I have a PostGIS database created from an old AutoCAD system. Because of this, every column is type varchar(254), even though actual data lengths are small. I need to export tables as shapefiles, and would like to minimize the extra space created in the shapefile due to the large column fields, in order to reduce the overall size of the .dbf file. Is there any way to reduce the size of fields when utilizing ogr2ogr to convert PostgreSQL data to Esri shapefiles?

  • dBase has a 4000 byte record limit, which you'd overwhelm with 16 254-width string fields.
    – Vince
    Jun 21 at 17:48
  • consider checking the length of the longest string in each column with a query like select max(LENGTH(my_column)) from my_table ; Jun 21 at 18:27

1 Answer 1


You can shrink the shapefile that ogr2ogr has created. Read https://gdal.org/drivers/vector/shapefile.html#field-sizes

It is also possible to force a resize of the fields to the optimal width by issuing a SQL ‘RESIZE ’ via the datasource ExecuteSQL() method. This is convenient in situations where the default column width (80 characters for a string field) is bigger than necessary.

Usage example with ogrinfo

ogrinfo -sql "RESIZE test" test.shp

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.