.1: make query TCs pass
    * Functional TCs postponed until a later iteration
  - All relevant TCs for the wiki subsystem have been translated
    * TODO-1.2: Some test cases for wiki formatter , macros and
      syntax are failing .
    * Functional TCs postponed until a later iteration
  - All relevant TCs for admin console have been translated
    * TODO 1.3: fix failures
    * Functional TCs postponed until a later iteration
  - All relevant TCs for MIME API have been translated . They seem to be ok.
  - There are no unit tests among preferences TCs , so they won't be
    considered in this iteration (unless otherwise stated) .
  - There are no tests for built-in search but this is not a concern as
    our free-text search plugin does a good job on this direction.
  - A relevant subset of TCs for core features has been translated (i.e.
    attachments, configuration, environments, notification, permissions)
    * TODO-4: Document and implement product-specific resources
      hierarchy
    * TODO-5: Translate Trac resources and wiki syntax TCs
  - There are no unit tests among timeline TCs , so they won't be
    considered in this iteration (unless otherwise stated) .
  - TCs for util package are at a lower level of abstraction so
    they won't be considered in this iteration (unless otherwise stated) .
  - TODO-6: Translate all TCs for Trac version control subsystem
  - TODO-7: Translate all TCs for Trac web API


@jure : if product-specific ticket ID sequences will be part of MP RC1
it'd be very helpful if you could start working on TODO-2 in parallel
as I make some progress working on TODO-1 , TODO-3 ... TODO-n . Now
you'll have enough TCs to detect regressions .

Happy hacking !


.. [1] Trend of Trac test case conversion towards Bloodhound
multi-product support
       (https://docs.google.com/spreadsheet/ccc?key=0AqniOh7YBqF4dHp6Sk0zWkNxS2FpMjZHQWllbmhYMnc#gid=0)


-- 
Regards,

Olemis.