#!/home/florent/.conda/envs/sixp/bin/python
#/usr/bin/env python
# file csv.py
# author Florent Guiotte <florent.guiotte@irisa.fr>
# version 0.0
# date 27 sept. 2022
"""Abstract

doc.
"""

import sys
from timewreport.parser import TimeWarriorParser
import pandas as pd
import json
from pathlib import Path

FILTER_FILE = '~/.timewarrior/tag_filter.json'
# ["tag1", "tag2"]


def filter_tag(df, tag):
    select = [tag in t for t in df.tags]
    return df[select]


def get_tag_summary(df, tags):
    summary_data = [(tag, filter_tag(df, tag).duration.sum()) for tag in tags]
    summary = pd.DataFrame(summary_data, columns=['tag', 'time'])
    duration_filter = [not not t for t in df.tags]
    summary['ratio'] = summary.time / df[duration_filter].duration.sum()

    return summary


def get_top_tag_summary(tag_summary, tags=None):
    tags_top = tags if tags else set([t.split()[0] for t in tag_summary.tag])

    top_data = []
    for tag in tags_top:
        top_df = summary[[tag == t.split()[0] for t in summary.tag]].sum()
        top_df.tag = tag
        top_data += [top_df]

    return pd.DataFrame(top_data)


def get_timew_df():
    parser = TimeWarriorParser(sys.stdin)

    data = [(i.get_start(), i.get_end(), i.get_duration(), i.get_tags()) for i in parser.get_intervals()]
    cols = ('start', 'end', 'duration', 'tags')
    df = pd.DataFrame(data, columns=cols)

    df.start = pd.to_datetime(df.start)
    df.end = pd.to_datetime(df.end)

    return df


if __name__ == '__main__':
    print(sys.stdin)
    df = get_timew_df()

    tags = set([t for tl in df.tags for t in tl])

    summary = get_tag_summary(df, tags)
    top_summary = get_top_tag_summary(summary)

    with Path(FILTER_FILE).expanduser().open() as ff:
        tag_filter = json.load(ff)

    filtered_summary = get_top_tag_summary(summary, tag_filter)


    for k, s in (('Detailed summary', summary), 
                 ('Top level summary', top_summary),
                 ('Filtered summary', filtered_summary)):
        print()
        print(f'--- {k} ---')
        print(s.sort_values(by=['ratio']).to_string(formatters={'ratio': '{:,.2%}'.format}))