abdera-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Martin Cooper" <mart...@apache.org>
Subject Re: Abdera Unit Tests and Network Access
Date Sun, 18 Jun 2006 16:51:38 GMT
On 6/18/06, Garrett Rooney <rooneg@electricjellyfish.net> wrote:
>
> One of the things that kind of bothers me about the current Abdera
> unit tests is that they require network access to function.  It's
> basically impossible (assuming that we have the tests running, which
> we don't for the ant build right now, but that's a separate issue) to
> run the tests (and by extension to build a release, since one step in
> building a release is running the tests) if you don't have unfetterred
> access to the internet.  It also means that our tests can now fail due
> to conditions totally outside our control.


If that's the case, then *they are not unit tests*! They are functional or
integration tests. It really bugs the crap out of me when people mix these
up.

I'd like to start working on bringing the Atom files we depend on into
> our source tree, at least the ones that are under a sufficiently clear
> license (Mark Pilgrim's Feed Parser tests, for example, are clearly
> under the Python Software Foundation license, which AFAICT is just a
> standard BSDish license and thus we can use that code as long as we
> clearly reproduce its license in our NOTICE file and don't remove any
> copyright statements).
>
> Does anyone have an objection to this?  If we want to preserve the
> ability to fetch the files from their cannonical locations online I
> can certainly look at a more complex system that provides that
> ability, but honestly it seems like the point here is to test our
> ability to parse the content, not the JDK's ability to download a file
> via http.


I'm definitely in favour of isolating the tests as much as possible.  Even
functional and integratoin tests. ;-)

--
Martin Cooper


-garrett
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message