Re: Code/test ratio wrt static vs dynamic typing

2008-05-21 Thread Bruno Desthuilliers
Ben Finney a écrit : greg <[EMAIL PROTECTED]> writes: Also, I don't think it's valid to equate the size of the tests with the amount of effort it took to develop them. For instance, the test suite for Pyrex is currently larger than the Pyrex compiler, but I've still spent far more time and effo

Re: Code/test ratio wrt static vs dynamic typing

2008-05-20 Thread Ben Finney
greg <[EMAIL PROTECTED]> writes: > Also, I don't think it's valid to equate the size of the tests with > the amount of effort it took to develop them. For instance, the test > suite for Pyrex is currently larger than the Pyrex compiler, but > I've still spent far more time and effort developing th

Re: Code/test ratio wrt static vs dynamic typing [was: Re: Python Success stories]

2008-05-20 Thread greg
Someone wrote: I'm just curious whether this argument against dynamic typing - that you end up doing the job of a static compiler in test code - holds in practice. I suspect that, although some of the things caught by the tests would be caught by static typing, the very *same* tests are also ca

Re: Code/test ratio wrt static vs dynamic typing [was: Re: Python Success stories]

2008-04-30 Thread MRAB
On Apr 30, 10:47 am, [EMAIL PROTECTED] wrote: > > A rather off-topic and perhaps naive question, but isn't a 1:4 > > production/test ratio a bit too much ? Is there a guesstimate of what > > percentage of this test code tests for things that you would get for > > free in a statically typed language

Re: Code/test ratio wrt static vs dynamic typing [was: Re: Python Success stories]

2008-04-30 Thread cokofreedom
> > A rather off-topic and perhaps naive question, but isn't a 1:4 > production/test ratio a bit too much ? Is there a guesstimate of what > percentage of this test code tests for things that you would get for > free in a statically typed language ? I'm just curious whether this > argument against

Code/test ratio wrt static vs dynamic typing [was: Re: Python Success stories]

2008-04-29 Thread George Sakkis
On Apr 29, 2:25 pm, Fuzzyman <[EMAIL PROTECTED]> wrote: > There are around 30 000 lines of Python in the production code and > about 120 000 lines of Python code in the test framework. A rather off-topic and perhaps naive question, but isn't a 1:4 production/test ratio a bit too much ? Is there a