Skip to content

Python utilities for scraping OSCN cases.

License

Notifications You must be signed in to change notification settings

openjusticeok/oscn

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

OSCN utilities

A python library for scraping case information from the Oklahoma State Courts Network.

Contents

oscn > Python package source to provide an api for retrieving and parsing case records.

scripts > Python scripts showing use of the oscn package

  • example.py: demonstrates use of the request Case and Caselist
  • retrieve-counts.py: saves a list of all counts for a list of counties and years
  • find-counts.py: saves a list of counts passing a test for a list of counties and years
  • soup_test.py: a stub for testing parsing attempts using BeautifulSoup
  • parse_test.py: a stub for developing using saved examples

OSCN package

oscn

  • counties: Returns a list of counties.
  • courts: Same as counties but more a accurate description.
  • judges: Returns a list of objects formated as {'name': 'Bond, James', 'number': '007'}
  • types: returns a dict of case type codes and descriptons
  • type: function to return case type description. Usage:
    >>> oscn.type("AO")
    'CIVIL ADMINISTRATIVE'
    

oscn.request

  • Case: Returns a single case. Case can be saved as files using Case.save() and retrieved using Case.open().

  • CaseList: Returns an iterator for retrieving cases for a county and year. CaseLists can be filtered using .find(). See scripts/example.py for details

  • Party: Returns information on parties available on OSCN.

  • Docket: Returns docket of cases for specific judges and date

oscn.parse

Parsers accept the html of an OSCN page and return python objects.

Case Page Parsers

  • filed: returns a string of the filing date (e.g. 12/25/2017)
  • closed: returns a string of the date the case was closed. Return None if not closed.
  • counts: returns of list of count dicts found in a case. Keys include 'description' of the count. If available 'violation' and 'disposed' are added.
  • judge: returns a string of the judge's name
  • parties: returns a list of dicts with these keys: id, name, type
  • docket: returns a list of rows in a docket
  • events: returns a list of dicts with these keys: event, party, docket, reporter, date, description. The keys date and description are cleaner versions of the event text. The event key will be deprecated some day so use date and description if you are starting a project.
  • attorneys: returns a list of dicts with these keys: name, address, and representing
  • issues: returns a list of dicts with issue information. Each issues includes a list of dicts for each party

Party Page Parsers

  • name: returns 'Requested Party'
  • alias: returns 'Alias or Alternate Names'
  • profile: returns dict of values in 'Personal Profile'
  • birth_month: returns string of 'Birth Month and Year'
  • addresses: returns a list of dicts for each address

Docket Page Parsers

  • cases: returns a list of case indexes
  • tables: returns the html table for each case in the docket

oscn.find

  • CaseIndexes: returns an iterator of case indexes (e.g. tulsa-CF-2019-12).

Usage

Create a CaseIndexes list using these key word arguments:

  • county: defaults to all,
  • last_name: use this for company or organization names
  • first_name: optional
  • middle_name: optional
  • filed_after: More readable than FiledDateL
  • filed_before: More readable than FiledDateH
  • closed_after: More readable than ClosedDateL
  • closed_before: More readable than ClosedDateH

Notes

  • The % wild card is added to all words in name, first and middle
  • Date arguments use MM/DD/YYY strings.

OSCN search parameters

If you are familar with the OSCN search parameters you can initialize CaseIndexes using these as key word arguments: db, number, lname, fname, mname, DoBMin, DoBMax, partytype, apct, dcct, FiledDate, FiledDateH, ClosedDateL, ClosedDateH, iLC, iLCType, iYear, iNumber, and citation

Using this will override init keyword values such as first or filed_after.

Development Install

  1. Create and activate a Python 3.6 virtual env
  2. git clone git@github.com:codefortulsa/oscn.git
  3. cd oscn
  4. pip install -e .

Usage

Install with pip install oscn

Script example:

import oscn

Request a single case:

oscn.request.Case(county='tulsa', year='2018', number=84)

or use case index notation:

oscn.request.Case('love-CF-2019-25')

To request a list of cases to iterate:

oscn.request.CaseList(county='adair', year='2016')

Run test scripts

  • pytest tests/

or with ipdb:

- `pytest -s tests/`

specify a test:

  • pytest -s tests/test_parse.py -k 'test_events'

Deployment steps

  1. python3 setup.py sdist bdist_wheel
  2. twine upload dist/*

User Agent

In some cases a custom user agent is required in the header of requests. Setting an environmental varialbe called OSCN_USER_AGENT will override the default.

About

Python utilities for scraping OSCN cases.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%