Slack - Direct Message Retrieval

Python script to utilize Slack's API to retrieve your direct message history with a user and output the conversation to a txt file in the following format:

DAY DATE, TIME: User_name - Message text

Example:

Wed. 10-21-2015, 12:55 PM: Corey Shott - Thanks for the great suggestion. Isn't coding fun?!? Wed. 10-21-2015, 12:57 PM: Joe Schmo - You are welcome. Yes! :simple_smile:

This program is configured to keep track of the timestamp of the last message retrieved so if you run this on a regular basis, you can retrieve only new messages and have them appended to the same file.

---

Required Info

• Python 2.7.1
• Slack API Token: Obtain a token from [Slack's Web API Site] (https://api.slack.com/web).
  • Scroll to the bottom of page and select "Issue Token". If you are part of multiple teams be sure the correct team for which you will be using this program is listed.
• Channel ID: Identify the direct message (DM) Channel Id (ex. D1234567890) for which you want to use this program.

####Getting Direct Message (DM) Channel ID

I haven't found an elegant solution to finding the channel ID. Here's how I did it, would appreciate feedback if you know a better way:

1. Sign-in to your Slack Team in a web browser (ex. TEAMNAME.slack.com/home)
2. Select "Team Directory" (or go directly to TEAMNAME.slack.com/team)
3. Find person for whom you want to retrieve DM history
4. On the right side of the row will be a menu button - click it. If there's a DM history, an option for "Open Message Archive" will appear. Select that option.
5. A new page will open. In the web address will be a string of numbers and letters - starting with "D". That is the channel id.

---

Set-Up

Create Config File This should store the information needed for the program to call the API and output correct information to a file. Format the file as follows:

config.ini

[SlackParams]
channel = DXXXXXX (channel id) 
slack_token = YOUR TOKEN HERE 
log_file = FILE NAME OF WHERE YOU WANT TIMESTAMP INFO LOGGED (i.e. log.txt)
output_file = FILE NAME OF WHERE YOU WANT MESSAGE INFO RECORDED (i.e. output.txt)

The only required information in the config file is the channel and the slack_token. The other items can be handled on command line or, if not given in either location, default log.txt and output.txt files will be created in first run of program.

Running Program

In directory for the project:

$ git clone https://github.com/jcshott/slack-retrieve-DM-info.git

If you have set up the config.ini file as outlined above, you can just run the program, specifying config file name on command line as such:

$ python slack_dmInfo_retrieve.py -c config.ini (or whatever name you gave it, as long as its a *.ini file) `

If you did not set a log file name and/or output file name in the config file. you can do so with the following command line arguments:

####Output filename specification

$ python slack_dmInfo_retrieve.py -c config.ini -o output.txt

####Log filename specification

$ python slack_dmInfo_retrieve.py -c config.ini -l log.txt

The log file will capture the last timestamp of the last message downloaded and outputted to file. If you plan to run the file more than once (like 1x/day or week), the program will read the last timestamp logged in the log file and only retrieve messages after that timestamp. Thus, it is in your best interest to keep the log file the same name after creation - either by specifying it in config file, on command line or keeping it the default.

Similarly, the output from retrieval will got to specified output file and will append each run, the new messages - if you keep that file name the same.

Lastly, the timestamp from which you want to retrieve messages after can be specified on the command line as well, though it is suggested to utilize the log file function instead for ongoing runs of the program so as to avoid duplication of message retrieval.

####Timestamp specification example

$ python slack_dmInfo_retrieve.py -c config.ini -t 1445457322.000006

Note: For the optional command line arguments, if you specify via command line, those specs will override specs in your config file.

By default, the API response only returns the first 100 messages, along with a boolean value of "has_more". If that is set to "true", retrieval script will run, with updated starting timestamp, until all messages are retrieved and "has_more" returns False.