There are three scripts here to allow for nice error handling in the Facebook APIs. After about 7000+ messages you will get throttled. So breaking this up into the scripts was the bets idea.

This creates a json file filled with all the messages it can find by paginating backward form the start of the conversation. To use this do the following:

Go here: https://developers.facebook.com/tools/explorer/?method=GET&path=me%2Finbox&version=v2.0 Click the "Get User Access Token" option under the "Get Access Token" dropdown Request "read_malibox" permissions under "Extended Permissions" Copy the access token from the input and put it in the access_token.dat file On the same site, grab the conversation id of the conversation you are interested in

Run the script:

python fetchMessages.py -m -c

This script will likely say something about "being throttled" and output a restart.dat file Repeat above process to get a new access token Edit the restart.dat file by removing the "access_token" query parameter Add "&access_token=" to the end of the file

Run the script:

python fetchMessages.py -m -r restart.dat

Repeat until all messages are downloaded.

Run the script: python createMessagesDB.py -d .db -m

Run the script: python analyzeMessages.py -d .db

Profit!