Wsrtjtyk

So I have two files file1 and file2:

file1:

my name is xyz.

my name is abc.

I am a doctor.

I am an engineer.

I like dogs.

I like cats.

I want to replace some of these sentences by shorter sentences. So I have created another file named file2.csv

file2.csv:

"my name is xyz.","name xyz"

"my name is abc.","name abc"

"I am a doctor.","doctor"

"I like dogs.","dogs"

I have used sed so far and if input all these lines individually in the sed command they work perfectly however the the contents of file1 and file2 may change according to my needs and i want a solution that doesn't need changing the script or the code.
Something like creating a 2 dimensional array and and then checking if the value in the first column of file2 exists in file1 and then replacing it with the corresponding entry in the second column of file2.csv

So after I run the shell script file 1 should look like:

name xyz.

name abc.

doctor.

I am an engineer.

dogs.

I like cats.

Note that the contents in file1 and file 2 can change or new entries can be added and hence using something like

sed -i 's/I like dogs/dogs/' file1.csv

is not feasible.

Using awk

awk -F'"(,")?' '

  NR==FNR { r[$2] = $3; next }

  { for (n in r) gsub(n, r[n]) } 1' file2.csv file1

• 
  -F'"(,")?' is the field separator, matches a " or ",", so that we don't need to remove double quotes from fields,


• 
  NR==FNR { r[$2] = $3; next } populates an array with content of file2.csv using the full sentence as key and replacement string as value,


• 
  { for (n in r) gsub(n, r[n]) } 1 searches for each full sentence in each input record and replaces it with the replacement string.

With bash and sed:

sed -f <(sed 's|","|/|; s|"|/|g; s|^|s|' file2.csv) file1

Output:

name xyz

name abc

doctor

I am an engineer.

dogs

I like cats.

The dot may be a problem because it is a special character in regex.

A concise ruby script:

ruby -rcsv -e '

    sentences = CSV.read(ARGV.shift).to_h

    File.foreach(ARGV.shift, chomp: true) {|line| puts sentences[line] || line}

' file2.csv file1

Using Perl One liner.

$ cat file1

my name is xyz.

my name is abc.

I am a doctor.

I am an engineer.

I like dogs.

I like cats.



$ cat file2.csv

"my name is xyz.","name xyz"

"my name is abc.","name abc"

"I am a doctor.","doctor"

"I like dogs.","dogs"



$ perl -ne ' BEGIN {%kvp=map{chomp;s/"//g;split "," } qx(cat file2.csv)} { chomp;print $kvp{$_}?"$kvp{$_}.n":"$_n"; } ' file1

name xyz.

name abc.

doctor.

I am an engineer.

dogs.

I like cats.



$