Re: Finding Errors in .csv Input Data

Rich Shepard <rshepard(at)appl-ecosys(dot)com> writes:
> I'm sure many of you have solved this problem in the past and can offer
> solutions that will work for me. The context is a 73-column postgres table
> of data that was originally in an Access .mdb file. A colleague loaded the
> file into Access and wrote a .csv file for me to use since we have nothing
> Microsoft here. There are 110,752 rows in the file/table. After a lot of
> cleaning with emacs and sed, the copy command accepted all but 80 rows of
> data. Now I need to figure out why postgres reports them as having too many
> columns.

Did you try pgloader yet?

http://pgloader.projects.postgresql.org/

http://pgfoundry.org/projects/pgloader/
https://github.com/dimitri/pgloader
http://packages.debian.org/sid/pgloader

Regards,
--
Dimitri Fontaine
http://2ndQuadrant.fr PostgreSQL : Expertise, Formation et Support