Remove a character between 2 known strings

If the quotes are balanced, you will want to remove commas between every other quote, this can be expressed in awk like this:

awk -F'"' -v OFS='' '{ for (i=2; i<=NF; i+=2) gsub(",", "", $i) } 1' infile

Output:

123,ABC DEV 23,345,534.202,NAME

Explanation

The -F" makes awk separate the line at the double-quote signs, which means every other field will be the inter-quote text. The for-loop runs gsub, short for globally substitute, on every other field, replacing comma (",") with nothing (""). The 1 at the end invokes the default code-block: { print $0 }.

This sed command should help you:

sed -e ':1;/<font>[[:space:]]*$/{N;s#<font>[[:space:]]\+</font>#<font></font>#g;b1}' file

The command is looking for <font> tag that is followed by whitespace up to the end of line. Then the next line is pulled into the pattern space; then the replacement of a possibly existing sequence <font>[[:space:]]\+</font> is performed and script restarts from the beginning. If the pattern space does not match the address /<font>[[:space:]]*$/, i.e. some non-space content is present after a <font> tag, then the pattern space is printed out and cleared by the end of sed script and the process restarts.

Edit: Performance measurement.

I filled a file with the following content repeated 10k times:

<font>
dejidewji
</font>
<font>



</font><font>





</font>

totally, 620Kb. The timings of the script above on 1.4Gz A8-4500M are:

real    0m0.361s
user    0m0.356s
sys 0m0.005s

Edit2:

Your last question update is much easier solved by perl and performance is 10 times better, as showed the other answer:

perl -0777 -pe 's|<font>\s+|<font>|g;s|\s+</font>|</font>|g' file

Credits to @spasic