I've been webscraping the chat room with as goal to retrieve some statistics. (have a look at this meta on security.SE (What do you guys want to know about what is said in the DMZ?)
After indexing all the messages, it turns out I'm missing a bunch messages. I'm not sure why I'm missing these messages, so I've got two questions:
- Is there an easier way to get the complete dataset rather than scraping the chatroom?
- How are these messages counted? Does an edit count as a new message? Do deleted messages still count?
I'm off by about 15 ~ 20 percent.
My source code is here: https://github.com/cloud101/StackExchangeChatScraper
Background
I'm currently loading all the messages in Elasticsearch which I can then query with Apache Spark. The goal is to get some really dumb, but funny statistics which I will display on a seperate site using React.js (this whole thing started with me wanting to learn react.js and thinking of a fun project).