Compare Python with Perl for QA purpose
claird at lairds.com
Sat Jan 4 19:05:44 CET 2003
In article <mailman.1041696487.26472.python-list at python.org>,
Bjorn Pettersen <BPettersen at NAREX.com> wrote:
>> From: Amol Sonaikar [mailto:amol at kenati.com]
>> Hi all
>> I am using Python to test the Our created Command Line
>> Interface. I would
>> like to know whether Python will be better option to work on
>> it or Perl. I
>> guess the python users can very well elaborate on it.
>Isn't this the canonical use case for Expect, i.e. Tcl? Python has of
>course developed features in that direction to, although still not quite
>as polished from what I understand (never used them myself, i.e. this is
>hearsay <wink>). In any case, your question should probably be: "Is
>either Perl or Python close enough to Tcl/Expect functionality to test
>our CLI and do they have other properties that would make up for any
>lack of specific features?" The answer, of course, depends on what your
Expect <URL: http://wiki.tcl.tk/Expect > is indeed the canon-
ical answer for test automation <URL: http://
I'm expert in Expect, and a big fan. HOWEVER, its advantages
over Perl and Python are commonly overestimated (and underesti-
mated, in other directions, but that's a different story).
There are many QA situations where Expect has no significant
advantage over Perl and Python.
Mr. Petersen is also right in emphasizing that "QA of command-
line interface" leaves a lot of free variables. At some point,
Mr. Sonaikar will need to decide whether a GUI for his automa-
tions is an advantage, whether his "Command Line Interface"
does tty-mangling, what kind of portability he requires, what
synchronicity, concurrency, and timing issues he has, and so on.
General summary: Python and Perl are roughly equally well suited
to the broad domain of QA.
Cameron Laird <Cameron at Lairds.com>
More information about the Python-list