Get starred threads dialogs from Slack

get_slack_threads.ipynb

{
  "cells": [
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### 1. Get your Slack token\n\nAccess https://api.slack.com/custom-integrations/legacy-tokens#legacy_token_generator, get the key for desired workspace and paste it below. "
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2018-07-25T15:24:43.089453Z",
          "start_time": "2018-07-25T15:24:43.085398Z"
        },
        "trusted": true
      },
      "cell_type": "code",
      "source": "TOKEN = \"your-token-here\"\n\noutput_folder = './data'",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### 2. Star some threads\n\n### 3. Run this script"
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2018-07-25T15:00:30.055455Z",
          "start_time": "2018-07-25T15:00:29.824649Z"
        },
        "trusted": true
      },
      "cell_type": "code",
      "source": "from datetime import datetime\nfrom pathlib import Path\nfrom re import findall\nfrom slackclient import SlackClient",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2018-07-25T15:00:30.062534Z",
          "start_time": "2018-07-25T15:00:30.057412Z"
        },
        "trusted": true
      },
      "cell_type": "code",
      "source": "sc = SlackClient(TOKEN)\n\ndata_path = Path(output_folder)",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2018-07-25T15:13:23.485155Z",
          "start_time": "2018-07-25T15:13:23.477133Z"
        },
        "trusted": true
      },
      "cell_type": "code",
      "source": "users = {}\n\ndef add_user(user_id):\n    user = sc.api_call(\"users.profile.get\", user=msg[\"user\"])[\"profile\"]\n    users[msg[\"user\"]] = {\n        \"real_name\": user['real_name'],\n        \"display_name\": f\"(@{user['display_name']})\" if user['display_name'] else \"\"\n    }\n\ndef clean_msg(text):\n    user_handles_in_msg = findall(r\"<@[^>]+>\", text)\n\n    for mention in user_handles_in_msg:\n        user_id = mention.replace(\"<@\", \"\").replace(\">\", \"\")\n        \n        add_user(user_id)\n        handle = user['display_name'] or user['real_name']\n        \n        text = text.replace(mention, handle)\n    \n    return text\n\ndef print_hr(file, num=1):\n    print(file=file)\n    \n    for _ in range(num):\n        print(\"=\" * 80, file=file)\n        \n    print(file=file)",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2018-07-25T15:22:33.508453Z",
          "start_time": "2018-07-25T15:20:21.180989Z"
        },
        "trusted": true
      },
      "cell_type": "code",
      "source": "data_path.mkdir(parents=True, exist_ok=True)\n\nstarreds = sc.api_call(\"stars.list\")[\"items\"]\n    \nfor i_star, starred in enumerate(starreds):\n    mess_id = \"_\".join(starred['message'][\"permalink\"].split(\"/\")[-2:])\n    print(f\"[{i_star+1} / {len(starreds)}] {mess_id}\")\n\n    msg = starred[\"message\"]\n    thread_ts = msg.get(\"thread_ts\")\n\n    with open(data_path / Path(f\"{mess_id}.txt\"), \"w+\") as f:\n        text = clean_msg(msg[\"text\"])\n        \n        print(text, file=f)\n        print_hr(f, 2)\n\n        if thread_ts:\n            msgs = sc.api_call(\"conversations.replies\", channel=starred[\"channel\"], ts=thread_ts)[\"messages\"]\n\n            for i, msg in enumerate(msgs):\n                add_user(msg[\"user\"])\n\n                the_date = datetime.fromtimestamp(starred[\"date_create\"]).isoformat()\n\n                print(f\"{i+1}/{len(msgs)}\", file=f)\n                print(\n                    f\"{the_date} :: [{msg['user']}] {users[msg['user']]['real_name']} {users[msg['user']]['display_name']}\",\n                    file=f\n                )\n\n                text = clean_msg(msg[\"text\"])\n\n                print(text, file=f)\n                print_hr(f)",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3",
      "language": "python"
    },
    "language_info": {
      "name": "python",
      "version": "3.7.0",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "gist": {
      "id": "d16a26e8a0de7a3038a2784b4b69502c",
      "data": {
        "description": "Get starred threads dialogs from Slack",
        "public": false
      }
    },
    "_draft": {
      "nbviewer_url": "https://gist.github.com/d16a26e8a0de7a3038a2784b4b69502c"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 2
}

script.py

# coding: utf-8

# ### 1. Get your Slack token
# 
# Access https://api.slack.com/custom-integrations/legacy-tokens#legacy_token_generator, get the key for desired workspace and paste it below. 

# In[ ]:


TOKEN = "your-token-here"

output_folder = './data'


# ### 2. Star some threads
# 
# ### 3. Run this script

# In[ ]:


from datetime import datetime
from pathlib import Path
from re import findall
from slackclient import SlackClient


# In[ ]:


sc = SlackClient(TOKEN)

data_path = Path(output_folder)


# In[ ]:


users = {}

def add_user(user_id):
    user = sc.api_call("users.profile.get", user=msg["user"])["profile"]
    users[msg["user"]] = {
        "real_name": user['real_name'],
        "display_name": f"(@{user['display_name']})" if user['display_name'] else ""
    }

def clean_msg(text):
    user_handles_in_msg = findall(r"<@[^>]+>", text)

    for mention in user_handles_in_msg:
        user_id = mention.replace("<@", "").replace(">", "")
        
        add_user(user_id)
        handle = user['display_name'] or user['real_name']
        
        text = text.replace(mention, handle)
    
    return text

def print_hr(file, num=1):
    print(file=file)
    
    for _ in range(num):
        print("=" * 80, file=file)
        
    print(file=file)


# In[ ]:


data_path.mkdir(parents=True, exist_ok=True)

starreds = sc.api_call("stars.list")["items"]
    
for i_star, starred in enumerate(starreds):
    mess_id = "_".join(starred['message']["permalink"].split("/")[-2:])
    print(f"[{i_star+1} / {len(starreds)}] {mess_id}")

    msg = starred["message"]
    thread_ts = msg.get("thread_ts")

    with open(data_path / Path(f"{mess_id}.txt"), "w+") as f:
        text = clean_msg(msg["text"])
        
        print(text, file=f)
        print_hr(f, 2)

        if thread_ts:
            msgs = sc.api_call("conversations.replies", channel=starred["channel"], ts=thread_ts)["messages"]

            for i, msg in enumerate(msgs):
                add_user(msg["user"])

                the_date = datetime.fromtimestamp(starred["date_create"]).isoformat()

                print(f"{i+1}/{len(msgs)}", file=f)
                print(
                    f"{the_date} :: [{msg['user']}] {users[msg['user']]['real_name']} {users[msg['user']]['display_name']}",
                    file=f
                )

                text = clean_msg(msg["text"])

                print(text, file=f)
                print_hr(f)

Used Python 3.7 . For other versions, use with caution.