#!/usr/bin/env python3
"""Queries wiki.openstreetmap.org for proposals and writes a JSON list of them to the given file."""
import argparse
import html
import json
import sys
import logging
from collections.abc import Container

import logformat
import pywikiapi
import mwparserfromhell
import requests

OSMWIKI_ENDPOINT = 'https://wiki.openstreetmap.org/w/api.php'

# https://wiki.openstreetmap.org/w/index.php?title=Template:Proposal_page&action=edit

logfmt_handler = logging.StreamHandler()
logfmt_handler.setFormatter(logformat.LogfmtFormatter())
logging.basicConfig(handlers=[logfmt_handler], level=logging.INFO)
logger = logformat.get_logger()


@logger.log_uncaught
def run():
    arg_parser = argparse.ArgumentParser(description=__doc__)
    arg_parser.add_argument("out_file")
    args = arg_parser.parse_args()

    res = requests.get(
        OSMWIKI_ENDPOINT,
        params=dict(
            action='expandtemplates',
            prop='wikitext',
            format='json',
            text='{{#invoke:languages/table|json}}',
        ),
    )
    if not res.ok:
        logger.error("expandtemplates request failed", status=res.status_code)
        sys.exit(1)

    data = res.json()
    langs: dict[str, dict] = json.loads(data['expandtemplates']['wikitext'])

    osmwiki = pywikiapi.Site(OSMWIKI_ENDPOINT)

    proposals = []
    # TODO: catch exception raised if HTTP request fails
    for page in osmwiki.query_pages(
        generator='embeddedin',
        geititle='Template:Proposal page',
        geilimit='max',
        prop='revisions',
        rvprop='content',
        rvslots='main',
    ):
        proposal = parse_proposal(
            page_title=page['title'],
            text=page['revisions'][0]['slots']['main']['content'],
            langs=langs,
        )
        if proposal:
            proposals.append(proposal)

    proposals.sort(key=sort_key, reverse=True)

    with open(args.out_file, 'w') as f:
        json.dump([{k: v for k, v in p.items() if v is not None} for p in proposals], f)

    logger.info(f"updated {args.out_file}")


def get_template_val(tpl, name):
    param = tpl.get(name, None)
    if param:
        value = param.value.strip()
        if value:
            # turn empty strings into None
            return value


def is_stub(doc):
    if any(
        doc.ifilter_templates(matches=lambda t: t.name.matches('Archived proposal'))
    ):
        return False

    if not any(doc.ifilter_headings()):
        # detect proposals without headings as stubs
        return True

    if not any(
        n
        for n in doc.nodes
        # any text
        if isinstance(n, mwparserfromhell.nodes.text.Text)
        # other than newlines
        and n.strip()
        # and "Please comment on the [[{{TALKPAGENAME}}|discussion page]]."
        and n.strip() not in ('Please comment on the', '.')
    ):
        # detect proposals without text as stubs
        return True

    return False


def parse_proposal(page_title: str, text: str, langs: Container[str]) -> dict | None:
    doc = mwparserfromhell.parse(text)
    proposal_page_templates = doc.filter_templates(
        matches=lambda t: t.name.matches('Proposal page')
        or t.name.matches('Proposal Page')
    )

    if not proposal_page_templates:
        logger.info('{{Proposal Page}} not found', page=page_title)
        return None

    for comment in doc.ifilter_comments():
        # remove comments like <!-- Date the RFC email is sent to the Tagging list: YYYY-MM-DD -->
        doc.remove(comment)

    tpl = proposal_page_templates[0]

    status = get_template_val(tpl, 'status')
    if status:
        status = status.lower()

    if is_stub(doc):
        if status in ('approved', 'rejected'):
            logger.info(f'{status} proposal is a stub', page=page_title)
        else:
            logger.info('skipping stub', page=page_title)
            return None

    name = get_template_val(tpl, 'name')
    if name:
        name = html.unescape(name)

    draft_start = get_template_val(tpl, 'draftStartDate')
    if draft_start in ('*', '-'):
        draft_start = None

    rfc_start = get_template_val(tpl, 'rfcStartDate')
    if rfc_start in ('*', '-'):
        rfc_start = None

    vote_start = get_template_val(tpl, 'voteStartDate')
    if vote_start in ('*', '-'):
        vote_start = None

    definition = get_template_val(tpl, 'definition')
    users = get_template_val(tpl, 'users') or get_template_val(tpl, 'user')

    parts = page_title.split(':', maxsplit=1)
    parts[0] = parts[0].lower()

    lang = None
    if parts[0] in langs:
        lang = parts[0]

    return dict(
        page_title=page_title,
        lang=lang,
        name=name,
        status=status,
        definition=definition,
        draft_start=draft_start,
        rfc_start=rfc_start,
        vote_start=vote_start,
        authors=users,
    )


STATUSES = {
    'voting': 0,
    'post-vote': 1,
    'proposed': 2,
    'draft': 3,
    'approved': 4,
    'inactive': 5,
    'rejected': 6,
    'abandoned': 7,
    'canceled': 8,
    'obsoleted': 9,
}


def sort_key(proposal):
    status = proposal['status']

    if status in ('voting', 'approved', 'rejected'):
        date = proposal['vote_start'] or ''
    elif status == 'proposed':
        date = proposal['rfc_start'] or ''
    else:
        date = proposal['draft_start'] or ''

    return (-STATUSES.get(proposal['status'], 10), date)