How to convert to title case a specific column - linux

Question

I have come up with this code:
cut -d';' -f4 columns.csv | sed 's/.*/\L&/; s/[a-z]*/\u&/g'
which actually does the job for the fourth column, but in the way I have lost the other columns..
I have unsuccessfully tried:
cut -d';' -f4 columns.csv | sed -i 's/.*/\L&/; s/[a-z]*/\u&/g'
So, how could I apply the change to that specific column in the file and keep other columns as they are?
Let say that columns.csv content is:
TEXT;more text;SoMe MoRe TeXt;THE FOURTH COLUMN;something else
Then, expected output should be:
TEXT;more text;SoMe MoRe TeXt;The Fourth Column;something else

Answer 1

GNU sed:
sed -ri 's/;/&\r/3;:1;s/\r([^; ]+\s*)/\L\u\1\r/;t1;s/\r//' columns.csv
update:
sed -i 's/; */&\n/3;:1;s/\n\([^; ]\+ *\)/\L\u\1\n/;t1;s/\n//' columns.csv
Place anchor \r (\n) at the beginning of field 4. We edit the whole word and move the anchor to the beginning of the next one. Jump by label t1 :1 is carried out as long as there are matches for the pattern in the substitution command. Removing the anchor.

Answer 2

Not a short simple awk, but should work:
awk -F";" '{t=split($4,a," ");$4="";for(i=1;i<=t;i++) {a[i]=substr(a[i],1,1) tolower(substr(a[i],2));$4=$4 sprintf("%s ",a[i])}$4=substr($4,1,length($4)-1)}1' OFS=";" file
TEXT;more text;SoMe MoRe TeXt;The Fourth Column;something else
Some shorter version
awk -F";" '{t=split($4,a," ");$4="";for(i=1;i<=t;i++) {a[i]=substr(a[i],1,1) tolower(substr(a[i],2));$4=$4 a[i](t==i?"":" ")}}1' OFS=";" file

Answer 3

With perl:
$ perl -F';' -lane '$F[3] =~ s/[a-z]+/\L\u$&/gi; print join ";", #F' columns.csv
TEXT;more text;SoMe MoRe TeXt;The Fourth Column;something else
-F';' use ; to split the input line
$F[3] =~ s/[a-z]+/\L\u$&/gi change case only for the 4th column
print join ";", #F print the modified fields
Unicode version:
perl -Mopen=locale -Mutf8 -F';' -lane '$F[3]=~s/\p{L}+/\L\u$&/gi;
print join ";", #F'

Answer 4

Using any awk in any shell on every Unix box:
$ cat tst.awk
BEGIN { FS=OFS=";" }
{
title = ""
numWords = split($4,words,/ /)
for (wordNr=1; wordNr<=numWords; wordNr++) {
word = words[wordNr]
word = toupper(substr(word,1,1)) tolower(substr(word,2))
title = (wordNr>1 ? title " " : "") word
}
$4 = title
print
}
$ awk -f tst.awk file
TEXT;more text;SoMe MoRe TeXt;The Fourth Column;something else
True capitalization in a title is much more complicated than that though.

Answer 5

This might work for you (GNU sed):
sed -E 's/[^;]*/\n&\n/4;h;s/\S*/\L\u&/g;H;g;s/\n.*\n(.*)\n.*\n(.*)\n.*/\2\1/' file
Delimit the fourth field by newlines and make a copy.
Uppercase the first character of each word.
Append the amended line to the original.
Using pattern matching, replace the original fourth field by the amended one.