Evaluates the HTML documents contained on the WestLaw website where prior permission has been granted to perform offline evaluation of the content.
These scripts run best with python3. Install python3 and pip3. Create a virtual environment and activate it. Install package prerequisites using:
python3 -V
python3 -m venv ./venv
source ./venv/bin/activate
pip3 install -r requirements.txt
These scripts use Selenium to access a Firefox browser and the gecko driver to query and navigate the WestLaw site. First ensure Firefox is available on your machine. The code and Selenium tools are largely browser agnostic but not currently configurable. Download Firefox, ensure it runs, then download the gecko driver for Selenium and add it to your system PATH.
mkdir -p ~/src/driver/gecko && cd ~/src/driver/gecko
wget https://github.com/mozilla/geckodriver/releases/download/v0.20.1/geckodriver-v0.20.1-macos.tar.gz
tar -xvf geckodriver-v0.20.1-macos.tar.gz
export PATH=$PATH:~/src/driver/gecko
which geckodriverLaunch a new browser and start parsing. You will be prompted for your WestLaw password.
python3 datachase.py --user myuser --infile my_cases.csv --outfile out.csvYou can describe cases you're interested in using an input file (--infile) in the following format:
2009 WL 82715
2011 WL 6157466
285 Fed. Appx. 326 (8th Cir. 2008)
"286 Fed. Appx. 383 (9th Cir. July 15, 2008)"
290 Fed. Appx. 203 (10th Cir. 2008)Or you can describe the cases you're interested in using any valid WestLaw query (--query)
adv: TI("big waiver")The cases are parsed and their summaries are written to a new CSV (--outfile)
This modified script pulls some data from case docket pages and downloads the HTML versions of party briefs. Run it like so:
python3 briefchase.py --user myuser --infile my_cases.csv --informat single OR fjcThe script takes two formats. The single format:
119 F.3d 772
2009 WL 82715
2011 WL 6157466
285 Fed. Appx. 326 (8th Cir. 2008)
286 Fed. Appx. 383 (9th Cir. July 15, 2008)
290 Fed. Appx. 203 (10th Cir. 2008)Or, a more specialized, FJC format, derived from a slimmed down version of the Federal Judicial Center's courts of appeals data:
| Circuit | Docket | Appellant Name | Appellee Name | Date of Decision |
|---|---|---|---|---|
| 9 | 03-99002 | Ryan | McMurtrey | 08/21/2008 |
| 9 | 03-99003 | Woodford | Pinholster | 05/02/2008 |
| 9 | 03-99009 | McMurtrey | Ryan | 08/21/2008 |
The dockets are parsed and their summaries are written to a CSV while the party briefs are downloaded as HTML files to directories named after either the docket or the case citation. All of this is saved in the output directory.
Selenium can be run in interactive mode, so you'll actually watch the browser navigating. Combined with a Python IDE with breakpoints and the browser's developer tools and view source, this is generally all you need. Firefox writes a log which can b useful as well.
These instructions assume a linux-based OS. Windows users now have Linux too.
