import plotly.express as px
import os 
import sys
ROOT_DIR = os.path.abspath(os.path.join(os.path.dirname(__file__), ".."))
sys.path.append(ROOT_DIR)
import numpy as np
import pandas as pd
import plotly.express as px
import numpy as np
from config.constants import VENUE_ORDER
import json
from core.paper import _parse_url_field

paper_dump = pd.read_csv('data/urls.csv', sep="\t")
paper_dump['urls_manual'] = paper_dump['urls_manual'].apply(
    lambda x: _parse_url_field(json.loads(x))
)
paper_dump['urls_auto'] = paper_dump['urls_auto'].apply(
    lambda x: _parse_url_field(json.loads(x))
)
paper_dump['url'] = paper_dump.apply(
    lambda row: next((u for u in [*row['urls_manual'], *row['urls_auto']] if "github.com" in u), None),
    axis=1
)
# Calculate total number of URLs per year and venue
total_titles_per_venue = paper_dump.groupby(['year', 'venue']).size().reset_index(name='total_titles')

# Calculate the number of URLs with errors per year and venue
total_url_per_venue = paper_dump[
    paper_dump["url"].notna() & (paper_dump["url"] != "")
].groupby(['year', 'venue']).size().reset_index(name='total_urls')

# Merge the DataFrames to calculate the error rate
merged_df = pd.merge(total_titles_per_venue, total_url_per_venue, on=['year', 'venue'], how='left')
merged_df['repo_rate'] = merged_df['total_urls'] / merged_df['total_titles']

# Plot the error rates using Plotly, with year on x-axis and color by venue
fig = px.bar(
    merged_df,
    x='year',
    y='total_titles',
    color='venue',
    barmode='group',
    title=f'Number of papers per venue',
    labels={'error_rate': 'Success Rate', 'year': 'Year'},
    category_orders={'venue': VENUE_ORDER}
)

fig.update_xaxes(range=[2018, 2025])
fig.show()


# Plot the error rates using Plotly, with year on x-axis and color by venue
fig = px.bar(
    merged_df,
    x='year',
    y='repo_rate',
    color='venue',
    barmode='group',
    title=f'Number of repositories per venue',
    labels={'error_rate': 'Success Rate', 'year': 'Year'},
    category_orders={'venue': VENUE_ORDER}
)
fig.update_xaxes(range=[2018, 2025])
fig.update_yaxes(range=[0, 1])

fig.show()