#!/usr/bin/env python3
# ===----------------------------------------------------------------------===##
#
# Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
# See https://llvm.org/LICENSE.txt for license information.
# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
#
# ===----------------------------------------------------------------------===##

from typing import List, Dict, Tuple, IO, Optional
import copy
import csv
import json
import logging
import re
import subprocess

import click

# Number of the 'Libc++ Standards Conformance' project on Github
LIBCXX_CONFORMANCE_PROJECT = '31'

# Project ID for 'Libc++ Standards Conformance'
LIBCXX_CONFORMANCE_PROJECT_ID = 'PVT_kwDOAQWwKc4AlOgt'

# Field ID for the 'Meeting Voted' field of the 'Libc++ Standards Conformance' project.
LIBCXX_CONFORMANCE_MEETING_VOTED_ID = 'PVTF_lADOAQWwKc4AlOgtzgdUEXI'

# Field ID for the 'Status' field of the 'Libc++ Standards Conformance' project.
LIBCXX_CONFORMANCE_STATUS_ID = 'PVTSSF_lADOAQWwKc4AlOgtzgdUBak'

def extract_between_markers(text: str, begin_marker: str, end_marker: str) -> Optional[str]:
    """
    Given a string containing special markers, extract everything located beetwen these markers.

    If the beginning marker is not found, None is returned. If the beginning marker is found but
    there is no end marker, it is an error (this is done to avoid silently accepting inputs that
    are erroneous by mistake).
    """
    start = text.find(begin_marker)
    if start == -1:
        return None

    start += len(begin_marker) # skip the marker itself
    end = text.find(end_marker, start)
    if end == -1:
        raise ArgumentError(f"Could not find end marker {end_marker} in: {text[start:]}")

    return text[start:end]

def rst_to_markdown(text: str) -> str:
    """Best effort conversion from RST to Markdown assuming a single line of text."""
    return text.replace('``', '`').replace('\\', '')

def markdown_to_rst(text: str) -> str:
    """Best effort conversion from Markdown to RST assuming a single line of text."""
    return re.sub(r'(?<!`)`(?!`)', '``', text)

class PaperStatus:
    TODO = 1
    IN_PROGRESS = 2
    PARTIAL = 3
    DONE = 4
    NOTHING_TO_DO = 5

    _status: int

    _original: Optional[str]
    """
    Optional string from which the paper status was created. This is used to carry additional
    information from CSV rows, like any notes associated to the status.
    """

    def __init__(self, status: int, original: Optional[str] = None):
        self._status = status
        self._original = original

    def __eq__(self, other) -> bool:
        return self._status == other._status

    def __lt__(self, other) -> bool:
        relative_order = {
            PaperStatus.TODO: 0,
            PaperStatus.IN_PROGRESS: 1,
            PaperStatus.PARTIAL: 2,
            PaperStatus.DONE: 3,
            PaperStatus.NOTHING_TO_DO: 3,
        }
        return relative_order[self._status] < relative_order[other._status]

    @staticmethod
    def from_csv_entry(entry: str):
        """
        Parse a paper status out of a CSV row entry. Entries can look like:
        - '' (an empty string, which means the paper is not done yet)
        - '|In Progress|'
        - '|Partial|'
        - '|Complete|'
        - '|Nothing To Do|'
        """
        if entry == '':
            return PaperStatus(PaperStatus.TODO, entry)
        elif entry == '|In Progress|':
            return PaperStatus(PaperStatus.IN_PROGRESS, entry)
        elif entry == '|Partial|':
            return PaperStatus(PaperStatus.PARTIAL, entry)
        elif entry == '|Complete|':
            return PaperStatus(PaperStatus.DONE, entry)
        elif entry == '|Nothing To Do|':
            return PaperStatus(PaperStatus.NOTHING_TO_DO, entry)
        else:
            raise RuntimeError(f'Unexpected CSV entry for status: {entry}')

    @staticmethod
    def from_github_issue(issue: Dict):
        """
        Parse a paper status out of a Github issue obtained from querying a Github project.
        """
        if 'status' not in issue:
            return PaperStatus(PaperStatus.TODO)
        elif issue['status'] == 'Todo':
            return PaperStatus(PaperStatus.TODO)
        elif issue['status'] == 'In Progress':
            return PaperStatus(PaperStatus.IN_PROGRESS)
        elif issue['status'] == 'Partial':
            return PaperStatus(PaperStatus.PARTIAL)
        elif issue['status'] == 'Done':
            return PaperStatus(PaperStatus.DONE)
        elif issue['status'] == 'Nothing To Do':
            return PaperStatus(PaperStatus.NOTHING_TO_DO)
        else:
            raise RuntimeError(f"Received unrecognizable Github issue status: {issue['status']}")

    def to_csv_entry(self) -> str:
        """
        Return the issue state formatted for a CSV entry. The status is formatted as '|Complete|',
        '|In Progress|', etc.
        """
        mapping = {
            PaperStatus.TODO: '',
            PaperStatus.IN_PROGRESS: '|In Progress|',
            PaperStatus.PARTIAL: '|Partial|',
            PaperStatus.DONE: '|Complete|',
            PaperStatus.NOTHING_TO_DO: '|Nothing To Do|',
        }
        return self._original if self._original is not None else mapping[self._status]

    def to_project_single_select_option(self) -> str:
        """
        Maps this Status to a single-select option in the libc++ Standards Conformance Github project.
        """
        # Values for the various options of the 'Status' field in the libc++ Standards Conformance project.
        mapping = {
            PaperStatus.TODO: 'f75ad846',
            PaperStatus.IN_PROGRESS: '47fc9ee4',
            PaperStatus.PARTIAL: '44a9df65',
            PaperStatus.DONE: '98236657',
            PaperStatus.NOTHING_TO_DO: '18422253',
        }
        return mapping[self._status]

class PaperInfo:
    paper_number: str
    """
    Identifier for the paper or the LWG issue. This must be something like 'PnnnnRx', 'Nxxxxx' or 'LWGxxxxx'.
    """

    paper_name: str
    """
    Plain text string representing the name of the paper.
    """

    status: PaperStatus
    """
    Status of the paper/issue. This can be complete, in progress, partial, or done.
    """

    meeting: Optional[str]
    """
    Plain text string representing the meeting at which the paper/issue was voted.
    """

    first_released_version: Optional[str]
    """
    First version of LLVM in which this paper/issue was resolved.
    """

    github_issue: Optional[str]
    """
    Optional number of the Github issue tracking the implementation status of this paper.
    This is used to cross-reference rows in the status pages with Github issues.
    """

    notes: Optional[str]
    """
    Optional plain text string representing notes to associate to the paper.
    This is used to populate the "Notes" column in the CSV status pages.
    """

    original: Optional[object]
    """
    Object from which this PaperInfo originated. This is used to track the CSV row or Github issue that
    was used to generate this PaperInfo and is useful for error reporting purposes.
    """

    def __init__(self, paper_number: str, paper_name: str,
                       status: PaperStatus,
                       meeting: Optional[str] = None,
                       first_released_version: Optional[str] = None,
                       github_issue: Optional[str] = None,
                       notes: Optional[str] = None,
                       original: Optional[object] = None):
        self.paper_number = paper_number
        self.paper_name = paper_name
        self.status = status
        self.meeting = meeting
        self.first_released_version = first_released_version
        self.github_issue = github_issue
        self.notes = notes
        self.original = original

    def for_printing(self) -> Tuple[str, str, str, str, str, str, str]:
        return (
            f'`{self.paper_number} <https://wg21.link/{self.paper_number}>`__',
            markdown_to_rst(self.paper_name),
            self.meeting if self.meeting is not None else '',
            self.status.to_csv_entry(),
            self.first_released_version if self.first_released_version is not None else '',
            f'`#{self.github_issue} <https://github.com/llvm/llvm-project/issues/{self.github_issue}>`__' if self.github_issue is not None else '',
            self.notes if self.notes is not None else '',
        )

    def __repr__(self) -> str:
        return repr(self.original) if self.original is not None else repr(self.for_printing())

    @staticmethod
    def from_csv_row(row: Tuple[str, str, str, str, str, str]):# -> PaperInfo:
        """
        Given a row from one of our status-tracking CSV files, create a PaperInfo object representing that row.
        """
        # Extract the paper number from the first column
        match = re.search(r"((P[0-9R]+)|(LWG[0-9]+)|(N[0-9]+))\s+", row[0])
        if match is None:
            raise RuntimeError(f"Can't parse paper/issue number out of row: {row}")

        # Match the issue number if present
        github_issue = re.search(r'#([0-9]+)', row[5])
        if github_issue:
            github_issue = github_issue.group(1)

        try:
            status = PaperStatus.from_csv_entry(row[3])
        except:
            logging.error(f'Could not extract a status from the provided CSV row: {row}')
            raise

        return PaperInfo(
            paper_number=match.group(1),
            paper_name=row[1],
            status=status,
            meeting=row[2] or None,
            first_released_version=row[4] or None,
            github_issue=github_issue,
            notes=row[6] or None,
            original=row,
        )

    @staticmethod
    def from_github_issue(issue: Dict):# -> PaperInfo:
        """
        Create a PaperInfo object from the Github issue information obtained from querying a Github Project.
        """
        # Extract the paper number from the issue title
        match = re.search(r"((P[0-9R]+)|(LWG[0-9]+)|(N[0-9]+)):", issue['title'])
        if match is None:
            raise RuntimeError(f"Issue doesn't have a title that we know how to parse: {issue}")
        paper = match.group(1)

        # Extract any notes from the Github issue and populate the RST notes with them
        issue_description = issue['content']['body']
        notes = extract_between_markers(issue_description, 'BEGIN-RST-NOTES', 'END-RST-NOTES')
        notes = notes.strip() if notes is not None else notes

        return PaperInfo(
            paper_number=paper,
            paper_name=issue['title'].removeprefix(paper + ': '),
            status=PaperStatus.from_github_issue(issue),
            meeting=issue.get('meeting Voted', None),
            first_released_version=None, # TODO
            github_issue=str(issue['content']['number']),
            notes=notes,
            original=issue,
        )

def merge(paper: PaperInfo, gh: PaperInfo) -> PaperInfo:
    """
    Merge a paper coming from a CSV row with a corresponding Github-tracked paper.

    If the CSV row has a status that is "less advanced" than the Github issue, simply update the CSV
    row with the newer status. Otherwise, report an error if they have a different status because
    something must be wrong.

    We don't update issues from 'To Do' to 'In Progress', since that only creates churn and the
    status files aim to document user-facing functionality in releases, for which 'In Progress'
    is not useful.

    In case we don't update the CSV row's status, we still take any updated notes coming
    from the Github issue and we add a link to the Github issue if it was previously missing.
    """
    took_gh_in_full = False # Whether we updated the entire PaperInfo from the Github version
    if paper.status == PaperStatus(PaperStatus.TODO) and gh.status == PaperStatus(PaperStatus.IN_PROGRESS):
        result = copy.deepcopy(paper)
    elif paper.status < gh.status:
        result = copy.deepcopy(gh)
        took_gh_in_full = True
    elif paper.status == gh.status:
        result = copy.deepcopy(paper)
    else:
        logging.info(f"We found a CSV row and a Github issue with different statuses:\nrow: {paper}\nGithub issue: {gh}")
        result = copy.deepcopy(paper)

    # If we didn't take the Github issue in full, make sure to update the notes, the link and anything else.
    if not took_gh_in_full:
        result.github_issue = gh.github_issue
        result.notes = gh.notes
    return result

def load_csv(file: IO) -> List[Tuple]:
    rows = []
    reader = csv.reader(file, delimiter=',')
    for row in reader:
        rows.append(row)
    return rows

def write_csv(output: IO, rows: List[Tuple]):
    writer = csv.writer(output, quoting=csv.QUOTE_ALL, lineterminator='\n')
    for row in rows:
        writer.writerow(row)

def link_github_issue(issue: str, meeting_voted: Optional[str], status: Optional[PaperStatus]) -> None:
    """
    Link an existing Github issue to the libc++ Conformance project.
    """
    # Figure out the URL of the issue
    url = subprocess.check_output(['gh', 'issue', 'view', issue, '--json', 'url', '--jq', '.url']).decode().strip()

    # Add the issue to the Conformance project
    cli = ['gh', 'project', 'item-add', LIBCXX_CONFORMANCE_PROJECT, '--owner', 'llvm', '--url', url, '--format', 'json', '--jq', '.id']
    id = subprocess.check_output(cli).decode().strip()

    # Adjust the 'Meeting Voted' field of the item
    if meeting_voted is not None:
        cli = ['gh', 'project', 'item-edit', '--id', id,
                                '--project-id', LIBCXX_CONFORMANCE_PROJECT_ID,
                                '--field-id', LIBCXX_CONFORMANCE_MEETING_VOTED_ID, '--text', meeting_voted]
        subprocess.check_call(cli)

    # Adjust the 'Status' field of the item
    if status is not None:
        option_id = status.to_project_single_select_option()
        cli = ['gh', 'project', 'item-edit', '--id', id,
                                '--project-id', LIBCXX_CONFORMANCE_PROJECT_ID,
                                '--field-id', LIBCXX_CONFORMANCE_STATUS_ID,
                                '--single-select-option-id', option_id]
        subprocess.check_call(cli)

def sync_csv(rows: List[Tuple], from_github: List[PaperInfo]) -> List[Tuple]:
    """
    Given a list of CSV rows representing an existing status file and a list of PaperInfos representing
    up-to-date (but potentially incomplete) tracking information from Github, this function returns the
    new CSV rows synchronized with the up-to-date information.

    Note that this only tracks changes from 'not implemented' issues to 'implemented'. If an up-to-date
    PaperInfo reports that a paper is not implemented but the existing CSV rows report it as implemented,
    it is an error (i.e. the result is not a CSV row where the paper is *not* implemented).
    """
    results = [rows[0]] # Start with the header
    for row in rows[1:]: # Skip the header
        # If the row contains empty entries, this is a "separator row" between meetings.
        # Preserve it as-is.
        if row[0] == "":
            results.append(row)
            continue

        paper = PaperInfo.from_csv_row(row)

        # Find any Github issues tracking this paper. Each row must have one and exactly one Github
        # issue tracking it, which we validate below.
        tracking = [gh for gh in from_github if paper.paper_number == gh.paper_number]

        # If there's more than one tracking issue, something is weird.
        if len(tracking) > 1:
            logging.warning(f"Found a row with more than one tracking issue: {row}")
            for t in tracking:
                logging.warning(f'  tracked by: {t}')
            results.append(row)
            continue

        # Validate the Github issue associated to the CSV row, if any
        if paper.github_issue is not None:
            if len(tracking) == 0:
                logging.warning(f"Found row claiming to have a tracking issue, but failed to find a tracking issue on Github: {row}")
                results.append(row)
                continue
            if len(tracking) == 1 and paper.github_issue != tracking[0].github_issue:
                logging.warning(f"Found row with incorrect tracking issue: {row}\ntracked by: {tracking[0]}")
                results.append(row)
                continue

        # If there is no tracking issue for that row, log that we're missing an issue.
        if len(tracking) == 0:
            logging.warning(f"Can't find any Github issue for CSV row: {row}")
            results.append(row)
            continue

        results.append(merge(paper, tracking[0]).for_printing())

    return results

class ConformanceProject:
    def __init__(self, _json):
        self._json = _json

    @staticmethod
    def query(): # -> ConformanceProject
        cli = ['gh', 'project', 'item-list', LIBCXX_CONFORMANCE_PROJECT, '--owner', 'llvm', '--format', 'json', '--limit', '9999999']
        js = json.loads(subprocess.check_output(cli))
        return ConformanceProject(js)

    @staticmethod
    def load_from(f: IO): # -> ConformanceProject
        js = json.load(f)
        return ConformanceProject(js)

    def save(self, output: IO) -> None:
        json.dump(self._json, output)

    def papers(self) -> List[PaperInfo]:
        """
        Return the list of papers tracked by the conformance project.
        """
        return [PaperInfo.from_github_issue(issue) for issue in self._json['items']]

    def tracks(self, paper_number: str) -> bool:
        """
        Return whether the conformance project is tracking the given paper number.
        """
        for paper in self.papers():
            if paper.paper_number == paper_number:
                return True
        return False


@click.group()
def main():
    """Manipulate libc++'s various source of truth for tracking conformance."""
    pass


@main.group('github')
def github_cmd():
    """Perform actions on Github issues that track libc++ conformance."""
    pass


@github_cmd.command('find-unlinked')
@click.option('--labels', multiple=True, default=('lwg-issue', 'wg21-paper'),
    help='The label(s) to search for. This can be lwg-issue or wg21-paper. '
         'Passing this flag multiple times will OR the provided labels.')
def github_find_unlinked(labels):
    """Find issues not linked to the Conformance project."""
    query = f'is:issue -project:llvm/{LIBCXX_CONFORMANCE_PROJECT} label:libc++ label:{",".join(labels)}'
    cli = ['gh', 'issue', 'list', '--search', query, '--limit', '1000']
    subprocess.check_call(cli)


@github_cmd.command('link')
@click.argument('issues', nargs=-1)
@click.option('--meeting-voted', required=False,
    help='The optional meeting at which these issues have been voted. If provided, this is used to set the '
         'Meeting Voted field in the Conformance project.')
@click.option('--status', required=False, type=click.Choice(['todo', 'inprogress', 'partial', 'done', 'nothingtodo'],
                                                            case_sensitive=False),
    help='The optional status to set the issue to in the Conformance project.')
def github_link(issues, meeting_voted, status):
    """Link issues to the Conformance project.

        ISSUES are the issues to link to the Conformance project. This can be anything that the gh
        command-line tool understands, such as an issue number or a URL.
    """
    mapping = {'todo': PaperStatus.TODO, 'inprogress': PaperStatus.IN_PROGRESS, 'partial': PaperStatus.PARTIAL,
               'done': PaperStatus.DONE, 'nothingtodo': PaperStatus.NOTHING_TO_DO}
    if status is not None:
        status = PaperStatus(mapping[status.lower()])
    for issue in issues:
        link_github_issue(issue, meeting_voted, status)

@github_cmd.command('create')
@click.argument('csvfiles', nargs=-1, type=click.File('r'))
@click.option('--load-github-from', type=click.File('r'), required=False,
    help='Optionally load the Github Conformance tracking data from the given file. Useful to avoid '
         'running into API call limits when testing locally. See `conformance github download`.')
@click.option('--labels', type=str, multiple=True,
    help='Optional labels to add to the issue being created. Either lwg-issue or wg21-paper must be provided, '
         'but both are mutually exclusive.')
def github_create(csvfiles, load_github_from, labels):
    """Create Github issues tracking libc++ conformance from CSV status files."""
    if 'lwg-issue' not in labels and 'wg21-paper' not in labels:
        raise click.BadParameter('Either lwg-issue or wg21-paper label should be provided', param_hint='--labels')

    if load_github_from is None:
        logging.info("Loading conformance information from Github")
        project = ConformanceProject.query()
    else:
        logging.info(f"Loading conformance information from {load_github_from.name}")
        project = ConformanceProject.load_from(load_github_from)

    def all_csv_rows():
        for file in csvfiles:
            csv = load_csv(file)
            for row in csv[1:]: # Skip the header
                # If the row contains empty entries, this is a "separator row" between meetings.
                if row[0] == "":
                    continue
                yield row

    for row in all_csv_rows():
        paper = PaperInfo.from_csv_row(row)

        # Only consider creating a Github issue for papers that are not already tracked.
        if project.tracks(paper.paper_number):
            continue

        # If the row pretends to be tracked by a Github issue but we can't find it, something is weird.
        if paper.github_issue is not None:
            logging.warning(f"Found row claiming to have a tracking issue, but failed to find a tracking issue on Github: {row}")
            continue

        # Create the actual Github issue
        cli = ['gh', 'issue', 'create', '--repo', 'llvm/llvm-project',
                    '--title', f'{paper.paper_number}: {rst_to_markdown(paper.paper_name)}',
                    '--body', f'**Link:** https://wg21.link/{paper.paper_number}',
                    '--project', 'libc++ Standards Conformance',
                    '--label', 'libc++']

        for label in labels:
            cli += ['--label', label]

        print("Do you want to create the following issue?")
        print(cli)
        answer = input("y/n: ")
        if answer == 'n':
            print("Not creating issue")
            continue
        elif answer != 'y':
            print(f"Invalid answer {answer}, skipping")
            continue

        logging.info("Creating issue")
        issue_link = subprocess.check_output(cli).decode().strip()
        logging.info(f"Created tracking issue for {paper.paper_number}: {issue_link}")

        logging.info(f"Linking {issue_link} to the libc++ Standards Conformance project")
        link_github_issue(issue_link, paper.meeting, paper.status)


@github_cmd.command('download')
@click.option('--output', '-o', type=click.File('w'), default='-',
    help='Save the data to the provided file. Default is stdout.')
def github_download(output):
    """Download libc++ conformance tracking data from Github issues and save it locally."""
    project = ConformanceProject.query()
    project.save(output)


@main.group('csv')
def csv_cmd():
    """Perform actions on the CSV files that track libc++ conformance."""
    pass


@csv_cmd.command('validate')
@click.argument('files', nargs=-1, type=click.File('r'))
def csv_validate(files):
    """Validate the format of the provided CSV files."""
    for file in files:
        csv = load_csv(file)
        for row in csv[1:]: # Skip the header
            if row[0] != "": # Skip separator rows
                PaperInfo.from_csv_row(row)


@csv_cmd.command('synchronize')
@click.argument('files', nargs=-1, type=click.File('r'))
@click.option('--output', '-o', type=click.File('w'), default='-',
    help='Write the generated CSV to the specified file. Defaults to stdout.')
@click.option('--load-github-from', type=click.File('r'), required=False,
    help='Optionally load the Github Conformance tracking data from the given file. Useful to avoid '
         'running into API call limits when testing locally. See `conformance github download`.')
def csv_synchronize(files, output, load_github_from):
    """Synchronize existing CSV status files based on existing Github issues."""
    if load_github_from is None:
        logging.info("Loading conformance information from Github")
        project = ConformanceProject.query()
    else:
        logging.info(f"Loading conformance information from {load_github_from.name}")
        project = ConformanceProject.load_from(load_github_from)

    for file in files:
        logging.info(f"Synchronizing {file.name} with Github issues")
        csv = load_csv(file)
        synced = sync_csv(csv, project.papers())
        write_csv(output, synced)


@csv_cmd.command('create')
@click.argument('issues', nargs=-1)
@click.option('--output', '-o', type=click.File('w'), default='-',
    help='Write the generated CSV to the specified file. Defaults to stdout.')
@click.option('--load-github-from', type=click.File('r'), required=False,
    help='Optionally load the Github Conformance tracking data from the given file. Useful to avoid '
         'running into API call limits when testing locally. See `conformance github download`.')
def csv_create(issues, output, load_github_from):
    """Produce CSV entries corresponding to the given Github issues."""
    if load_github_from is None:
        logging.info("Loading conformance information from Github")
        project = ConformanceProject.query()
    else:
        logging.info(f"Loading conformance information from {load_github_from.name}")
        project = ConformanceProject.load_from(load_github_from)

    rows = []
    for issue in issues:
        # Figure out the issue number
        number = subprocess.check_output(['gh', 'issue', 'view', issue, '--json', 'number', '--jq', '.number']).decode().strip()

        # Find the PaperInfo corresponding to that issue. If there is none, the issue provided by the user is invalid.
        for paper in project.papers():
            if paper.github_issue == number:
                rows.append(paper.for_printing())
                break
        else:
            raise RuntimeError(f'Issue {issue} does not have a corresponding tracking issue in Github, aborting')

    rows.sort()
    write_csv(output, rows)


if __name__ == '__main__':
    main()