9

Could someone please explain this code that deletes all leading blank lines at the top of a file:

sed '/./,$!d' file

I understand that it is a regex, matching only the first character, but then don't understand the ,$!d part. Is this what it's being replaced by, or are they options for the match?

Is this even a search command if it does not start with 's/'...?

Code source (from another question)

1 Answer 1

13
sed '/./,$!d'

From the first line which contains a character (blank or not) to the end of the file - negate (which then means from the beginning of the file to the line before the first line which contains a character) - delete.

This deletes leading empty lines, not blank lines. To delete leading blank lines (lines which are empty or contain only whitespace characters) say '/\S/,$!d'.

Read "Sed, an introduction and tutorial" at http://www.grymoire.com/Unix/Sed.html. Then read the reference manual at https://www.gnu.org/software/sed/manual/sed.html.

In short:

  • The general form of a sed command is [selector][negation]command[flags] (square brackets indicate optional parts)

  • The selector, if present, selects the lines on which the command applies

  • If ! appears it negates the selector, that is, makes the command apply to the lines which do not match the selector.

  • If no selector is present the command applies to all lines.

  • A selector can select one line (by number) or a set of lines (by regular expression), or the lines between a start line (by number or regular expression) and an end line (by number or regular expression).

  • In our case the selector is /./,$ which means from the first line found which matches /./ (that is, contains at least one character) to the end of the file ($ is used as a line number and means the last line in the file).

  • It is negated by !, so that the command applies to the lines from the beginning of the file to the line before the first line matching /./.

  • The command d deletes the selected lines.

3
  • Could you please explain each thing individually?What does the comma do? I know the $ means the end of the string. I'm guessing the ! means the negate? How did it get to the point of defining the "beginning of the file to the line before the first line which contains a character" point?
    – Emerson
    Nov 17, 2016 at 10:01
  • Are the ,$!d part of the replace section, or are they options like where the g for a global search would go?
    – Emerson
    Nov 17, 2016 at 10:02
  • I had both of those tabs open, but couldn't find a reference to these options/commands in the table of contents, or through a quick skim, so I didn't know where to look. It looks like I just need a more global understanding of sed first. Thank you for going into detail. I will use this as reference when studying later. :)
    – Emerson
    Nov 17, 2016 at 10:18

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .