On Monday, July 25, 2016 at 8:42:21 AM UTC+5:30, Chris Angelico wrote: > On Mon, Jul 25, 2016 at 12:48 PM, Michael Torrie wrote: > > Far more often I'm bitten by the dynamic nature of Python (would happen > > in any dynamic language). I'll be using a particular member attribute > > which I accidentally misspell somewhere and sometimes that results in > > silent failure. Something doesn't work, but no exception is thrown. > > Unit tests, and perhaps lint, are required to catch these errors. That > > is one thing about a dynamic language: comprehensive testing is required > > as you go along. > > > > To be quite honest, comprehensive testing is needed in more static > languages too. There are certain categories of error which can be > detected by a compiler/linter, and certain which cannot; a language > that forces you to declare variables will catch variable name > misspellings, but only if they don't land you on an existing variable, > and still won't catch the dynamic places like dict keys (imagine > getting a block of JSON from somewhere, converting it into a > dictionary, and looking up stuff in it - the compiler can't know > whether your incoming data is correct and your code wrong, or the > other way around). A language with less dynamism might be able to > catch more, but still not everything,
So far — Fine! > so ultimately, it all comes down to testing anyway. All?? There is a famous quote by Dijkstra: «Testing shows the presence, not the absence of bugs» Or if you prefer things of a more ‘practical’ (so-called_ nature: http://www.testingexcellence.com/reasons-automated-tests-fail-to-find-regression-bugs/ -- https://mail.python.org/mailman/listinfo/python-list
