Current Software Review Stats

# https://github.com/ryantam626/jupyterlab_code_formatter
# TODO: calculate days open in the script that grabs the data!!
import os
import warnings
from datetime import datetime, timezone

import altair as alt
import pandas as pd
import pytz
from pyosmeta import ProcessIssues
from pyosmeta.github_api import GitHubAPI

from pyosmetrics.functions import count_edits_by_quarter

# Suppress all warnings
warnings.filterwarnings("ignore")
# Make tables nicer
pd.set_option("display.max_colwidth", None)
today = datetime.now(timezone.utc)

# Get all current reviews
reviews = pd.read_csv(
    "../../_data/review_submissions.csv",
    parse_dates=["date_opened", "date_closed", "last_comment_date"],
)
reviews["date_opened"] = reviews["date_opened"]
reviews["last_comment_date"] = reviews["last_comment_date"]
reviews = reviews.rename(
    columns={
        "package_name": "Name",
        "date_opened": "Date Opened",
        "date_closed": "Date Closed",
        "issue_num": "Issue",
        "description": "Description",
        "categories": "Categories",
        "last_comment_date": "Last Comment",
        "last_comment_user": "Last User to Comment",
    }
)

open_reviews = reviews[reviews["Date Closed"].isna()]
open_reviews.drop(columns=["Date Closed","Unnamed: 0"], inplace=True)
total_open = len(open_reviews)

open_reviews["Days Open"] = (today - open_reviews["Date Opened"]).dt.days
open_reviews["Date Opened"] = open_reviews["Date Opened"].dt.date
open_reviews["Last Comment"] = open_reviews["Last Comment"].dt.date

seeking_editor = open_reviews[
    open_reviews["labels"].str.contains("0/seeking-editor", na=False)
].copy()
seeking_editor.drop(
    columns=[ "labels", "status"], inplace=True
)
seeking_editor.reset_index(drop=True, inplace=True)

seeking_editor.style.set_properties(
    **{"text-align": "left", "white-space": "normal"}
)

seeking_reviewers = open_reviews[
    open_reviews["labels"].str.contains("2/seeking-reviewers", na=False)
].copy()
seeking_reviewers

presubmissions = pd.read_csv(
    "../../_data/review_presubmissions.csv",
    parse_dates=["date_opened", "date_closed", "last_comment_date"],
)
presubmissions["date_opened"] = presubmissions["date_opened"]
presubmissions["last_comment_date"] = presubmissions["last_comment_date"]
presubmissions = presubmissions.rename(
    columns={
        "package_name": "Name",
        "date_opened": "Date Opened",
        "date_closed": "Date Closed",
        "issue_num": "Issue",
        "description": "Description",
        "categories": "Categories",
        "last_comment_date": "Last Comment",
        "last_comment_user": "Last User to Comment",
    }
)

all_presubmissions = len(presubmissions)

# Get all currently open presubmissions
open_presubmissions = presubmissions[presubmissions["Date Closed"].isna()]
today = datetime.now(timezone.utc)
open_presubmissions["days_open"] = (
    today - open_presubmissions["Date Opened"]
).dt.days
open_presubmissions["Date Opened"] = open_presubmissions["Date Opened"].dt.date
open_presubmissions["Date Opened"] = open_presubmissions["Last Comment"].dt.date
open_presubmissions.reset_index(drop=True, inplace=True)

total_open = len(open_presubmissions)

# Cleanup
open_presubmissions.sort_values(by="Date Opened", ascending=False, inplace=True)
open_presubmissions.drop(columns=["Date Closed", "Unnamed: 0", "editor"], inplace=True)

open_presubmissions

# Static list of all editors, updated 7/13/2024
# Let's pull this list from unique people above who are active. 
# TODO: get this list of current editors dynamically
all_editors = [
    "cmarmo",
    "dhomeier",
    "ocefpaf",
    "NikleDave",
    "SimonMolinsky",
    "Batalex",
    "sneakers-the-rat",
    "tomalrussel",
    "ctb",
    "mjhajharia",
    "hamogu",
    "isabelizimm",
    "yeelauren",
    "banesullivan",
]
all_editors_df = pd.DataFrame({"editor": all_editors})

all_editor_activity = all_editors_df.merge(busy_editors, on="editor", how="outer").fillna(0)

all_editor_activity["count"]= all_editor_activity["count"].astype(int)
all_editor_activity.sort_values(by="count", inplace=True)
all_editor_activity = all_editor_activity.reset_index(drop=True)

# Get counts of available and unavailable editors
available_editors = all_editor_activity[all_editor_activity["count"] == 0]
all_busy_editors = all_editor_activity[all_editor_activity["count"] != 0]

num_available_editors = len(available_editors)
available_editors

# Display editor table
all_editor_activity

# TODO: calculate time that they were in review.

# Get a list of all editors over time that have supported pyOpenSci
ignore_editors = ["TBD"]
# ignore lwasser and xmnlab to bring min date to a more recent date
ignore_editors += ["lwasser", "xmnlab"]
editors = [
    editor
    for editor in pd.unique(reviews.editor)
    if editor not in ignore_editors
]

n_edits = (
    reviews.groupby([reviews["editor"], reviews["Date Opened"].dt.to_period("Q")])
    .size()
    .reset_index(name="n_edits")  # Ensures a DataFrame with column name
)

n_edits_filtered = n_edits[~n_edits["editor"].isin(ignore_editors)]
n_edits_filtered = n_edits_filtered.rename(columns={"Date Opened": "Date", "n_edits": "count"})
# Date should be a standard date time
df = n_edits_filtered.copy()
df[["Year", "Quarter"]] = df["Date"].astype(str).str.extract(r"(\d{4})Q(\d)").astype(int)
df["Date"] = pd.to_datetime(df["Year"].astype(str) + "-" + (3 * (df["Quarter"].astype(int) - 1) + 1).astype(str) + "-01")
df

# This likely should be active edits?
edits = reviews.rename(columns={"Date Opened": "Date"}).copy()

edits = reviews[["editor", "Name", "Date Opened"]]
edits = edits.rename(columns={"Date Opened": "Date", "Name":"package_name"})
edits = edits[edits["editor"] != "TBD"]

charts = [
    alt.Chart(edits.loc[edits.editor == editor])
    .mark_bar(color="purple")
    .encode(
        x=alt.X("yearquarter(Date):T"),
        y=alt.Y("count(package_name)", title="Number of edits per quarter"),
        tooltip=["yearquarter(Date)", "count(package_name)"],
    )
    .properties(
        title=alt.TitleParams(
            text=f"{editor}",
            fontSize=18,
            orient="right",
            angle=0,
            align="right",
        ),
        width=600,
        height=200,
    )
    for editor in editors
]

full_chart = alt.vconcat(*charts).resolve_scale(x="shared", y="shared")
full_chart.show()