Examples

Imports and initialization

We start with some imports, which aren’t particularly interesting, so they’ve been folded.

Code for importing packages

import pandas as pd
from rcounting import side_threads, thread_navigation as tn

Validating side threads

The rcounting tools have various pieces of functionality to validate the side threads it knows about, both in terms of whether the counts follow the expected format, and whether the thread obeys any special rules that might apply. Here’s an example of what that looks like in a python script.

Code

# Pick a comment early in a chain
comments = pd.DataFrame(tn.fetch_comments("gja0ehe"))
side_thread = side_threads.get_side_thread('slowestest')
print("The thread is valid!" if side_thread.is_valid_thread(comments) else "The thread is invalid")

The thread is valid!

The thread is valid - excellent! There’s also a script that you can run from the command line to validate the most common threads. Try typing rcounting validate -h in a terminal to see how to use it.

Network analysis

The rcounting tools also have functionality to do some network analysis. The following snippet will generate the (comment, replying to, weight) graph for the top 250 counters. The heavy lifting is done by the response graph [[file:../rcounting/analysis.py::def response_graph(df, n=250, username_column=“username”):][response_graph]] function in analysis.py.

Code

from rcounting import analysis, counters
import sqlite3
from pathlib import Path
data_directory = Path("../data")
db = sqlite3.connect(data_directory / "counting.sqlite")
counts = pd.read_sql("select comments.username "
                       "from comments join submissions "
                       "on comments.submission_id = submissions.submission_id "
                       "where comments.position > 0 "
                       "order by submissions.timestamp, comments.position", db)
counts["username"] = counts["username"].apply(counters.apply_alias)
n = 250
graph = analysis.response_graph(counts, n, username_column="username")
graph.sort_values(ascending=False, by="Weight").head()

	Source	Target	Weight
853	Antichess	Countletics	109808
1633	Countletics	Antichess	109793
14934	thephilsblogbar2	GarlicoinAccount	98844
2943	GarlicoinAccount	thephilsblogbar2	98821
12825	nonsensy	Countletics	93082