Re: delete columns csv file

Ed Morton wrote:

> Depends if by "it" in "it considers.." you mean "print" or "awk".
> 
> Awk considers "a" to have the value 0 or "" depending on the context in
> which you use it, just like with input data. "print" prints strings. A
> numeric argument is converted to a string before printing, so printing
an
> uninitialised variable (which can be a string or a numeric) using print
> (which prints strings) means that the variable is taken to be a string
IN
> THAT CONTEXT.
> 
> So "print" considers a to have the value "" but "awk" still considers a
to
> have the value "" or 0.

After the other replies I went reading the standard, and saw that this is
exactly what it (the standard :-)) says. 
What I meant to say in my previous post was that *I* mistakenly believed
that variables are strings by default based on the outcome of "print"
which, actually, is required to convert its arguments to strings, so I was
basing my understanding upon wrong premises.
Thanks for further clarification.

> Yeah, there's a lot to think about when dealing with string/numeric
> conversions and comparisons using variables vs explicit strings vs
> explicit numbers vs input data. Even if you understand it at the time,
> it's often best to write your code to force explicit conversions so it's
> clearer later...

And yes, another thing I was missing is that "0.0e0" is among those
strings
awk considers that "look like a number" (as you called them in your
previous post), and this is why I wrote my other answer to Hermann (again,
based upon wrong understanding).
Thus, now I see that your "" to force awk to treat it like a string was
correct.

Thanks again.

-- 
All the commands are tested with bash and GNU tools, so they may use
nonstandard features. I try to mention when something is nonstandard (if
I'm aware of that), but I may miss something. Corrections are welcome.