ArcHacks (WIP)

Extensions and wrappers for ESRI's arcpy Python package (not included).

Purpose

This work-in-progress suite of wrappers and utilities was primarily created for working with ESRI ArcMap/Catalog on a virtual machine (or any painfully slow computer).
Further, this package is engineered to prevent lag when using the Python Window interactively by skipping ESRI's built-in parameter validation.
These tools can be used in ArcMap or in stand-alone scripts.

The highlight of this package is in-memory processing using object-oriented data objects.
This allows us, for instance, to wrap JOIN and SELECT operations as a methods of the in-memory data like so:

import archacks
# Instantiate the memory workspace
mem = archacks.MemoryWorkspace()
# Add the data; this method renames the data to "mem_<data>" by default  
mem.add_layer("C:/workspace/parcels.shp")  
# Get the data as a MemoryLayer (an object-oriented data object)  
parcels = mem.get_memorylayer("mem_parcels")  
# Add a table to memory
mem.add_table("C:/workspace/OwnerInfo.csv")
# Join parcel layer to OwnerInfo table  
parcels.join("mem_OwnerInfo", "ParcelID", "StateGeo")  
# Select by intersection with city_limits data object  
# Assume we've done the same sequence of steps for a city limits layer
parcels.selection.Intersect(city_limits)  
# Make a selection using a query  
parcels.selection.where('"Subdivision" LIKE \'APPLE%\'")  

Convenient, right?

Then there's the problem of reordering and editing the schema of the attribute table -- which ESRI calls a Field Mapping for some reason... ArcHacks makes this easy by wrapping the nonsensical FieldMap and FieldMapping objects (which are really just strings). Let's take our joined parcels-owner data from above and limit the fields to only those we want in our output data:

# Get the field mapping/schema  
schema = archacks.EZFieldMap(parcels)  
# Set the new order of fields by index; by name is planned
new_order = [0, 14, 15, 16, 17, 18, 20, 21, 22, 23]  
schema.reorder(new_order, drop=True)  
# Export the data using the new schema
schema.export("parcel_owners.shp", "C:/workspace")  

Re-mapping the fields by name is planned, but not yet available. Get the current order of the fields using the .current_order property. The export method supports selections and only those selected features are exported with the new schema.

Installation

Unzip this repository and place in your Python's site-packages folder.
e.g.
C:\Python27\ArcGIS10.3\Lib\site-packages\archacks

Brief Overview of Current Features

• Integration with Pandas
• Makes use of the in_memory workspace and treats it as a Python object
• Get field names by regular expression (available, but WIP)
• Sane field-mapping handlers (see above)
• Data as "MemoryLayer" objects
• A Service object (available, but WIP)

Brief List of Planned Features

• Remap schema/field mapping by field name rather than just index
• Use osgeo as faster alternative to common arcpy bottlenecks
• MXD and PDF metadata handler (and append MXD metadata to PDF on export)
• Some kind of testing suite
• Documentation