Skip to main content

pure-python

Project description

Open In Colab Binder

	    ___    __                          __         
	   /   |  / /___  _________  _________/ /__  _____
	  / /| | / / __ `/ ___/ __ \/ ___/ __  / _ \/ ___/
	 / ___ |/ / /_/ / /__/ /_/ / /  / /_/ /  __/ /    
	/_/  |_/_/\__,_/\___/\____/_/   \__,_/\___/_/     

		ALACORDER beta 70.5

Getting Started with Alacorder

GitHub | PyPI | Report an issue

Alacorder processes case detail PDFs into data tables suitable for research purposes. Alacorder also generates compressed text archives from the source PDFs to speed future data collection from the same set of cases.

Installation

Alacorder can run on most devices. If your device can run Python 3.7 or later, it can run Alacorder.

  • To install on Windows, open Command Prompt and enter pip install alacorder.
    • To start the interface, enter python -m alacorder or python3 -m alacorder.
  • On Mac, open the Terminal and enter pip3 install alacorder then python3 -m alacorder.
    • To start the interface, enter python3 -m alacorder or python -m alacorder.
  • Install Anaconda Distribution to install Alacorder if the above methods do not work, or if you would like to open and edit this interactive guide on your desktop.
    • After installation, create a virtual environment, open a terminal, and then repeat these instructions. If your copy of Alacorder is corrupted, use pip uninstall alacorder or pip3 uninstall alacorder and then reinstall it. There may be a newer version.

Alacorder should automatically download and install dependencies upon setup, but you can also install the full list of dependencies yourself with pip: pandas, numpy, PyPDF2, openpyxl, xlrd, xlwt, build, setuptools, xarray, jupyter, numexpr, bottleneck.

pip uninstall -y alacorder
pip install alacorder

Using the guided interface

Once you have a Python environment up and running, you can launch the guided interface in two ways:

  1. Import the library from your command line: Depending on your Python configuration, enter python -m alacorder or python3 -m alacorder to launch the command line interface in module mode.

  2. Import the alacorder module in Python: Use the import statement from alacorder import __main__ to start the command line interface.

Alacorder can be used without writing any code, and exports to common formats like Excel (.xls), Stata (.dta), CSV (.csv), and JSON (.json).

  • Alacorder compresses case text into pickle archives (.pkl.xz) to save storage and processing time. If you need to unpack a pickle archive without importing alac, use a .xz compression tool, then read the pickle into Python with the standard library module pickle.

  • Once installed, enter python -m alacorder or python3 -m alacorder to start the interface. If you are using iPython, launch the iPython shell and enter from alacorder import __main__ to launch the guided interface.

from alacorder import __main__

Special Queries with alac

For more advanced queries, the alac module can extract fields and tables from case records with just a few lines of code.

  • Call alac.config(in_path, out_path='', flags='', print_log=True, warn=False) and assign it to a variable to hold your config() object. This tells the imported Alacorder methods where and how it should input and output. If out_path is omitted, alac.write…() and alac.parse() methods will print to console instead of writing to out_path.

  • Call alac.writeArchive(config) to export a full text archive. It's recommended that you create a full text archive and save it as a .pkl.xz file before making tables from your data. Full text archives can be scanned faster than PDF directories and require significantly less storage. Full text archives can be imported to Alacorder the same way as PDF directories.

  • Call alac.writeTables(config) to export detailed case information tables. If export type is .xls or .pkl.xz, the cases, fees, and charges tables will be exported. Otherwise, you can select which table you would like to export.

  • Call alac.writeCharges(config) to export charges table only.

  • Call alac.writeFees(config) to export fee tables only.

import warnings
warnings.filterwarnings('ignore')

from alacorder import alac

pdf_directory = "/Users/crimson/Desktop/Tutwiler/"
archive = "/Users/crimson/Desktop/Tutwiler.pkl.xz"
tables = "/Users/crimson/Desktop/Tutwiler.xls"

# make full text archive from PDF directory 
c = alac.config(pdf_directory, archive)
alac.writeArchive(c)

print("Full text archive complete. Now processing case information into tables at " + tables)

# then scan full text archive for spreadsheet
d = alac.config(archive, tables)
alac.writeTables(d)

Custom Parsing with alac.parse()

If you need to conduct a custom search of case records, Alacorder has the tools you need to extract custom fields from case PDFs without any fuss. Try out alac.parse() to search thousands of cases in just a few minutes.

from alacorder import alac
import pandas as pd
import re

archive = "/Users/crimson/Desktop/Tutwiler.pkl.xz"
tables = "/Users/crimson/Desktop/Tutwiler.xls"

def findName(text):
    name = ""
    if bool(re.search(r'(?a)(VS\.|V\.{1})(.+)(Case)*', text, re.MULTILINE)) == True:
        name = re.search(r'(?a)(VS\.|V\.{1})(.+)(Case)*', text, re.MULTILINE).group(2).replace("Case Number:","").
	strip()
    else:
        if bool(re.search(r'(?:DOB)(.+)(?:Name)', text, re.MULTILINE)) == True:
            name = re.search(r'(?:DOB)(.+)(?:Name)', text, re.MULTILINE).group(1).replace(":","").replace("Case Number:","").strip()
    return name

c = alac.config(archive, tables)

alac.parse(c, findName)
Method Description
getPDFText(path) -> text Returns full text of case
getCaseInfo(text) -> [case_number, name, alias, date_of_birth, race, sex, address, phone] Returns basic case details
getFeeSheet(text, cnum = '') -> [total_amtdue, total_balance, total_d999, feecodes_w_bal, all_fee_codes, table_string, feesheet: pd.DataFrame()] Returns fee sheet and summary as str and pd.DataFrame
getCharges(text, cnum = '') -> [convictions_string, disposition_charges, filing_charges, cerv_eligible_convictions, pardon_to_vote_convictions, permanently_disqualifying_convictions, conviction_count, charge_count, cerv_charge_count, pardontovote_charge_count, permanent_dq_charge_count, cerv_convictions_count, pardontovote_convictions_count, charge_codes, conviction_codes, all_charges_string, charges: pd.DataFrame()] Returns charges table and summary as str, int, and pd.DataFrame
getCaseNumber(text) -> case_number Returns case number
getName(text) -> name Returns name
getFeeTotals(text) -> [total_row, tdue, tpaid, tbal, tdue] Return totals without parsing fee sheet

Working with case data in Python

Out of the box, Alacorder exports to .xls, .csv, .json, .dta, and .pkl.xz. But you can use alac, pandas, and other python libraries to create your own data collection workflows and design custom exports.

The snippet below prints the fee sheets from a directory of case PDFs as it reads them.

from alacorder import alac

c = alac.config("/Users/crimson/Desktop/Tutwiler/","/Users/crimson/Desktop/Tutwiler.xls")

for path in c['contents']:
    text = alac.getPDFText(path)
    cnum = alac.getCaseNumber(text)
    charges_outputs = alac.getCharges(text, cnum)
    if len(charges_outputs[0]) > 1:
        print(charges_outputs[0])

Extending Alacorder with pandas and other tools

Alacorder runs on pandas, a python library you can use to perform calculations, process text data, and make tables and charts. pandas can read from and write to all major data storage formats. It can connect to a wide variety of services to provide for easy export. When Alacorder table data is exported to .pkl.xz, it is stored as a pd.DataFrame and can be imported into other python modules and scripts with pd.read_pickle() like below:

import pandas as pd
contents = pd.read_pickle("/path/to/pkl")

If you would like to visualize data without exporting to Excel or another format, create a jupyter notebook and import a data visualization library like matplotlib to get started. The resources below can help you get started. jupyter is a Python kernel you can use to create interactive notebooks for data analysis and other purposes. It can be installed using pip install jupyter or pip3 install jupyter and launched using jupyter notebook. Your device may already be equipped to view .ipynb notebooks.

Resources

© 2023 Sam Robson

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

alacorder-70.5.9.9.2.tar.gz (19.1 kB view hashes)

Uploaded Source

Built Distribution

alacorder-70.5.9.9.2-py3-none-any.whl (16.1 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page