Q: Analyse data and provide a report - Arrays?

[Troll]

Hi,

I need to write a script which reads some data and reports the findings.
Just to give you an idea the structure is similar to the following.

Data input example:

HEADING 1
**********
ColumnA ColumnB ColumnC ColumnD ColumnE
Pete Male Marketing Single 40
Kate Female Marketing Married 30
John Male Sales Married 38
Pete Male Sales Single 52
John Male Sales Single 24


HEADING 2
**********
ColumnF ColumnG ColumnH ColumnI
whatever
whatever
whatever
whatever


Report Output example:
# of Pete's =
# of Males =
# of Salespeople =
# of Singles =
# of over 35s =


Since this is the first time I'm even writing such a script I would
appreciate some pointers.
1) Do I use arrays or associate arrays for this? Why or why not?
2) Is it possible for someone to give me a code example of counting how many
Singles we have?
3) What happens when I have read all the data under HEADING 1 and need to
move onto HEADING 2?
That is, how do I accomplish the jump from what I think is one loop onto the
next?

I imagine that there will be many more posts following this one so there's
no need to get into too much detail. Some guidance would be nice as I will
need to utilise Google and my references for the rest.

Thanks in advance.

 

[Troll]

Ga Mu,
Great stuff - thanks very much.

The headings differentiate blocks of data so once we count everything under
HEADING 1 we move onto HEADING 2 then HEADING 3 etc.

Does this help a bit?

 

[Troll]

Ga Mu,

Pls disregard last post.

With regard to the jump between HEADINGS, will it be enough to do something
like:
while (<>)
....
if (/HEADING 1/ .. /HEADING 2/) {
# line falls between HEADING 1 and HEADING 2 in the text, inclusive.
# then do the string extraction
# then increment stuff
elsif (/HEADING 2/ .. /HEADING 3/) {
# line falls between HEADING 2 and HEADING 3 in the text, inclusive.
# then do the string extraction
# then increment stuff
etc?

I quite like the code example you provided - actually found a similar one in
http://www.oreilly.com/catalog/perlwsmng/chapter/ch08.html
Up until now I was under the impression that I would have to use split - can
you elaborate why you chose a different approach?

One other task I have to do is similar to:

If a line contains Single in the column then get the single person's name.

I sort of came up with:
foreach $m_statuses{'Single'}
print $names{$name}

but that's probably totally wrong. Can you advise?

Thanks again.

 

[Ga Mu]

Ga Mu,

Pls disregard last post.

With regard to the jump between HEADINGS, will it be enough to do something
like:
while (<>)
...
if (/HEADING 1/ .. /HEADING 2/) {
# line falls between HEADING 1 and HEADING 2 in the text, inclusive.
# then do the string extraction
# then increment stuff
elsif (/HEADING 2/ .. /HEADING 3/) {
# line falls between HEADING 2 and HEADING 3 in the text, inclusive.
# then do the string extraction
# then increment stuff
etc?

I am unclear as to the distinction between blocks. Are there a separate
group of totals for each heading or is everyting totalled up together?
If the latter, then simply ignore the headings. If the former, then you
could parse out the heading name and use a multidimensional hash. I.e.,
replace this:

$names{$name}++;

with this:

$names{$heading}{$name}++;

I quite like the code example you provided - actually found a similar one in
http://www.oreilly.com/catalog/perlwsmng/chapter/ch08.html
Up until now I was under the impression that I would have to use split - can
you elaborate why you chose a different approach?

Either method produces the same results. If you plan on incorporating
error checking, m// allows to specifically define a format, e.g., four
words and a number, whereas split simply breaks a string up into a list.
Whichever method makes you happy.

One other task I have to do is similar to:


I sort of came up with:
foreach $m_statuses{'Single'}
print $names{$name}

but that's probably totally wrong. Can you advise?

Yes, it is totally wrong. $m_statuses{'Single'} is a scalar. It is the
count of lines where the marital status is 'Single'. Your foreach loop
above would produce a syntax error. Although it is not what you're
after, a valid foreach loop could look like this:

foreach $m_status ( keys %m_statuses ) {
#
# $m_status will be 'female' for one iteration of the loop and 'male'
# for the other. (Unless you have more than two sexes...)
#
}

Perhaps a more meaningful foreach loop would look like this:

foreach $age ( keys %ages ) {
#
# For each iteration, $age will one the ages that was found in the data
# -->> IN NO PARTICULAR ORDER <<-- unless you sort it.
#
}

To do what you propose, i.e., print the name of all single people, you
would have to include the logic for that in the parsing loop:

# extract four words and a number into scalars:
my ($name, $sex, $dept, $m_status, $age) =
/^(\w+) (\w+) (\w+) (\w+) (\d+)$/;

# increment counts for each:
$names{$name}++;
$sexes{$sex}++;
$depts{$dept}++;
$m_statuses{$m_status}++;
$ages{$age}++;

# take special actions:
if ( $m_status eq 'Single' ) print "$name is single.\n";
if ( $age >= 40 ) print "$name is over the hill!\n";


Hope this helps!

Greg

 

[Troll]

Thanks again !

1)
Sorry for being too vague. With regard to the HEADINGS they separate blocks
of data. But because the column names will be different [data is different]
then I'm not quite sure I could use:
$names{$heading}{$name}++;

So I'm looking at creating separate my () definitions for each HEADING and
just wanted to confirm how to jump out of one HEADING loop and start with
the next.

For example, under HEADING 1 we have these columns:
Name, Sex, Dept, M_Status, Age

and under HEADING 2we have:
Address, Phone#, Mobile#, Salary

So at the beginning of the script I would have
my (%names, %sexes, %depts, %m_statuses, %ages)
my (%addresses, %phones, %mobiles, %salaries)
#then I have my while (<>) and parsing here
#I have my output at the end

Is that a little more clearer?


2)
With my last question regarding the printing of the names of single people,
if we include a print statement in the parsing loop would that give us
something like:
Pete is single.
John is single.
while the parsing is still running?

What I'm after is hopefully feeding that output into something else
[@array?] which can then print a list of the names [line by line] at the end
of the script, something like:
#this is the output structure
Number of Petes =
Number of Males =
Singles are:
Pete
John
Number of Salespeople =


Does this make sense?

Thanks Greg.

 

[Ga Mu]

Thanks again !

1)
Sorry for being too vague. With regard to the HEADINGS they separate blocks
of data. But because the column names will be different [data is different]
then I'm not quite sure I could use:
$names{$heading}{$name}++;

So I'm looking at creating separate my () definitions for each HEADING and
just wanted to confirm how to jump out of one HEADING loop and start with
the next.

For example, under HEADING 1 we have these columns:
Name, Sex, Dept, M_Status, Age

and under HEADING 2we have:
Address, Phone#, Mobile#, Salary

So at the beginning of the script I would have
my (%names, %sexes, %depts, %m_statuses, %ages)
my (%addresses, %phones, %mobiles, %salaries)
#then I have my while (<>) and parsing here
#I have my output at the end

Is that a little more clearer?

Yes. Much clearer. There are a couple of different ways you could do
this. One is to use a single loop that reads through the file and uses
a state variable (e.g., $heading) to keep track of where you are in the
parsing process. The other is to have a separate loop for each heading.
Again, six of one, half a dozen of another. It's more a matter of
preference than anything else.

An example of the first approach:

my $heading = 'initial';
my $fin_name = '/usr/local/blah/blah/blah';
open FIN,$fin_name || die "Can't open $fin_name\n";

while (<FIN>) {

# check for a new heading
# I am assuming single word heading names
if ( /HEADING (\S+)/ {

$heading = $1; # set $heading equal to word extracted above

# take appropriate action based on the heading we are under

} elsif ( $heading eq 'NAMES' ) {

( $name, $sex, $dept, $m_status, $age ) =
/(\w+) (\w+) (\w+) (\w+) (\d+)/;

# update counts, append to lists, etc...

} elsif ( $heading eq 'ADDRESSES' ) {

# I am assuming the address field is limited to 30 characters
# here:
( $address,$phone, $mobile, $salary ) =
/(\.{30}) (\S+) (\S+) (\d+)/;

# update counts, append to lists, etc...

}

}


And the second approach:

my $heading = 'initial';
my $fin_name = '/usr/local/blah/blah/blah';
open FIN,$fin_name || die "Can't open $fin_name\n";

# scan for first heading
while ( <FIN> && ! /HEADING NAMES/ );

# parse the names, etc...
while ( <FIN> && ! /HEADING ADDRESSES/ ) {

( $name, $sex, $dept, $m_status, $age ) =
/(\w+) (\w+) (\w+) (\w+) (\d+)/;

# update counts, append to lists, etc...


# parse the addresses, etc...
# for brevity , I am assuming only two headings
while ( <FIN> ) {

( $address,$phone, $mobile, $salary ) =
/(\.{30}) (\S+) (\S+) (\d+)/;

# update counts, append to lists, etc...

}

2)
With my last question regarding the printing of the names of single people,
if we include a print statement in the parsing loop would that give us
something like:
Pete is single.
John is single.
while the parsing is still running?
Yes.


What I'm after is hopefully feeding that output into something else
[@array?] which can then print a list of the names [line by line] at the end
of the script, something like:
#this is the output structure
Number of Petes =
Number of Males =
Singles are:
Pete
John
Number of Salespeople =


Does this make sense?

Yes. It would be easy to create a list/array of, e.g., single people.
Prior to the loop, declare the array. Within the loop, test each person
for being single. If they are, push them onto the list:

# prior to your parsing loop, declare array @singles:

my @singles;

# within your parsing loop, after parsing out name, status, etc.:

if ( $m_status eq 'Single' ) push @singles,($name);

# after loop, to print the list of singles:

print "Single persons:\n";
foreach $single_person ( @singles ) print " $single_person\n";


Greg

 

[Troll]

Wow. I don't know how you get the time to respond to my queries in such
detail. It is greatly appreciated.
I just came back from work and it's like 2:30 am so I'll crash out soon and
have a closer read tomorrow [especially of the HEADINGS part].

With the push @array stuff I actually got to this today in my readings. I
saw an example of appending an array onto another array with a push and I
was wondering if we could just substitute a $variable for one of the arrays.
I'm glad you confirmed this.

I was also wondering if doing this at the beginning of the script:

my (%names, %sexes, %depts, %m_statuses, %ages) # declaring things
locally

would be considered bad practice. I thought that one should declare things
as my ( ) if one is using things within a loop so as not to impact anything
external to the loop. But if one uses variables/arrays both within and
outside the loops, should we then still declare stuff as my ( )?
Maybe I'm just confused about my ( )...

Greg, if you could possibly keep an eye on this thread for the next few days
I would be very much in your debt. Your help has been invaluabe so far in
allowing me to visualise quite a few things.

Thanks very much.

Thanks again !

1)
Sorry for being too vague. With regard to the HEADINGS they separate blocks
of data. But because the column names will be different [data is different]
then I'm not quite sure I could use:
$names{$heading}{$name}++;

So I'm looking at creating separate my () definitions for each HEADING and
just wanted to confirm how to jump out of one HEADING loop and start with
the next.

For example, under HEADING 1 we have these columns:
Name, Sex, Dept, M_Status, Age

and under HEADING 2we have:
Address, Phone#, Mobile#, Salary

So at the beginning of the script I would have
my (%names, %sexes, %depts, %m_statuses, %ages)
my (%addresses, %phones, %mobiles, %salaries)
#then I have my while (<>) and parsing here
#I have my output at the end

Is that a little more clearer?

Yes. Much clearer. There are a couple of different ways you could do
this. One is to use a single loop that reads through the file and uses
a state variable (e.g., $heading) to keep track of where you are in the
parsing process. The other is to have a separate loop for each heading.
Again, six of one, half a dozen of another. It's more a matter of
preference than anything else.

An example of the first approach:

my $heading = 'initial';
my $fin_name = '/usr/local/blah/blah/blah';
open FIN,$fin_name || die "Can't open $fin_name\n";

while (<FIN>) {

# check for a new heading
# I am assuming single word heading names
if ( /HEADING (\S+)/ {

$heading = $1; # set $heading equal to word extracted above

# take appropriate action based on the heading we are under

} elsif ( $heading eq 'NAMES' ) {

( $name, $sex, $dept, $m_status, $age ) =
/(\w+) (\w+) (\w+) (\w+) (\d+)/;

# update counts, append to lists, etc...

} elsif ( $heading eq 'ADDRESSES' ) {

# I am assuming the address field is limited to 30 characters
# here:
( $address,$phone, $mobile, $salary ) =
/(\.{30}) (\S+) (\S+) (\d+)/;

# update counts, append to lists, etc...

}

}


And the second approach:

my $heading = 'initial';
my $fin_name = '/usr/local/blah/blah/blah';
open FIN,$fin_name || die "Can't open $fin_name\n";

# scan for first heading
while ( <FIN> && ! /HEADING NAMES/ );

# parse the names, etc...
while ( <FIN> && ! /HEADING ADDRESSES/ ) {

( $name, $sex, $dept, $m_status, $age ) =
/(\w+) (\w+) (\w+) (\w+) (\d+)/;

# update counts, append to lists, etc...


# parse the addresses, etc...
# for brevity , I am assuming only two headings
while ( <FIN> ) {

( $address,$phone, $mobile, $salary ) =
/(\.{30}) (\S+) (\S+) (\d+)/;

# update counts, append to lists, etc...

}

2)
With my last question regarding the printing of the names of single people,
if we include a print statement in the parsing loop would that give us
something like:
Pete is single.
John is single.
while the parsing is still running?
Yes.


What I'm after is hopefully feeding that output into something else
[@array?] which can then print a list of the names [line by line] at the end
of the script, something like:
#this is the output structure
Number of Petes =
Number of Males =
Singles are:
Pete
John
Number of Salespeople =


Does this make sense?

Yes. It would be easy to create a list/array of, e.g., single people.
Prior to the loop, declare the array. Within the loop, test each person
for being single. If they are, push them onto the list:

# prior to your parsing loop, declare array @singles:

my @singles;

# within your parsing loop, after parsing out name, status, etc.:

if ( $m_status eq 'Single' ) push @singles,($name);

# after loop, to print the list of singles:

print "Single persons:\n";
foreach $single_person ( @singles ) print " $single_person\n";


Greg

 

[Troll]

Now time for some stupid Qs:

Let's say that the data I have is in a file called employees.
How can I call this file so that I can parse it?

1) Can I do:
@HRdata = `cat employees`;
while (<@HRdata>) {


2) With regard to the HEADING sections, the script has to be able to
recognise the different sections by the following rules:
# there's a blank line
before each heading
HEADING 1 # this is the name of the heading -
this is a string with a special character and a blank space as part of it
ColumnA ColumnB ColumnC # these are the column names - these are
strings which also can inlude a blank space if they have 2 or more words
******* # a sort of an underlining
pattern

I guess this is to make sure that one does not include any silly heading
data as part of the arrays created and the parsing only takes place on
'real' data. Can you pls advise? Or do you need more info? I'm more in
favour of creating separate 'if' loops due to my 'newbie' status. I'll get
lost otherwise...

Thanks.

Wow. I don't know how you get the time to respond to my queries in such
detail. It is greatly appreciated.
I just came back from work and it's like 2:30 am so I'll crash out soon and
have a closer read tomorrow [especially of the HEADINGS part].

With the push @array stuff I actually got to this today in my readings. I
saw an example of appending an array onto another array with a push and I
was wondering if we could just substitute a $variable for one of the arrays.
I'm glad you confirmed this.

I was also wondering if doing this at the beginning of the script:

my (%names, %sexes, %depts, %m_statuses, %ages) # declaring things
locally

would be considered bad practice. I thought that one should declare things
as my ( ) if one is using things within a loop so as not to impact anything
external to the loop. But if one uses variables/arrays both within and
outside the loops, should we then still declare stuff as my ( )?
Maybe I'm just confused about my ( )...

Greg, if you could possibly keep an eye on this thread for the next few days
I would be very much in your debt. Your help has been invaluabe so far in
allowing me to visualise quite a few things.

Thanks very much.

Thanks again !

1)
Sorry for being too vague. With regard to the HEADINGS they separate blocks
of data. But because the column names will be different [data is different]
then I'm not quite sure I could use:
$names{$heading}{$name}++;

So I'm looking at creating separate my () definitions for each HEADING and
just wanted to confirm how to jump out of one HEADING loop and start with
the next.

For example, under HEADING 1 we have these columns:
Name, Sex, Dept, M_Status, Age

and under HEADING 2we have:
Address, Phone#, Mobile#, Salary

So at the beginning of the script I would have
my (%names, %sexes, %depts, %m_statuses, %ages)
my (%addresses, %phones, %mobiles, %salaries)
#then I have my while (<>) and parsing here
#I have my output at the end

Is that a little more clearer?

Yes. Much clearer. There are a couple of different ways you could do
this. One is to use a single loop that reads through the file and uses
a state variable (e.g., $heading) to keep track of where you are in the
parsing process. The other is to have a separate loop for each heading.
Again, six of one, half a dozen of another. It's more a matter of
preference than anything else.

An example of the first approach:

my $heading = 'initial';
my $fin_name = '/usr/local/blah/blah/blah';
open FIN,$fin_name || die "Can't open $fin_name\n";

while (<FIN>) {

# check for a new heading
# I am assuming single word heading names
if ( /HEADING (\S+)/ {

$heading = $1; # set $heading equal to word extracted above

# take appropriate action based on the heading we are under

} elsif ( $heading eq 'NAMES' ) {

( $name, $sex, $dept, $m_status, $age ) =
/(\w+) (\w+) (\w+) (\w+) (\d+)/;

# update counts, append to lists, etc...

} elsif ( $heading eq 'ADDRESSES' ) {

# I am assuming the address field is limited to 30 characters
# here:
( $address,$phone, $mobile, $salary ) =
/(\.{30}) (\S+) (\S+) (\d+)/;

# update counts, append to lists, etc...

}

}


And the second approach:

my $heading = 'initial';
my $fin_name = '/usr/local/blah/blah/blah';
open FIN,$fin_name || die "Can't open $fin_name\n";

# scan for first heading
while ( <FIN> && ! /HEADING NAMES/ );

# parse the names, etc...
while ( <FIN> && ! /HEADING ADDRESSES/ ) {

( $name, $sex, $dept, $m_status, $age ) =
/(\w+) (\w+) (\w+) (\w+) (\d+)/;

# update counts, append to lists, etc...


# parse the addresses, etc...
# for brevity , I am assuming only two headings
while ( <FIN> ) {

( $address,$phone, $mobile, $salary ) =
/(\.{30}) (\S+) (\S+) (\d+)/;

# update counts, append to lists, etc...

}

2)
With my last question regarding the printing of the names of single people,
if we include a print statement in the parsing loop would that give us
something like:
Pete is single.
John is single.
while the parsing is still running?
Yes.


What I'm after is hopefully feeding that output into something else
[@array?] which can then print a list of the names [line by line] at

the

end

Yes. It would be easy to create a list/array of, e.g., single people.
Prior to the loop, declare the array. Within the loop, test each person
for being single. If they are, push them onto the list:

# prior to your parsing loop, declare array @singles:

my @singles;

# within your parsing loop, after parsing out name, status, etc.:

if ( $m_status eq 'Single' ) push @singles,($name);

# after loop, to print the list of singles:

print "Single persons:\n";
foreach $single_person ( @singles ) print " $single_person\n";


Greg

 

[Troll]

I'm getting heaps of the following errors when I run my script:
Use of uninitialized value in hash element at ...

The beginning of my script looks like:
my(%names, %sexes, %depts);
%names = ("name" => "0");
%sexes = ("sex" => "0");
%depts = ("dept" => "0");

$names = '0';
$sexes = '0';
$depts = '0';
$name = '0';
$sex = '0';
$dept = '0';

while (<>)
#and the parsing loop here...


The hash errors relate to only these 3 lines which are part of the parsing
loop:
$names{$name}++;
$sexes{$sex}++;
$depts{$dept}++;


Can you run over the variable declarations/initializations for me as I'm not
sure I'm doing this right?
Thanks.

Now time for some stupid Qs:

Let's say that the data I have is in a file called employees.
How can I call this file so that I can parse it?

1) Can I do:
@HRdata = `cat employees`;
while (<@HRdata>) {


2) With regard to the HEADING sections, the script has to be able to
recognise the different sections by the following rules:
# there's a blank line
before each heading
HEADING 1 # this is the name of the heading -
this is a string with a special character and a blank space as part of it
ColumnA ColumnB ColumnC # these are the column names - these are
strings which also can inlude a blank space if they have 2 or more words
******* # a sort of an underlining
pattern

I guess this is to make sure that one does not include any silly heading
data as part of the arrays created and the parsing only takes place on
'real' data. Can you pls advise? Or do you need more info? I'm more in
favour of creating separate 'if' loops due to my 'newbie' status. I'll get
lost otherwise...

Thanks.

Wow. I don't know how you get the time to respond to my queries in such
detail. It is greatly appreciated.
I just came back from work and it's like 2:30 am so I'll crash out soon and
have a closer read tomorrow [especially of the HEADINGS part].

With the push @array stuff I actually got to this today in my readings. I
saw an example of appending an array onto another array with a push and I
was wondering if we could just substitute a $variable for one of the arrays.
I'm glad you confirmed this.

I was also wondering if doing this at the beginning of the script:

my (%names, %sexes, %depts, %m_statuses, %ages) # declaring things
locally

would be considered bad practice. I thought that one should declare things
as my ( ) if one is using things within a loop so as not to impact anything
external to the loop. But if one uses variables/arrays both within and
outside the loops, should we then still declare stuff as my ( )?
Maybe I'm just confused about my ( )...

Greg, if you could possibly keep an eye on this thread for the next few days
I would be very much in your debt. Your help has been invaluabe so far in
allowing me to visualise quite a few things.

Thanks very much.

Troll wrote:
Thanks again !

1)
Sorry for being too vague. With regard to the HEADINGS they separate blocks
of data. But because the column names will be different [data is different]
then I'm not quite sure I could use:
$names{$heading}{$name}++;

So I'm looking at creating separate my () definitions for each

HEADING
and

just wanted to confirm how to jump out of one HEADING loop and start with
the next.

For example, under HEADING 1 we have these columns:
Name, Sex, Dept, M_Status, Age

and under HEADING 2we have:
Address, Phone#, Mobile#, Salary

So at the beginning of the script I would have
my (%names, %sexes, %depts, %m_statuses, %ages)
my (%addresses, %phones, %mobiles, %salaries)
#then I have my while (<>) and parsing here
#I have my output at the end

Is that a little more clearer?

Yes. Much clearer. There are a couple of different ways you could do
this. One is to use a single loop that reads through the file and uses
a state variable (e.g., $heading) to keep track of where you are in the
parsing process. The other is to have a separate loop for each heading.
Again, six of one, half a dozen of another. It's more a matter of
preference than anything else.

An example of the first approach:

my $heading = 'initial';
my $fin_name = '/usr/local/blah/blah/blah';
open FIN,$fin_name || die "Can't open $fin_name\n";

while (<FIN>) {

# check for a new heading
# I am assuming single word heading names
if ( /HEADING (\S+)/ {

$heading = $1; # set $heading equal to word extracted above

# take appropriate action based on the heading we are under

} elsif ( $heading eq 'NAMES' ) {

( $name, $sex, $dept, $m_status, $age ) =
/(\w+) (\w+) (\w+) (\w+) (\d+)/;

# update counts, append to lists, etc...

} elsif ( $heading eq 'ADDRESSES' ) {

# I am assuming the address field is limited to 30 characters
# here:
( $address,$phone, $mobile, $salary ) =
/(\.{30}) (\S+) (\S+) (\d+)/;

# update counts, append to lists, etc...

}

}


And the second approach:

my $heading = 'initial';
my $fin_name = '/usr/local/blah/blah/blah';
open FIN,$fin_name || die "Can't open $fin_name\n";

# scan for first heading
while ( <FIN> && ! /HEADING NAMES/ );

# parse the names, etc...
while ( <FIN> && ! /HEADING ADDRESSES/ ) {

( $name, $sex, $dept, $m_status, $age ) =
/(\w+) (\w+) (\w+) (\w+) (\d+)/;

# update counts, append to lists, etc...


# parse the addresses, etc...
# for brevity , I am assuming only two headings
while ( <FIN> ) {

( $address,$phone, $mobile, $salary ) =
/(\.{30}) (\S+) (\S+) (\d+)/;

# update counts, append to lists, etc...

}



2)
With my last question regarding the printing of the names of single people,
if we include a print statement in the parsing loop would that give us
something like:
Pete is single.
John is single.
while the parsing is still running?

Yes.


What I'm after is hopefully feeding that output into something else
[@array?] which can then print a list of the names [line by line] at

the

end
of the script, something like:
#this is the output structure
Number of Petes =
Number of Males =
Singles are:
Pete
John
Number of Salespeople =


Does this make sense?


Yes. It would be easy to create a list/array of, e.g., single people.
Prior to the loop, declare the array. Within the loop, test each person
for being single. If they are, push them onto the list:

# prior to your parsing loop, declare array @singles:

my @singles;

# within your parsing loop, after parsing out name, status, etc.:

if ( $m_status eq 'Single' ) push @singles,($name);

# after loop, to print the list of singles:

print "Single persons:\n";
foreach $single_person ( @singles ) print " $single_person\n";


Greg

 

[Ga Mu]

Greg,
I decided to give you a glimpse at the code itself so as to make it clearer.
Just be aware that the variable/array names have changed but the general
idea is the same.
The hash errors refer to the variables in the increment section.

#!/usr/bin/perl -w

open(NET, "netstat|") || die ("Cannot run netstat: $!");

my(%UDP4localaddresses, %UDP4remoteaddresses, %UDP4states);

$UDP4localaddress = '0';
$UDP4remoteaddress = '0';
$UDP4state = '0';

Why are you doing this (above)? This is initializing three variables to
zero. These three variables have nothing to do with the three variables
of the same name in the while loop.

$UDP4localaddresses = '0';
$UDP4remoteaddresses = '0';
$UDP4states = '0';

Why are you doing this (above)? This is initializing three scalars to
zero. These three scalars have the same name, but have nothing else to
do with the hashes of the same name.

$UDP4localaddresses{$UDP4localaddress} = '0';
$UDP4remoteaddresses{$UDP4remoteaddress} = '0';
$UDP4states = ($UDP4state} = '0';

Instances of hash keys are automatically initialized to zero. That is
what makes them perfect for counting occurences of unknown words,
numbers, etc. And even if you had to initialize them, you are
initilizing $UDP4localaddresses{0} to zero.

while (<NET>) {
my($UDP4localaddress, $UDP4remoteaddress, $UDP4state)=
/(\s+) (\s+) (\s+)$/;

#increments start here
$UDP4localaddresses{$UDP4localaddress}++;
$UDP4remoteaddresses{$UDP4remoteaddress}++;
$UDP4states = ($UDP4state}++;

If the increments above are failing, it is probably because your m// is
failing and one or more of the keys (variable inside the {}) are
undefined. Try putting a print statement before the increments and
print each of the variables you are extracting, then play with the
regular expression until you get values for ALL of them.

 

[Ga Mu]

Now time for some stupid Qs:

Let's say that the data I have is in a file called employees.
How can I call this file so that I can parse it?

1) Can I do:
@HRdata = `cat employees`;
while (<@HRdata>) {

The above is considered bad practice, especially if the file is large.
Why read the entire file into memory when you can read, process, and
discard a line at a time..? To open and read a file:

open (FIN, '<employess') || die "blah blah blah...";

while (<FIN>) {


}

2) With regard to the HEADING sections, the script has to be able to
recognise the different sections by the following rules:
# there's a blank line
before each heading
HEADING 1 # this is the name of the heading -
this is a string with a special character and a blank space as part of it
ColumnA ColumnB ColumnC # these are the column names - these are
strings which also can inlude a blank space if they have 2 or more words
******* # a sort of an underlining
pattern

while (<FIN>) {

if ( /^$/ ) {

# this is a blank line, don't do anything

} elsif ( /HEADING (\.+)/ ) {

# this is a heading, with the heading name in $1

} elsif ( (($name, $sex, $status, $age) = /(\s+) (\s+) (\s+) (\d+)/) ==
4 ) {

# this line contains three words and a number, do whatever
# (I'm not really sure if this will work. My Linux box is
# down and I have no way of testing.)

}

I guess this is to make sure that one does not include any silly heading
data as part of the arrays created and the parsing only takes place on
'real' data. Can you pls advise? Or do you need more info? I'm more in
favour of creating separate 'if' loops due to my 'newbie' status. I'll get
lost otherwise...

"if loops"...? How does one make an if loop?

Thanks.

Wow. I don't know how you get the time to respond to my queries in such
detail. It is greatly appreciated.
I just came back from work and it's like 2:30 am so I'll crash out soon
and

have a closer read tomorrow [especially of the HEADINGS part].

With the push @array stuff I actually got to this today in my readings. I
saw an example of appending an array onto another array with a push and I
was wondering if we could just substitute a $variable for one of the
arrays.

I'm glad you confirmed this.

I was also wondering if doing this at the beginning of the script:

my (%names, %sexes, %depts, %m_statuses, %ages) # declaring things
locally

would be considered bad practice. I thought that one should declare things
as my ( ) if one is using things within a loop so as not to impact
anything

external to the loop. But if one uses variables/arrays both within and
outside the loops, should we then still declare stuff as my ( )?
Maybe I'm just confused about my ( )...

Greg, if you could possibly keep an eye on this thread for the next few
days

I would be very much in your debt. Your help has been invaluabe so far in
allowing me to visualise quite a few things.

Thanks very much.

Troll wrote:

Thanks again !

1)
Sorry for being too vague. With regard to the HEADINGS they separate
blocks

of data. But because the column names will be different [data is
different]

then I'm not quite sure I could use:
$names{$heading}{$name}++;

So I'm looking at creating separate my () definitions for each HEADING
and

just wanted to confirm how to jump out of one HEADING loop and start
with

the next.

For example, under HEADING 1 we have these columns:
Name, Sex, Dept, M_Status, Age

and under HEADING 2we have:
Address, Phone#, Mobile#, Salary

So at the beginning of the script I would have
my (%names, %sexes, %depts, %m_statuses, %ages)
my (%addresses, %phones, %mobiles, %salaries)
#then I have my while (<>) and parsing here
#I have my output at the end

Is that a little more clearer?

Yes. Much clearer. There are a couple of different ways you could do
this. One is to use a single loop that reads through the file and uses
a state variable (e.g., $heading) to keep track of where you are in the
parsing process. The other is to have a separate loop for each heading.
Again, six of one, half a dozen of another. It's more a matter of
preference than anything else.

An example of the first approach:

my $heading = 'initial';
my $fin_name = '/usr/local/blah/blah/blah';
open FIN,$fin_name || die "Can't open $fin_name\n";

while (<FIN>) {

# check for a new heading
# I am assuming single word heading names
if ( /HEADING (\S+)/ {

$heading = $1; # set $heading equal to word extracted above

# take appropriate action based on the heading we are under

} elsif ( $heading eq 'NAMES' ) {

( $name, $sex, $dept, $m_status, $age ) =
/(\w+) (\w+) (\w+) (\w+) (\d+)/;

# update counts, append to lists, etc...

} elsif ( $heading eq 'ADDRESSES' ) {

# I am assuming the address field is limited to 30 characters
# here:
( $address,$phone, $mobile, $salary ) =
/(\.{30}) (\S+) (\S+) (\d+)/;

# update counts, append to lists, etc...

}

}


And the second approach:

my $heading = 'initial';
my $fin_name = '/usr/local/blah/blah/blah';
open FIN,$fin_name || die "Can't open $fin_name\n";

# scan for first heading
while ( <FIN> && ! /HEADING NAMES/ );

# parse the names, etc...
while ( <FIN> && ! /HEADING ADDRESSES/ ) {

( $name, $sex, $dept, $m_status, $age ) =
/(\w+) (\w+) (\w+) (\w+) (\d+)/;

# update counts, append to lists, etc...


# parse the addresses, etc...
# for brevity , I am assuming only two headings
while ( <FIN> ) {

( $address,$phone, $mobile, $salary ) =
/(\.{30}) (\S+) (\S+) (\d+)/;

# update counts, append to lists, etc...

}



2)
With my last question regarding the printing of the names of single
people,

if we include a print statement in the parsing loop would that give us
something like:
Pete is single.
John is single.
while the parsing is still running?

Yes.


What I'm after is hopefully feeding that output into something else
[@array?] which can then print a list of the names [line by line] at

the

end

of the script, something like:
#this is the output structure
Number of Petes =
Number of Males =
Singles are:
Pete
John
Number of Salespeople =


Does this make sense?


Yes. It would be easy to create a list/array of, e.g., single people.
Prior to the loop, declare the array. Within the loop, test each person
for being single. If they are, push them onto the list:

# prior to your parsing loop, declare array @singles:

my @singles;

# within your parsing loop, after parsing out name, status, etc.:

if ( $m_status eq 'Single' ) push @singles,($name);

# after loop, to print the list of singles:

print "Single persons:\n";
foreach $single_person ( @singles ) print " $single_person\n";


Greg

 

[Troll]

Thanks again

Will I get these errors:
Use of uninitialized value in print at ./netstat.pl line 16, <NET> line 1.
Use of uninitialized value in print at ./netstat.pl line 17, <NET> line 1.
Use of uninitialized value in print at ./netstat.pl line 18, <NET> line 1.
....etc

if an undefined value is passed, for example, to $UDP4localaddress?
Because if that's the case then all I need to do is to make sure that
whatever I'm passing as part of the m()// is correctly split and defined as
a string, digit, word etc, yes?

 

[John Bokma]

Ga Mu wrote:

while (<FIN>) {

if ( /^$/ ) {

# this is a blank line, don't do anything

next if /^\s*$/; # skip blank lines (or consisting of white space
# only)

} elsif ( /HEADING (\.+)/ ) {

# this is a heading, with the heading name in $1

if (/ .....) {

# this is a heading
next;
}

} elsif ( (($name, $sex, $status, $age) = /(\s+) (\s+) (\s+) (\d+)/) ==

if (......) {

# bla bla
next;
}

next moves on to the next "while step".

 

[Ga Mu]

Thanks again

Will I get these errors:
Use of uninitialized value in print at ./netstat.pl line 16, <NET> line 1.
Use of uninitialized value in print at ./netstat.pl line 17, <NET> line 1.
Use of uninitialized value in print at ./netstat.pl line 18, <NET> line 1.
...etc

if an undefined value is passed, for example, to $UDP4localaddress?
Because if that's the case then all I need to do is to make sure that
whatever I'm passing as part of the m()// is correctly split and defined as
a string, digit, word etc, yes?

Exactly. Experiment with your re in the m// until you get values.

 

[Ga Mu]

Thanks again. No reading files into memory from now on [unless necessary]

The data will actually be read from stdin in the form of
$ netstat | netstat.pl
or
$ netstat.pl < netstat

Will something like this suffice?
#!/usr/bin/perl -w
while (<STDIN>) {

STDIN is the default file handle, so all you need is:

while (<>) {

}

"if loops"...? How does one make an if loop?

What I meant here is that I'll create 4 separate 'if' sections [with their
own elsif branches], one for each HEADING section [there are 4 of them].
So I think I meant 'if' statements...is that better or I am still confusing
my terminology?

Makes more sense...

 

[Troll]

Thanks very much.

I'm having a bit of drama within my parsing loop.

If I'm trying to look for a specific pattern [ie. tcp] then I am able to
find it [by printing a 'found' message]. This message is then printed each
and every time 'tcp' is found [for a total of 6 times on 6 separate lines].
The script then finishes.

But if I'm trying to increment the number of times this pattern was found I
get the dreaded error:
Use of uninitialized value in hash element at ...

Here's the code extract:
while (<>) {
my($Proto)=
/(\s+)*$/;

if (/tcp/) {
print 'found';
$Protos{$Proto}++;


where am I failing ?

 

[Troll]

OK, I had some luck getting the first value incremented but no more.

Version which works:
*****************
if (/tcp/) {
my($Proto)=
/^(\w+)/;
$Protos{$Proto}++;
}
print "TCP = $Protos{'tcp'}\n";

#output section
TCP = 6 # all is correct here


Version which does not work:
**********************
if (/tcp/) {
my($Proto, $RecvQ)=
/^(\w+) (\s+)/;
$Protos{$Proto}++;
$RecvQs{$RecvQ)++;
}
print "TCP = $Protos{'tcp'}\n";
print "RecvQ = $RecvQs{'0'}\n";

#output section
TCP = 6 # all is correct here
Use of uninitialized value in concatenation (.) or string at... # error
time - this refers to the 2nd print statement
RecvQ = # this is blank

I have tried reading the second parameter as a (\s+) and as a (\d+) with no
luck. If you run netstat you will probably see that all items in the RecvQ
column are 0.
What have I done wrong now?

Can a number of whitespaces be represented by:
/^(\w+) (\s+)/; # this is a word followed by some spaces followed by a
string
or is the above only ONE whitespace?

 

[John Bokma]

OK, I had some luck getting the first value incremented but no more.

Version which works:
*****************
if (/tcp/) {
my($Proto)=
/^(\w+)/;
$Protos{$Proto}++;
}
print "TCP = $Protos{'tcp'}\n";

#output section
TCP = 6 # all is correct here


Version which does not work:
**********************
if (/tcp/) {
my($Proto, $RecvQ)=
/^(\w+) (\s+)/;
$Protos{$Proto}++;
$RecvQs{$RecvQ)++;
}
print "TCP = $Protos{'tcp'}\n";
print "RecvQ = $RecvQs{'0'}\n";

#output section
TCP = 6 # all is correct here
Use of uninitialized value in concatenation (.) or string at... # error
time - this refers to the 2nd print statement
RecvQ = # this is blank

I have tried reading the second parameter as a (\s+) and as a (\d+) with no
luck. If you run netstat you will probably see that all items in the RecvQ
column are 0.
What have I done wrong now?

I guess you want (\S+) ie, non-whitespace. If it are always digits you
should use (\d+). If the number of spaces between proto and recvq can be
more than one you should use something like:

(\w+)\s+(\d+)

print the values of $proto and $recvq

Also, you can't be sure there are any recvqs{'0'} so check this
same for protos.

print "TCP = ...." if defined $Protos{'tcp'};
print "RecvQ = ..." if defined $RecvQs{'0'};

Can a number of whitespaces be represented by:
/^(\w+) (\s+)/; # this is a word followed by some spaces followed by a
string

nope. \s+ means one or more whitespaces. Not *string*
and it is a word followed by exactly one space (white space?).
See above.

HTH

 

[Ga Mu]

Here's the code extract:
while (<>) {
my($Proto)=
/(\s+)*$/;

Your m// above is saying find an occurence of one or more spaces, zero
or more times, terminated by an end-of-line.

if (/tcp/) {
print 'found';
$Protos{$Proto}++;

This m// has nothing to do with the value, if any, that was extracted
into $proto. It is looking at the last line read for "tcp".

I'll continue in you next post...

 

[Troll]

Looks like I had some typos there but after correcting them it's still a no
go
/^(\w+) (\s+)/;
was changed to
/^(\w+)(\s+)(\S+)(\s+)(\S+)/;
# looking for word(s), 1 or more spaces, non-space(s), space(s),
non-space(s)


Still get the same output tho:
#output section
TCP = 6 # all is correct here
Use of uninitialized value in concatenation (.) or string at... # error
time - this refers to the 2nd print statement
RecvQ = # this is blank

What am I missing?