Merging Lines using sed - sed

Question

I have text file that consists of 45999 lines. Each line has a word (unigram). I want to create two-sequential words (bigrams). For example:
apple
pie
red
vine
I want 'apple pie', 'pie red', 'red vine'. I tried with sed 'N;s/\n/ /' but it creates just 'apple pie' and 'red vine'. How can I solve this problem? Thank you..

Answer 1

Could you please try following if you are ok with awk.
awk -v RS="" '
BEGIN{
OFS=","
s1="\047"
}
{
for(i=2;i<=NF;i++){
print s1 $(i-1) s1, s1 $i s1
}
}' Input_file
Output will be as follows.
'apple','pie'
'pie','red'
'red','vine'
2nd solution: since output of OP is not clear so adding this one too.
awk -v RS="" '
BEGIN{
OFS=","
s1="\047"
}
{
for(i=2;i<=NF;i++){
val=(val?val OFS:"")s1 $(i-1) s1 OFS s1 $i s1
}
}
END{
print val
}' Input_file
Output will be as follows.
'apple','pie','pie','red','red','vine'

Answer 2

This might work for you (GNU sed):
sed -nE 'N;s/\n(.*)/ \1&/;P;D' file
Append the next line to the current line, then replace the newline by a space and append the second line again. Print/delete the first line and repeat.
N.B. This does not print the last line as it is not a pair, if the last line is needed use:
sed -E 'N;s/\n(.*)/ \1&/;P;D' file
If the output is to be printed as a single line with each pair surrounded by single quotes and separated by a comma, use:
sed -E ':a;$!N;s/(\S+)\n(.*)/'\''\1 \2'\'', \2/;ta;s/ (\S+)$/ '\''\1'\''/' file
Or:
sed -E ':a;$!N;s/(\S+)\n(.*)/'\''\1 \2'\'', \2/;ta;s/, \S+$/' file