Shell Programming and Scripting

Hi All,

I have files:

1. abc.sql

'This is a sample file for testing'
This does not have quotations
this also does not have quotations.
and this 'has quotations'.

here I need to list the hard coded strings 'This is a sample file for testing' and
'has quotations'.

So i have written the shell script as below:

#!/bin/sh
x=$(basename $1)
y=$(dirname $1)
echo "filename: $x"
echo "Dir name: $y"
file_ext=${x##*.}
if [ ! -f $x ]; then
printf "File $x is not a valid file"
exit
fi
if [ $file_ext = sql ] || [ $file_ext = pls ] || [ $file_ext = xml ]; then
printf "This file has extension $file_ext\n";
grep -i "'" $x > hard_coded_strings.txt
fi

Here after the script had been executed, the file "hard_coded_strings.txt" has the text,

'This is a sample file for testing'
and this 'has quotations'.

Now i need to get only the data in the quotation marks, and i have used the awk command and i don't understand how do we compare quotation mark(').

awk command is as below:

awk '{a="/'" ; print index ( $0,a ) > "hardcoded.txt"}' hard_coded_strings.txt

I have used all kinds of possibilitis as,

awk '/"'"||*||"'"/ {a=""'"" ; print index ( $0,a ) > "hardcoded.txt"}' hard_coded_strings.txt

Can any one suggest me how do we get the quoted strings.

Thanks,
Kiran.

search this way for single Quote

'\'' { Read as single quote \ single quote single quote }

Try...

Code:

gawk -v r="'[^']*'" 'match($0,r,a){print a[0]}' $x > hard_coded_strings.txt

Code:

awk -F "'" '{ for (f=1; f<=(NF-1)/2; f++) print $(f*2) }' input_text

Input text:

Code:

'This is a sample file for testing'
This does not have quotations
this also does not have quotations.
and this 'has quotations'.
first 'quotation 1' and second 'quotation 2', no more...
complete 'complete quotation' and 'incomplete

Output:

Code:

This is a sample file for testing
has quotations
quotation 1
quotation 2
complete quotation

Hi,

Iam not able to understand the code,

can u plz explain me how does it work

What does f pint to ? field or word.
is the lopping same as c(print getting executed after each and every increment of f ).

Its working, but i need to understand it.
Iam too testing with examples...

Thanks,
Kiran.

Code:

awk -F "'" '{ for (f=1; f<=(NF-1)/2; f++) print $(f*2) }' input_text

-F "'"
Define the quote ' as the Field Separator for Input Records.
for (f=1; f<=(NF-1)/2; f++)
Loop same as in C.
As the field separator is ', the quoted strings are in fields 2, 4, 6, ...
For example, input=first 'quotation 1' and second 'quotation 2', no more...
Field1=first f=1
Field2=quotation 1 f=1, field f*2=2
Field3= and second f=2 =(NF-1)/2
Field4=quotation 2' f=2, field f*2=4
Field5=, no more...
If the number of fields (NF) is even, there is a missing quote.
print $(f*2)
Print the quoted string.


Jean-Pierre.

Hi Jean Pierrie,

Very thanx for ur explanation. This is a great logic.
Please correct me if am wrong anywhere below:

Suppose the string is,

'First Sample' has Quotations 'Comments'

This is totally divided into 5 fields:

field1 = is blank precedding the quotation mark(').
field2 = first Sample
field3 = has Quotations (with spaces on both sides)
field4 = Comments
field5 = is blank after the last quotation.

And if the line is,

First 'Sample has' quotations 'comments'

field1 = First
field2 = Sample has
field3 = quotations (with spaces on both sides)
field4 = comments
field5 = is blank after the last quotation.

always the quotation strings are at even places.

Very thanx for ur logic.

Kiran