|
| 1 | +from compression import gzip |
| 2 | +import concurrent.futures |
| 3 | +from pathlib import Path |
| 4 | +import html.parser |
| 5 | +import functools |
| 6 | +import argparse |
| 7 | +import json |
| 8 | +import sys |
| 9 | +import re |
| 10 | + |
| 11 | + |
| 12 | +IGNORED_ID_RE = re.compile(r""" |
| 13 | + index-\d+ |
| 14 | + | id\d+ |
| 15 | + | [_a-z]+_\d+ |
| 16 | +""", re.VERBOSE) |
| 17 | + |
| 18 | + |
| 19 | +class IDGatherer(html.parser.HTMLParser): |
| 20 | + def __init__(self, ids): |
| 21 | + super().__init__() |
| 22 | + self.__ids = ids |
| 23 | + |
| 24 | + def handle_starttag(self, tag, attrs): |
| 25 | + for name, value in attrs: |
| 26 | + if name == 'id': |
| 27 | + if not IGNORED_ID_RE.fullmatch(value): |
| 28 | + self.__ids.add(value) |
| 29 | + |
| 30 | + |
| 31 | +def get_ids_from_file(path): |
| 32 | + ids = set() |
| 33 | + gatherer = IDGatherer(ids) |
| 34 | + with path.open() as file: |
| 35 | + while chunk := file.read(4096): |
| 36 | + gatherer.feed(chunk) |
| 37 | + return ids |
| 38 | + |
| 39 | + |
| 40 | +def gather_ids(htmldir, *, verbose_print): |
| 41 | + if not htmldir.joinpath('objects.inv').exists(): |
| 42 | + raise ValueError(f'{htmldir!r} is not a Sphinx HTML output directory') |
| 43 | + |
| 44 | + if sys._is_gil_enabled: |
| 45 | + pool = concurrent.futures.ProcessPoolExecutor() |
| 46 | + else: |
| 47 | + pool = concurrent.futures.ThreadPoolExecutor() |
| 48 | + tasks = {} |
| 49 | + for path in htmldir.glob('**/*.html'): |
| 50 | + relative_path = path.relative_to(htmldir) |
| 51 | + if '_static' in relative_path.parts: |
| 52 | + continue |
| 53 | + if 'whatsnew' in relative_path.parts: |
| 54 | + continue |
| 55 | + tasks[relative_path] = pool.submit(get_ids_from_file, path=path) |
| 56 | + |
| 57 | + ids_by_page = {} |
| 58 | + for relative_path, future in tasks.items(): |
| 59 | + verbose_print(relative_path) |
| 60 | + ids = future.result() |
| 61 | + ids_by_page[str(relative_path)] = future.result() |
| 62 | + verbose_print(f' - {len(ids)} ids found') |
| 63 | + |
| 64 | + common = set.intersection(*ids_by_page.values()) |
| 65 | + verbose_print(f'Filtering out {len(common)} common ids') |
| 66 | + for key, page_ids in ids_by_page.items(): |
| 67 | + ids_by_page[key] = sorted(page_ids - common) |
| 68 | + |
| 69 | + return ids_by_page |
| 70 | + |
| 71 | + |
| 72 | +def do_check(baseline, checked, excluded, *, verbose_print): |
| 73 | + successful = True |
| 74 | + for name, baseline_ids in sorted(baseline.items()): |
| 75 | + try: |
| 76 | + checked_ids = checked[name] |
| 77 | + except KeyError: |
| 78 | + successful = False |
| 79 | + print(f'{name}: (page missing)') |
| 80 | + print() |
| 81 | + else: |
| 82 | + missing_ids = set(baseline_ids) - set(checked_ids) |
| 83 | + if missing_ids: |
| 84 | + missing_ids = { |
| 85 | + a for a in missing_ids |
| 86 | + if not IGNORED_ID_RE.fullmatch(a) |
| 87 | + and (name, a) not in excluded |
| 88 | + } |
| 89 | + if missing_ids: |
| 90 | + successful = False |
| 91 | + for missing_id in sorted(missing_ids): |
| 92 | + print(f'{name}: {missing_id}') |
| 93 | + print() |
| 94 | + return successful |
| 95 | + |
| 96 | + |
| 97 | +def main(argv): |
| 98 | + parser = argparse.ArgumentParser() |
| 99 | + parser.add_argument( |
| 100 | + '-v', '--verbose', action='store_true', |
| 101 | + help='print out more information') |
| 102 | + subparsers = parser.add_subparsers(dest='command', required=True) |
| 103 | + |
| 104 | + collect = subparsers.add_parser( |
| 105 | + 'collect', |
| 106 | + help='collect IDs from a set of HTML files') |
| 107 | + collect.add_argument( |
| 108 | + 'htmldir', type=Path, |
| 109 | + help='directory with HTML documentation') |
| 110 | + collect.add_argument( |
| 111 | + '-o', '--outfile', |
| 112 | + help='File to save the result in; default <htmldir>/html-ids.json.gz') |
| 113 | + |
| 114 | + check = subparsers.add_parser( |
| 115 | + 'check', |
| 116 | + help='check two archives of IDs') |
| 117 | + check.add_argument( |
| 118 | + 'baseline_file', type=Path, |
| 119 | + help='file with baseline IDs') |
| 120 | + check.add_argument( |
| 121 | + 'checked_file', type=Path, |
| 122 | + help='file with checked IDs') |
| 123 | + check.add_argument( |
| 124 | + '-x', '--exclude-file', type=Path, |
| 125 | + help='file with IDs to exclude from the check') |
| 126 | + |
| 127 | + args = parser.parse_args(argv[1:]) |
| 128 | + |
| 129 | + if args.verbose: |
| 130 | + verbose_print = functools.partial(print, file=sys.stderr) |
| 131 | + else: |
| 132 | + def verbose_print(*args, **kwargs): |
| 133 | + """do nothing""" |
| 134 | + |
| 135 | + if args.command == 'collect': |
| 136 | + ids = gather_ids(args.htmldir, verbose_print=verbose_print) |
| 137 | + if args.outfile is None: |
| 138 | + args.outfile = args.htmldir / 'html-ids.json.gz' |
| 139 | + with gzip.open(args.outfile, 'wt') as zfile: |
| 140 | + json.dump({'ids_by_page': ids}, zfile) |
| 141 | + |
| 142 | + if args.command == 'check': |
| 143 | + with gzip.open(args.baseline_file) as zfile: |
| 144 | + baseline = json.load(zfile)['ids_by_page'] |
| 145 | + with gzip.open(args.checked_file) as zfile: |
| 146 | + checked = json.load(zfile)['ids_by_page'] |
| 147 | + excluded = set() |
| 148 | + if args.exclude_file: |
| 149 | + with open(args.exclude_file) as file: |
| 150 | + for line in file: |
| 151 | + line = line.strip() |
| 152 | + if line and not line.startswith('#'): |
| 153 | + name, sep, excluded_id = line.partition(':') |
| 154 | + if sep: |
| 155 | + excluded.add((name.strip(), excluded_id.strip())) |
| 156 | + if do_check(baseline, checked, excluded, verbose_print=verbose_print): |
| 157 | + verbose_print('All OK') |
| 158 | + else: |
| 159 | + sys.stdout.flush() |
| 160 | + print( |
| 161 | + 'ERROR: Removed IDs found', |
| 162 | + 'The above HTML IDs were removed from the documentation, ' |
| 163 | + + 'resulting in broken links. Please add them back.', |
| 164 | + sep='\n', |
| 165 | + file=sys.stderr) |
| 166 | + if args.exclude_file: |
| 167 | + print(f'Alternatively, add them to {args.exclude_file}.') |
| 168 | + |
| 169 | + |
| 170 | +if __name__ == '__main__': |
| 171 | + main(sys.argv) |
0 commit comments