#!/usr/bin/env python3

import re

import common
from shell_helpers import LF
import os
import subprocess

class Main(common.LkmcCliFunction):
    def __init__(self):
        super().__init__(
            defaults = {
                'show_time': False,
            },
            description='''\
https://cirosantilli.com/linux-kernel-module-cheat#build-the-documentation
''',
        )
        self.add_argument(
            '--github-pages',
            default=False,
            help='''
Build for GitHub pages instead of a local build. This redirects all links
from the README to example sources to GitHub rather than locally.
'''
        )

    def timed_main(self):
        asciidoctor_dir = os.path.join(self.env['root_dir'], 'asciidoctor')
        if self.env['github_pages']:
            link_target_script = 'link-target-github.rb'
        else:
            link_target_script = 'link-target-up.rb'
        common_cmd = [
            'bundle', LF,
            'exec', LF,
            'asciidoctor', LF,
            '--attribute', 'docinfo=shared', LF,
            '--failure-level', 'info', LF,
            '--require', os.path.join(asciidoctor_dir, link_target_script), LF,
            '--trace', LF,
            '--verbose', LF,
        ]
        exit_status = self.sh.run_cmd(
            common_cmd +
            [
                '--out-file', self.env['readme_out'], LF,
                self.env['readme'], LF,
            ],
            out_file=self.env['build_doc_log'],
        )
        if exit_status == 0:
            exit_status = self.sh.run_cmd(
                common_cmd +
                [
                    '-D', self.env['out_doc_dir'], LF,
                    '-a', 'multipage-level=6', LF,
                    '-b', 'multipage_html5', LF,
                    '-r', 'asciidoctor-multipage', LF,
                    self.env['readme'], LF,
                ],
                out_file=self.env['build_doc_multipage_log'],
            )

        # Check that all local files linked from README exist.
        external_link_re = re.compile('^https?://')
        for link in self.sh.check_output([
            os.path.join(asciidoctor_dir, 'extract-link-targets'),
            self.env['readme']
        ]).decode().splitlines():
            if not external_link_re.match(link):
                if not os.path.lexists(os.path.join(self.env['root_dir'], link)):
                    self.log_error('broken link to local file: ' + link)
                    exit_status = 1

        # Check that there are not links to the GitHub README.
        # https://github.com/isaacs/github/issues/1610
        for grep_line in self.sh.check_output(
            [
                'git',
                'grep',
                '--fixed-strings',
                self.env['github_repo_url'] + '#',
                LF
            ],
            cwd=self.env['root_dir'],
            raise_on_failure=False
        ).decode().splitlines():
            self.log_error('link to GitHub readme: {}'.format(
                grep_line
            ))
            exit_status = 1

        # Check that non-README links to README IDs exit.
        header_ids = set()
        grep_line_location_re = re.compile('^(.*?:\d+):')
        grep_line_hash_re = re.compile('^([a-z0-9_-]+)')
        for header_id in self.sh.check_output([
            os.path.join(asciidoctor_dir, 'extract-header-ids'),
            self.env['readme']
        ]).decode().splitlines():
            header_ids.add(header_id)
        for grep_line in self.sh.check_output(
            [
                'git',
                'grep',
                '--fixed-strings',
                self.env['homepage_url'] + '#',
                LF
            ],
            cwd=self.env['root_dir']
        ).decode().splitlines():
            url_index = grep_line.index(self.env['homepage_url'])
            hash_start_index = url_index + len(self.env['homepage_url'])
            if len(grep_line) > hash_start_index:
                hash_str = grep_line_hash_re.search(grep_line[hash_start_index + 1:]).group(1)
                if not hash_str in header_ids:
                    self.log_error('broken link to {} at {}'.format(
                        hash_str,
                        grep_line_location_re.search(grep_line).group(1))
                    )
                    exit_status = 1

        return exit_status

if __name__ == '__main__':
    Main().cli()