I was thinking why not Mart?! ;-)

There is no need to rush. I would wait for more feedback and Massimo
ideas on this, in the midtime we can use googlecode better,
while *eventually* starting a web2py centric bugtracker...


2010/8/25 mart <msenecal...@gmail.com>:
> What a great thing! Now we just need to define requirements... Someone
> should lead this to avoid problems... (N cooks in one kitchen where N
> should = 1) Michele is it? :)
>
> Mart :)
>
> On Aug 25, 1:52 pm, Michele Comitini <michele.comit...@gmail.com>
> wrote:
>> *plugin_wiki*  could be used as a basis for this eventual bugtraker?
>>
>> 2010/8/25 mart <msenecal...@gmail.com>:
>>
>> > yes, in any bug tracking system, "work on bug" data is provided by
>> > assigned dev user (status/fix description/fix available in build X/
>> > etc.).
>>
>> > Mart :)
>>
>> > On Aug 25, 1:23 pm, Alexandre Andrade <alexandrema...@gmail.com>
>> > wrote:
>> >> Since we can se the web2py tickets as bugs (of our apps), its be nice to
>> >> incorporate a management of this tickets, not only registering them.
>>
>> >> 2010/8/25 mart <msenecal...@gmail.com>
>>
>> >> > bug tracking app is available on web2py.com? I think opening a bug
>> >> > should be the first that happens.... possible to add validation code
>> >> > within the bug tracking as a first layer filter (is this /or hast this
>> >> > ever been a bug ?) or... perhaps an easy way for users for query the
>> >> > db.bug_history (perhaps optimal in this case?)
>>
>> >> > Mart :)
>>
>> >> > On Aug 25, 1:10 pm, Michele Comitini <michele.comit...@gmail.com>
>> >> > wrote:
>> >> > > Point 2) is nice for all of us because you (Massimo) are very quick,
>> >> > > but how much does take off of your web2py time?
>> >> > > Would not be better to open the ticket before testing and eventually
>> >> > > close it as "works for me"?
>> >> > > This way someone among the developers  could take care of the ticket
>> >> > > and test it, if he is able to fix it good, he makes a patch
>> >> > > and he closes it when the patch is put in the trunk (by you).  In case
>> >> > > the patch cannot be applied either you have time to fix it
>> >> > > or inform the submitter to fix it.
>>
>> >> > > A slight modified version of the process (very imperfect):
>>
>> >> > > 1) people post a question about a possible bug
>> >> > > 2) if it looks (without test) like it, you (or a developer) ask them
>> >> > > to open a ticket
>> >> > > 3) you or a developer take care of the ticket (becoming the ticket
>> >> > > holder) locking others out
>> >> > > 4) the holder tests: if it is not a bug then 6)
>> >> > > 5) you fix it in trunk or a developer sends you a patch
>> >> > > 5.1) if you cannot apply the patch in trunk then 5) again
>> >> > > 6) the ticket holder closes the ticket making a reference to the
>> >> > > revision in trunk.
>>
>> >> > > Point 1) and 2) can be optional? could a user open the ticket right 
>> >> > > away?
>>
>> >> > > For me a plus of a ticket system would be the automatic assignement of
>> >> > > tickets to developers based on the area of the problem
>> >> > > or some other criteria.
>>
>> >> > > Of course there is some work for Massimo... for instance finding
>> >> > > stalled tickets and bashing lazy developers ;)
>> >> > > One advantage would be that users can search for similar bugs in
>> >> > > googlecode and see that they are already fixed at
>> >> > > some revision and would check that they have updated their copy of
>> >> > > web2py before asking.
>> >> > > Also the changelog of a stable release could include a list of closed
>> >> > > tickets (do not know how on googlecode, but *there must be some
>> >> > > way*!!).
>>
>> >> > > BTW: patch generation should be something with a procedure by itself,
>> >> > > using plain files or others means such as mercurial
>>
>> >> > > mic
>>
>> >> > > 2010/8/25 mdipierro <mdipie...@cs.depaul.edu>:
>>
>> >> > > > we do use googlecode for not. Here is the current (imperfect) 
>> >> > > > process:
>>
>> >> > > > 1) people post a question about a possible bug
>> >> > > > 2) somebody checks that it is a bug, usually me
>> >> > > > 3) if the bug does not get fixed in 24h, the original poster opens a
>> >> > > > googlecode ticket
>> >> > > > 4) when the bug is fixed the ticket is closed
>>
>> >> > > > because many bugs are dealt with in <24hrs there is no record. 
>> >> > > > Because
>> >> > > > bugs are fixed in trunk and trunk takes a couple of weeks to become
>> >> > > > steable and because most users never upgrade to the latest stable,
>> >> > > > occasionally there are multiple questions related to the same fixed
>> >> > > > bug. I am not sure better workflow management fixes this latter
>> >> > > > problem.
>>
>> >> > > > I have not read all messages on this thread carefully yet. 
>> >> > > > Eventually
>> >> > > > I will but I am happy to hear more of your ideas.
>>
>> >> > > > On Aug 24, 5:52 pm, Michele Comitini <michele.comit...@gmail.com>
>> >> > > > wrote:
>> >> > > >> Actually I would like to ask if bug tracking is used on web2py?
>>
>> >> > > >> Code is available from either (btw Massimo how do you keep those 2 
>> >> > > >> in
>> >> > > >> sync? just too curious :-) ):
>> >> > > >> a) googlecode (with hg)
>> >> > > >> b) launchpad (with bzr)
>>
>> >> > > >> both have some sort of bugtracking ticket system I do not know 
>> >> > > >> which
>> >> > > >> one is best (or worst),  we could start with one those, but
>> >> > > >> the choice must taken with care and other systems must be evaluated
>> >> > > >> (on: usability, independece, web2py phylosophy ...), and first
>> >> > > >> they must meet Massimo needs.
>>
>> >> > > >> BTW: I would like to see  a web2py application for doing serious
>> >> > > >> bugtracking in the future... so that submitting
>> >> > > >> a bug would be just one click on the ticket reported by any web2py
>> >> > > >> installation! mmm too easy... that would be dangerous! ;-)
>>
>> >> > > >> ciao,
>> >> > > >> mic
>>
>> >> > > >> 2010/8/24 mart <msenecal...@gmail.com>:
>>
>> >> > > >> > I don't know if you are currently using a specific bug tracking
>> >> > > >> > system, but they are typically easy to interface with and made 
>> >> > > >> > part
>> >> > of
>> >> > > >> > build/release & test processes/automation. I.e. As part of a 
>> >> > > >> > release
>> >> > > >> > process, I would set rules with the source control system where 
>> >> > > >> > non-
>> >> > > >> > bugTraking releated changes can either be automatically 
>> >> > > >> > rejected, or
>> >> > > >> > moved to another set of prioritiesArea, etc... the build (or
>> >> > packaged
>> >> > > >> > fileset, or whatever the output is) contains a detailed 
>> >> > > >> > inventory of
>> >> > > >> > bug fixes/features/etc... as part of an automated delivery system
>> >> > > >> > (these are part of the build notes)...
>>
>> >> --
>> >> Atenciosamente
>>
>> >> --
>> >> =========================
>> >> Alexandre Andrade
>> >> Hipercenter.com

Reply via email to