GnuCash Testplan First Draft
warlord at MIT.EDU
Thu Jun 14 11:03:17 EDT 2007
ahmad sayed <ahmadsayed83 at yahoo.com> writes:
> when writing a test case we always has setup and teardown, the both these methods will be different according to the testcase, some testcase will need to have a gnucash without any data file, others will require this data file, these operation will be done in the setup,
> in general
> the basic testcase will be
> setup() # do the required setup (e.g start gnucash if not)
> teardown() # do the required clean up close gnucash, kill gnucash process, .... )
> setup will be different from test case to other, in some cases we need to open gnucash without a file, In other cases will have gnucash open with a loaded data file, I miss writing the setup for these test cases, it will be fixed in the next test plan version.
Note that these setups should be shared across multiple tests. I.e.,
you shouldn't need to write the "create a random data file" multiple
times. And we have some C code that does a lot of this already
which you might want to leverage.
> Testing the register,
> Regardless our ability to test the register it is relatively important, but how it could be done
> 1 - Create a custom class that wrap some raw events, this class used to do some higher level actions
> def write_something_on_cell(x ,y, something):
> press enter key for x times
> press tab key for y times.
> raw type (something)
> also use all possible accelerator and hot keys.
> 2 - Modify the widget code to be accessible, I prefer this but not sure how feasible is it? ,
That's the big question.. It's written on a GnomeCanvas and.. quite
annoying. The new register-rewrite will make it easier to test,
obvioualy, but that's not finished yet.
> Best regards,
> Ahmed Sayed
Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
Member, MIT Student Information Processing Board (SIPB)
URL: http://web.mit.edu/warlord/ PP-ASEL-IA N1NWH
warlord at MIT.EDU PGP key available
More information about the gnucash-devel