PyShp Version 1.1.7 Release

PyShp 1.1.7 is out after several months of virtually no updates.  This release fixes a bunch of minor issues

plus a couple of important features.  You can get it through setuptools or source from the CheeseShop: https://pypi.python.org/pypi/pyshp/1.1.7.  The Google Code page is here:https://code.google.com/p/pyshp/

And as usual there are no dependencies other than Python itself.  Updates include:

• Added Python geo_interface convention to export shapefiles as GeoJSON.

• Used is_string() method to detect file names passed as unicode strings (failed on unicode strings before).

• Added Reader.iterShapes() method to iterate through geometry records for parsing large files efficiently.

• Added Reader.iterRecords() method to iterate through dbf records efficiently in large files.

• Modified shape() method to use iterShapes() if shx file is not available as well as record() method.

• Fixed bug which prevents writing the number 0 to dbf fields.

• Updated shape() method to calculate and seek the start of the next record. The shapefile spec does not require the content of a geometry record to be as long as the content length defined in the header. The result is you can delete features without modifying the record header allowing for empty space in records.

• Added enforcement of closed polygons in the Writer.poly() method.

• Added unique file name generator to use if no file names are passed to a writer instance when saving (ex. w.save()). The unique file name is returned as a string.

• Updated "bbox" property documentation to match Esri specification.

The __geo_interface__ update required a polygon area calculator.  This method is undocumented but you can feed a list of points representing a polygon to shapefile.signed_area(coords) and get an area calculation back. If the area is a positive number the points are clockwise (outer ring).  If the area is negative then the points are in counter-clockwise order (i.e. an inner polygon ring).