Cfrtjryk

<wpt lat="1.345529841" lon="103.7577152"><time>2010-01-01T00:00:00Z</time</wpt> 

<wpt lat="1.345529841" lon="103.7577152"><time>2010-01-01T00:00:00Z</time></wpt> 

<wpt lat="1.3982529841" lon="103.90877152"><time>2010-01-01T00:00:00Z</time></wpt> 

I have a file like having the above lines which needed to be converted into

         1.345529841,103.7577152,2010-01-01 00:00:00

         1.345529841,103.7577152,2010-01-01 00:00:00

         1.3982529841,103.90877152,2010-01-01 00:00:00

You could use sed to strip out the characters you don't want:

sed 's/[^0-9.T:-]+/,/g;s/T/ /;s/^,|,$//g' file

s/[^0-9.T:-]+/,/g is replacing unwanted characters with a comma

s/T/ / is replacing the character T with a space

s/^,|,$//g is removing the first and last comma

GPX is an XML format, so you can't use awk or sed to parse it reliably.

Instead, use something like XMLStarlet:

$ xml sel -t -m '//wpt' 

          -v '@lat' -o ',' -v '@lon' -o ',' 

          -v 'time' -nl data.gpx

1.345529841,103.7577152,2010-01-01T00:00:00Z

1.345529841,103.7577152,2010-01-01T00:00:00Z

1.3982529841,103.90877152,2010-01-01T00:00:00Z

Alternatively:

$ xml sel -t -m '//wpt' -v 'concat(@lat, ",", @lon, ",", time)' -nl data.wpx

this answer is Based on the input given...

awk -F"[<>"]" '{print $3,$5,$9}' OFS=, input.txt | sed "s/[TZ]/ /g"

1.345529841,103.7577152,2010-01-01 00:00:00

1.345529841,103.7577152,2010-01-01 00:00:00

1.3982529841,103.90877152,2010-01-01 00:00:00

awk -F"[<>"]" '{gsub(/T|Z/," ",$9);print $3,$5,$9}' OFS=, input.txt

Please, please - don't use a regular expression based solution, like awk or sed.

XML is contextual, where regular expressions are not - so they can NEVER work properly, they're only at best a bit of a hack.

But XML does have a solution to this problem - it's called xpath, that lets you 'search' in a contextual way.

So to take your example:

#!/usr/bin/perl



use warnings;

use strict;

use XML::Twig;



my $xml = XML::Twig -> new -> parsefile('your_file.xml'); 



foreach my $wpt ( $xml -> get_xpath('//wpt') ) {

   print join ",", $wpt -> att('lat'), 

                   $wpt -> att('lon'),

                   $wpt -> first_child_text('time'), "n";

}

Which gives the desired result, but it will also handle a variety of otherwise perfectly valid and semantically identical forms of your XML.

Like indented:

<xml>

  <wpt lat="1.345529841" lon="103.7577152">

    <time>2010-01-01T00:00:00Z</time>

  </wpt>

  <wpt lat="1.345529841" lon="103.7577152">

    <time>2010-01-01T00:00:00Z</time>

  </wpt>

  <wpt lat="1.3982529841" lon="103.90877152">

    <time>2010-01-01T00:00:00Z</time>

  </wpt>

</xml>

All on a single line:

<xml><wpt lat="1.345529841" lon="103.7577152"><time>2010-01-01T00:00:00Z</time></wpt><wpt lat="1.345529841" lon="103.7577152"><time>2010-01-01T00:00:00Z</time></wpt><wpt lat="1.3982529841" lon="103.90877152"><time>2010-01-01T00:00:00Z</time></wpt></xml>

Another style of indenting:

<xml>

  <wpt

      lat="1.345529841"

      lon="103.7577152">

    <time>2010-01-01T00:00:00Z</time>

  </wpt>

  <wpt

      lat="1.345529841"

      lon="103.7577152">

    <time>2010-01-01T00:00:00Z</time>

  </wpt>

  <wpt

      lat="1.3982529841"

      lon="103.90877152">

    <time>2010-01-01T00:00:00Z</time>

  </wpt>

</xml>

Or even:

<xml

><wpt

lat="1.345529841"

lon="103.7577152"

><time

>2010-01-01T00:00:00Z</time></wpt><wpt

lat="1.345529841"

lon="103.7577152"

><time

>2010-01-01T00:00:00Z</time></wpt><wpt

lat="1.3982529841"

lon="103.90877152"

><time

>2010-01-01T00:00:00Z</time></wpt></xml>

These are all semantically identical, and should be parsed the same way. Hopefully it's fairly clear that a regular expression to do this is a LOT more complicated than just using an XML parser.

For the sake of being concise though:

perl -MXML::Twig -0777 -e 'XML::Twig->new(twig_handlers=>{wpt=>sub{print join ",", $_->att("lat", $_->att("lon"),$_->first_child_text("time"), "n" }})->parse(<>)'

Assuming f.xml is our input (a valid xml):

$ perl -MXML::DT -E 'dt("f.xml",

                         time=>sub{$a=father;

                                   $c =~ s/[TZ]/ /g;

                                   say "$a->{lat},$a->{lon},$c"}

                       )'

• 
  -MXML::DT load XML::DT module (xml down translator)


• 
  dt( file, time => sub{....}) : parse file and every time we see a time execute the correspondent sub


• 
  $a=father : get the attributes from father


• 
  $c : is the current element content

Warning: I am one of the authors of XML::DT (install with cpan XML::DT)