Re: truncate data before importing

I searched and found they are for whitespace characters. But they deleted sentence coordinates from text for me. Still no idea how.

On Wed, Nov 18, 2015 at 7:18 PM, Felipe Santos <felipepts@gmail.com> wrote:

2015-11-18 16:08 GMT-02:00 Shmagi Kavtaradze <kavtaradze.s@gmail.com>:

I was not able to find any details about "\S+\s+ ", can u explain them? Also thanks a lot, this worked perfectly!

On Wed, Nov 18, 2015 at 4:36 PM, Hans Ginzel <hans@matfyz.cz> wrote:

On Wed, Nov 18, 2015 at 01:49:35PM +0100, Shmagi Kavtaradze wrote:

  I am importing sentences from txt file. They look like:
  "0,170     A       recent  statistical     analysis        by      David
  Barton  graphically     illustrates     how     America         has
  plummeted       from    righteous       living  ,       prosperity
  and     success         in      the     last    quarter         century
       .
  Each Sentence starts with coordinates and each word is delimited with tab.
  I want to import data to tables without coordinates, just text and if
  possible to convert tab delimited space with just 'space', not to have
  such a gap between words. Any solutions how to do it? maybe with shell
  script?

You can use the 'PROGRAM' in COPY syntax
http://www.postgresql.org/docs/current/static/sql-copy.html

-- DROP TABLE IF EXISTS Sentence;
CREATE TABLE IF NOT EXISTS Sentence (s text);
COPY Sentence
FROM PROGRAM 'sed -re ''s/\t/ /g; s/^\S+\s+//'' file.txt'
WITH (FORMAT text, NULL '');

Take care of escape sequences – backslashes in the file.

If the file is on the client side see the \copy command of psql client instaed.
http://www.postgresql.org/docs/9.4/static/app-psql.html#APP-PSQL-META-COMMANDS-COPY

https://www.gnu.org/software/sed/manual/sed.html

H.

These ("S+" and "s+") are regular expressions (regex).

They are being used as part of the SED tool, which is available from your Linux bash.