Skip to content

Latest commit

 

History

History
67 lines (48 loc) · 3.57 KB

README.md

File metadata and controls

67 lines (48 loc) · 3.57 KB
Raw

Python Scripts to Parse Email Files

This repository contains different Python scripts to parse email files. The scripts output single pdf files for each email file and attachment as well as a summary csv file with email details and keywords.

So far this repository includes scripts to parse:

These scripts were created by Juan Francisco Saldarriaga at the Brown Institute for Media Innovation

Installation instructions

The scripts require Python 3.x and the following libraries:

  • Pandas
  • Numpy
  • NLTK
  • Rake_NLTK
  • Beautiful Soup
  • pdfkit

When working on MacOS we suggest you install the necessary packages through Homebrew:

  1. Install Homebrew: /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install.sh)"
  2. Install Python 3.x: brew install python
  3. Install wkhtmltopdf (pre-requisite for pdfkit: brew install Caskroom/cask/wkhtmltopdf
  4. Install Pandas, Numpy, NLTK, Rake_NLTK, Beautiful Soup, pdfkit:
pip3 install pandas
pip3 install numpy
pip3 install nltk
pip3 install rake-nltk
pip3 install bs4
pip3 install pdfkit
python3 -m spacy download en_core_web_sm

Running the scripts

To run the scripts download the script files and do: python3 <script name>.

When prompted for the input and output paths, please type (or copy and paste) the full absolute path. Something like /Users/juanfrans/Google Drive/08_Brown/00_TestingGroundMagicGrants/01_DocumentingCovid19/01_ParsingEmails/01_ParsingEML/input.

Note that the input files must be unzipped.

The scripts will produce an output pdf file with the same file name as the original email file. In addition, it will also output the attachments with the name of the original email file pre-appended to their name.

The script will also produce a file called summaryFile.csv which contains the following fields:

  • fileName: the name of the original email file
  • from: the sender's email address
  • to: the receiver's email address
  • cc: any cc'd email address
  • subject: the email subject
  • attachments: the names of any attachment files in the email
  • attachmentTypes: the types of attachments in the email
  • keywords_10: the 10 most important keywords in the email, according to the rake-nltk library
  • persons: a list of persons (and the number of times) mentioned in the email, based on SpaCy named entity recognition model
  • orgs: a list of organizations (and the number of times) mentioned in the email, based on SpaCy named entity recognition model
  • nat_rel_polt: a list of nationalities, religious or political groups (and the number of times) mentioned in the email, based on SpaCy named entity recognition model
  • countries_cities_states: a list of countries, cities or states (and the number of times) mentioned in the email, based on SpaCy named entity recognition model
  • laws: a list of laws (and the number of times) mentioned in the email, based on SpaCy named entity recognition model

Parsing .pst files

To parse .pst files, open Microsoft-Outlook and import them by selecting File / Import / Outlook for Windows archive file (.pst). Once the messages are imported, select all the messages and drag them to a finder window on the folder where you wish to save the files. This should convert them to .eml files.