Saturday, 12 January 2013

Converting date formats in a flat-file using Regular expressions

I use two free text editors for preparing flat-files (.csv, etc.) prior to importing into MySQL.

I use Notepad++ in M$ Windows and Kate in Linux.  Both can do Regular expression Search & Replace upon text.

I find regular expressions very useful for preparing date fields prior to import into a database table.  So, converting the UK date format
dd/mm/yyyy (12/01/2013) into a MySQL DATE, DATETIME or, TIMESTAMP datatype (2013-01-12) is easy.

Search:
([0-9]+)/+([0-9]+)/+([0-9]+)


Replace (converts UK date format):
\3-\2-\1

Replace (converts US date format):
\3-\1-\2


So the above UK example searches for dates in the UK date format dd/mm/yyyy and converts them to yyyy-mm-dd.

Before:
Click image to enlarge
After:
Click image to enlarge
Datatypes:
Unlike M$ Excel, these text-editors can cope with large files and do this relatively quickly.

Nice tutorials on Regular expressions at http://www.regular-expressions.info/



No comments: