Removing ONLY UNIQUE blank spaces with sed?

Linux and Unix-Like Emergency UNIX and Linux Support
1

Hi, I have data that looks similar to this:
In which the sentences are written horizontally and the beginning of a sentence is indicated by a 1 in the first column and the number increments until the last item of the sentence. The end of the sentence and the beginning of the next is then indicate by two white lines that separate them.

However, my data has some single white lines within the sentence which is causing problems. Is there a sed or a grep that will remove ONLY those instances of SINGLE white lines, leaving in place all of the double white lines that indicate the sentence change?

Please let me know if you can help.

1       ...     -       ...     :       -       0       ROOT    -       -


1       This    -       this    DT      -       3       NMOD    -       -
2       concerns        -       concern NNS     -       3       NMOD    -      -
3       finger  -       finger  NN      -       0       ROOT    -       -

4       ,       -       ,       ,       -       3       P       -       -
5       or      -       or      CC      -       3       CC      -       -
6       rather  -       rather  RB      -       8       NMOD    -       -
7       the     -       the     DT      -       8       NMOD    -       -
8       lack    -       lack    NN      -       3       COORD   -       -
9       of      -       of      IN      -       8       NMOD    -       -
10      finger  -       finger  NN      -       9       PMOD    -       -

11      .       -       .       SENT    -       0       ROOT    -       -

1       And     -       and     CC      -       3       CC      -       -
2       tattoo  -       tattoo  NN      -       3       DEP     -       -

3       .       -       .       SENT    -       0       ROOT    -       -


1       Here    -       here    RB      -       6       ADV     -       -
2       the     -       the     DT      -       5       NMOD    -       -
3       highlighted     -       highlight       VVN     -       5       NMOD   --
4       nouns   -       noun    NNS     -       6       SBJ     -       -
5       are     -       be      VBP     -       0       ROOT    -       -
6       not     -       not     RB      -       6       VMOD    -       -
7       for     -       for     IN      -       8       ADV     -       -
8      number  -       number  NN      -       9       PMOD    -       -
9      in      -       in      IN      -       10      ADV     -       -
10      any     -       any     DT      -       13      NMOD    -       -
11      way     -       way     NN      -       11      PMOD    -       -

The desired result is as follows:

1       ...     -       ...     :       -       0       ROOT    -       -


1       This    -       this    DT      -       3       NMOD    -       -
2       concerns        -       concern NNS     -       3       NMOD    -      -
3       finger  -       finger  NN      -       0       ROOT    -       -
4       ,       -       ,       ,       -       3       P       -       -
5       or      -       or      CC      -       3       CC      -       -
6       rather  -       rather  RB      -       8       NMOD    -       -
7       the     -       the     DT      -       8       NMOD    -       -
8       lack    -       lack    NN      -       3       COORD   -       -
9       of      -       of      IN      -       8       NMOD    -       -
10      finger  -       finger  NN      -       9       PMOD    -       -
11      .       -       .       SENT    -       0       ROOT    -       -


1       And     -       and     CC      -       3       CC      -       -
2       tattoo  -       tattoo  NN      -       3       DEP     -       -
3       .       -       .       SENT    -       0       ROOT    -       -


1       Here    -       here    RB      -       6       ADV     -       -
2       the     -       the     DT      -       5       NMOD    -       -
3       highlighted     -       highlight       VVN     -       5       NMOD   --
4       nouns   -       noun    NNS     -       6       SBJ     -       -
5       are     -       be      VBP     -       0       ROOT    -       -
6       not     -       not     RB      -       6       VMOD    -       -
7       for     -       for     IN      -       8       ADV     -       -
8      number  -       number  NN      -       9       PMOD    -       -
9      in      -       in      IN      -       10      ADV     -       -
10      any     -       any     DT      -       13      NMOD    -       -
11      way     -       way     NN      -       11      PMOD    -       -
2

Try

awk 'NF==0 {next} $1==1 && NR>1 {print "\n"}1' file
1 Like