PyShp cuts the shapefile

Question

I have a .las file and .shp file with polygons. The shapefile looks like follows:

For each polygon, I calculated statistics based on elevation data vertices from the .las. Now I want to add fields into .shp with statistics. The problem occurs when I just try to add extra fields with the following function:

    def writeDims(self, params):
       r = shapefile.Reader(self.shapefile)
       w = shapefile.Writer("new.shp")
       w.fields = list(r.fields)
       for key in params[1].keys():
           w.field(key, "F")
       for shaperec in r.iterShapeRecords():
           w.record(*shaperec.record)
           w.shape(shaperec.shape)
       w.close()

the result shapefile has only part of the full shapefile however with fields:

how to prevent such file-cropping?

Answer 1

With the caveat that I have not used this library, you might check your code usage.

The documentation says to only use the tail of list if the field list is from the Reader class.

w.fields = r.fields[1:] # skip first deletion field

You can see that the first items is a field that is not a type you want.

fields = sf.fields

assert fields == [("DeletionFlag", "C", 1, 0), ["AREA", "N", 18, 5],
["BKG_KEY", "C", 12, 0], ["POP1990", "N", 9, 0], ["POP90_SQMI", "N", 10, 1],
["HOUSEHOLDS", "N", 9, 0],

Try changing you code to:

def writeDims(self, params):
   r = shapefile.Reader(self.shapefile)
   w = shapefile.Writer("new.shp")
   w.fields = r.fields[1:] # You should not need to recast r.fields to a list.
   for key in params[1].keys():
       w.field(key, "F")
   for shaperec in r.iterShapeRecords():
       w.record(*shaperec.record)
       w.shape(shaperec.shape)
   w.close()

Additionally, your key/value pairs in QGIS suggest there is a an issue with the field types in your final shapefile. Your float fields should not have spaces and then more numbers.

This suggest that numbers are being inserted as strings and not floats. Not sure if this will fix the problem or it might exist else where, DBF database are really simple (you can edit them in a spreadsheet program) and if the library is not doing any type checking it might be performing incorrect inserts.