understanding sys.argv[]

[Steve Holden]

"Don Low" <mt at open2web.com> wrote ...
> >> OK, so the first thing I do is import getopt. The getopt module is for
> >> parsing command line options so I guess it makes sense to import it.
Next, I
> >> modify
> >>
> >> if argc != 3:
> >>          usage()
> >>
> >> to
> >>
> >> if argc != 4:
> >>          usage()
> >>
> >> since there's 4 arguments now.
> >
> > Not always. -i is an *option*... maybe it's specified, maybe not.
>
> OK, then I would do
>
> if argc == 3 || argc == 4

That will work in this case. However, in general you may want to implement a
number of options for any given program. Since by their nature these are
optional, checking the argumnet count is often best left until you've
processed the options (see below).

> >
> > Then there's the following:
> >>
> >> filefind(sys.argv[1], sys.argv[2]).
> >>
> >> The author says "the command-line arguments are stored in sys.argv
list."
> >> That makes sense. "The first argument is the program name and
presumably,
> >> the second is the string we are looking for, and the final argument is
the
> >> name of the file to search." Wait, there are 3 arguments but the code
only
> >> lists two. So I add a third one as in:
> >>
> >> filefind(sys.argv[1], sys.argv[2], sys.argv[3])
> >>
> >> This doesn't work, so I try:
> >>
> >> filefind(sys.argv[1], sys.argv[3]).
> >>
> >> This works, although honestly I don't know why. Does the sys.argv[x]
where x
> >> equals a number signify 1st arg, 2nd arg, etc, or what?  I don't get
this.
> >
> > sys.argv is a list of the command line arguments passed to the program.
> > sys.argv[0] is the name of the program itself; sys.argv[1] and further
are used
> > for the arguments.
>
> How come sys.argv[0] isn't mentioned? Is it implicit?  In this case
>
> sys.argv[1] represents the word I want to look up
> sys.argv[2] reps name of the document in which I want to look up the word.

sys.argv[0] is the path/name of the script you are running. Sometimes it's
useful, but most often you just want to ignore it. So sys.argv[1] is the
first argument, and so on. That's why you use sys.argv[1:] below ... this is
all arguments except argument zero.

> >
> > If you use getopt, as the author suggests, things should be a bit
easier, since
> > you don't have to figure out where exactly the -i switch is in the list,
or if
> > it's there at all. Instead, getopt.getopt() returns two lists, one of
> > options/switches, and one of "normal" arguments. See the documentation:
> >
> >    http://www.python.org/doc/current/lib/module-getopt.html
> >
> OK, so if I understand correctly, I'd have something like this
>
> opts, args = getopt.getopt(sys.argv[1:], "i:")
>
> Not sure why 2 variables (opts and args) get initialized with getopt...

This is because getopts returns two things: first there's the options,
represented as a list of (option, value) pairs. Since the "-i" option
doesn't take a value, when it's present then opts will receive [("i",
None)], when it's absent opts will receive []. Second there's the remaining
arguments, which presumably have to be processed by the program (in your
case, as file names).

>
> def checkargs():
>
>     argc = len(sys.argv)
>
>     if argc == 3 || argc == 4 :
>
> filefind(getopt.getopt(sys.argv[1:], "i"), sys.argv[2])
>
>     else:
>         usage()
>
>
> Still haven't figured out how the -i parameter allows a case-insensitive
> search.
>
It doesn't. You are supposed to detect the presence or absence of the "-i"
command-line option, and pass this as an additional parameter to filefind().
In turn, filefind()'s logic has to be modified so that when the parameter is
true it operates in a case-insensitive way.

regards
-----------------------------------------------------------------------
Steve Holden                                 http://www.holdenweb.com/
Python Web Programming                http://pydish.holdenweb.com/pwp/
-----------------------------------------------------------------------