Hello, Codeforces!
Today i gathered some statistics about comments, and I'm going to present them to you.
Midway through writing this post, I discovered that there is a website called codeforces.comment.statistic. The author, Igorjan94, has done much more than i did and you can check out theirs blog, its absolutely epic. They also created a type-script SDK for CF, which is even more epic.
The data pool consisted only of comments from all the blogs on the main page. That means that only a small portion of all blogs were used. I didn't find a good method to get all blogs from all users.
Data pool
- Blogs amount: $$$2059$$$
- Comments amount: $$$297853$$$
- *Root comments: $$$117293$$$
*Root comments are comments that are not a reply to another comment
The most up-voted comments
The most down-voted comments
Reply depth
First comments
If anyone needs the data pool, DM me; I will give it for free. It includes the reply-depth of the comment on top of the API's default comment JSON structure.
Have a nice day, and feel free to down-vote as this post is not really useful.
why not giving honorary mention to this epic comment??
Sadly, this comment was not in the data pool, as the blog post it is on doesn't appear on the main page of Codeforces.
Gotta say tho, nice score!
Also, there exists a comment on CF that has -1055 rating. And there is a deleted comment with -1280 rating. The information is taken from codeforces.comment.statistic
I am profoundly grateful for your efforts in crafting this — it is indeed a privilege of the highest order to find my name included in this distinguished list.
I love to see someone from top 5 standings!
I did all of data gathering in matter of 2 hours and then i spent a lot of time writing the post itself.
bruh, that rotavirus comment is definitely toxic
I don't really see how it is toxic. It is useless and quite manipulate though
Hi, creator of mentioned site is here.
My DB contains:
First comments: Here
You can scrap all comments/blogs from codeforces API just by simple oneliner:
for i in range(127969): get blog with id i from API
If you have some ideas to visualize data in more convenient way or play with data interactively or anything else feel free to DM me
very nice work man
First of all, epic.
I have read all of your blogs, and it seams that you have made much more than I have. You definitely deserve more recognition.
I thought about this way of scraping every blog, but i wasn't sure if it was okay to send 100k requests to the API. I guess it is fine as long as it is doing 1 req/2 seconds. I have some ideas, i will DM you later.
Opened CF after a long time just to see a comment of mine being one of the most upvoted :)
Please give me some advice Sir. Want to be a commenter like you.
drake high quality pictures