2 Check lesson files and their contents.
9 from argparse import ArgumentParser
11 from util import (Reporter, read_markdown, load_yaml, check_unwanted_files,
16 # Where to look for source Markdown files.
17 SOURCE_DIRS = ['', '_episodes', '_extras']
19 # Where to look for source Rmd files.
20 SOURCE_RMD_DIRS = ['_episodes_rmd']
22 # Required files: each entry is ('path': YAML_required).
23 # FIXME: We do not yet validate whether any files have the required
24 # YAML headers, but should in the future.
25 # The '%' is replaced with the source directory path for checking.
26 # Episodes are handled specially, and extra files in '_extras' are also handled
27 # specially. This list must include all the Markdown files listed in the
28 # 'bin/initialize' script.
30 'CODE_OF_CONDUCT.md': True,
31 'CONTRIBUTING.md': False,
34 os.path.join('_extras', 'discuss.md'): True,
35 os.path.join('_extras', 'guide.md'): True,
41 # Episode filename pattern.
42 P_EPISODE_FILENAME = re.compile(r'(\d\d)-[-\w]+.md$')
44 # Pattern to match lines ending with whitespace.
45 P_TRAILING_WHITESPACE = re.compile(r'\s+$')
47 # Pattern to match figure references in HTML.
48 P_FIGURE_REFS = re.compile(r'<img[^>]+src="([^"]+)"[^>]*>')
50 # Pattern to match internally-defined Markdown links.
51 P_INTERNAL_LINK_REF = re.compile(r'\[([^\]]+)\]\[([^\]]+)\]')
53 # Pattern to match reference links (to resolve internally-defined references).
54 P_INTERNAL_LINK_DEF = re.compile(r'^\[([^\]]+)\]:\s*(.+)')
56 # Pattern to match {% include ... %} statements
57 P_INTERNAL_INCLUDE_LINK = re.compile(r'^{% include ([^ ]*) %}$')
59 # What kinds of blockquotes are allowed?
73 # What kinds of code fragments are allowed?
91 # What fields are required in teaching episode metadata?
92 TEACHING_METADATA_FIELDS = {
101 # What fields are required in break episode metadata?
102 BREAK_METADATA_FIELDS = {
108 # How long are lines allowed to be?
109 # Please keep this in sync with .editorconfig!
117 args.reporter = Reporter()
118 check_config(args.reporter, args.source_dir)
119 check_source_rmd(args.reporter, args.source_dir, args.parser)
120 args.references = read_references(args.reporter, args.reference_path)
122 docs = read_all_markdown(args.source_dir, args.parser)
123 check_fileset(args.source_dir, args.reporter, list(docs.keys()))
124 check_unwanted_files(args.source_dir, args.reporter)
125 for filename in list(docs.keys()):
126 checker = create_checker(args, filename, docs[filename])
129 args.reporter.report()
130 if args.reporter.messages and not args.permissive:
135 """Parse command-line arguments."""
137 parser = ArgumentParser(description="""Check episode files in a lesson.""")
138 parser.add_argument('-l', '--linelen',
142 help='Check line lengths')
143 parser.add_argument('-p', '--parser',
146 help='path to Markdown parser')
147 parser.add_argument('-r', '--references',
149 dest='reference_path',
150 help='path to Markdown file of external references')
151 parser.add_argument('-s', '--source',
154 help='source directory')
155 parser.add_argument('-w', '--whitespace',
158 dest='trailing_whitespace',
159 help='Check for trailing whitespace')
160 parser.add_argument('--permissive',
164 help='Do not raise an error even if issues are detected')
166 args, extras = parser.parse_known_args()
167 require(args.parser is not None,
168 'Path to Markdown parser not provided')
170 'Unexpected trailing command-line arguments "{0}"'.format(extras))
175 def check_config(reporter, source_dir):
176 """Check configuration file."""
178 config_file = os.path.join(source_dir, '_config.yml')
179 config = load_yaml(config_file)
180 reporter.check_field(config_file, 'configuration',
181 config, 'kind', 'lesson')
182 reporter.check_field(config_file, 'configuration',
183 config, 'carpentry', ('swc', 'dc', 'lc', 'cp'))
184 reporter.check_field(config_file, 'configuration', config, 'title')
185 reporter.check_field(config_file, 'configuration', config, 'email')
188 {'values': {'root': '.', 'layout': 'page'}},
189 {'values': {'root': '..', 'layout': 'episode'}, 'scope': {'type': 'episodes', 'path': ''}},
190 {'values': {'root': '..', 'layout': 'page'}, 'scope': {'type': 'extras', 'path': ''}}
192 reporter.check(defaults in config.get('defaults', []),
194 '"root" not set to "." in configuration')
196 def check_source_rmd(reporter, source_dir, parser):
197 """Check that Rmd episode files include `source: Rmd`"""
199 episode_rmd_dir = [os.path.join(source_dir, d) for d in SOURCE_RMD_DIRS]
200 episode_rmd_files = [os.path.join(d, '*.Rmd') for d in episode_rmd_dir]
202 for pat in episode_rmd_files:
203 for f in glob.glob(pat):
204 data = read_markdown(parser, f)
205 dy = data['metadata']
207 reporter.check_field(f, 'episode_rmd',
210 def read_references(reporter, ref_path):
211 """Read shared file of reference links, returning dictionary of valid references
212 {symbolic_name : URL}
216 raise Warning("No filename has been provided.")
221 with open(ref_path, 'r', encoding='utf-8') as reader:
222 for (num, line) in enumerate(reader, 1):
224 if P_INTERNAL_INCLUDE_LINK.search(line): continue
226 m = P_INTERNAL_LINK_DEF.search(line)
228 message = '{}: {} not a valid reference: {}'
229 require(m, message.format(ref_path, num, line.rstrip()))
234 message = 'Empty reference at {0}:{1}'
235 require(name, message.format(ref_path, num))
237 unique_name = name not in result
238 unique_url = url not in urls_seen
240 reporter.check(unique_name,
242 'Duplicate reference name {0} at line {1}',
245 reporter.check(unique_url,
247 'Duplicate definition of URL {0} at line {1}',
256 def read_all_markdown(source_dir, parser):
257 """Read source files, returning
258 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
261 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
262 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
264 for pat in all_patterns:
265 for filename in glob.glob(pat):
266 data = read_markdown(parser, filename)
268 result[filename] = data
272 def check_fileset(source_dir, reporter, filenames_present):
273 """Are all required files present? Are extraneous files present?"""
275 # Check files with predictable names.
276 required = [os.path.join(source_dir, p) for p in REQUIRED_FILES]
277 missing = set(required) - set(filenames_present)
279 reporter.add(None, 'Missing required file {0}', m)
281 # Check episode files' names.
283 for filename in filenames_present:
284 if '_episodes' not in filename:
287 # split path to check episode name
288 base_name = os.path.basename(filename)
289 m = P_EPISODE_FILENAME.search(base_name)
291 seen.append(m.group(1))
294 None, 'Episode {0} has badly-formatted filename', filename)
296 # Check for duplicate episode numbers.
297 reporter.check(len(seen) == len(set(seen)),
299 'Duplicate episode numbers {0} vs {1}',
300 sorted(seen), sorted(set(seen)))
302 # Check that numbers are consecutive.
303 seen = sorted([int(s) for s in seen])
305 for i in range(len(seen) - 1):
306 clean = clean and ((seen[i+1] - seen[i]) == 1)
307 reporter.check(clean,
309 'Missing or non-consecutive episode numbers {0}',
313 def create_checker(args, filename, info):
314 """Create appropriate checker for file."""
316 for (pat, cls) in CHECKERS:
317 if pat.search(filename):
318 return cls(args, filename, **info)
319 return NotImplemented
322 """Base class for checking Markdown files."""
324 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
325 """Cache arguments for checking."""
328 self.reporter = self.args.reporter # for convenience
329 self.filename = filename
330 self.metadata = metadata
331 self.metadata_len = metadata_len
341 self.check_metadata()
342 self.check_line_lengths()
343 self.check_trailing_whitespace()
344 self.check_blockquote_classes()
345 self.check_codeblock_classes()
346 self.check_defined_link_references()
348 def check_metadata(self):
349 """Check the YAML metadata."""
351 self.reporter.check(self.metadata is not None,
353 'Missing metadata entirely')
355 if self.metadata and (self.layout is not None):
356 self.reporter.check_field(
357 self.filename, 'metadata', self.metadata, 'layout', self.layout)
359 def check_line_lengths(self):
360 """Check the raw text of the lesson body."""
362 if self.args.line_lengths:
363 over = [i for (i, l, n) in self.lines if (
364 n > MAX_LINE_LEN) and (not l.startswith('!'))]
365 self.reporter.check(not over,
367 'Line(s) too long: {0}',
368 ', '.join([str(i) for i in over]))
370 def check_trailing_whitespace(self):
371 """Check for whitespace at the ends of lines."""
373 if self.args.trailing_whitespace:
375 i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
376 self.reporter.check(not trailing,
378 'Line(s) end with whitespace: {0}',
379 ', '.join([str(i) for i in trailing]))
381 def check_blockquote_classes(self):
382 """Check that all blockquotes have known classes."""
384 for node in self.find_all(self.doc, {'type': 'blockquote'}):
385 cls = self.get_val(node, 'attr', 'class')
386 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
387 (self.filename, self.get_loc(node)),
388 'Unknown or missing blockquote type {0}',
391 def check_codeblock_classes(self):
392 """Check that all code blocks have known classes."""
394 for node in self.find_all(self.doc, {'type': 'codeblock'}):
395 cls = self.get_val(node, 'attr', 'class')
396 self.reporter.check(cls in KNOWN_CODEBLOCKS,
397 (self.filename, self.get_loc(node)),
398 'Unknown or missing code block type {0}',
401 def check_defined_link_references(self):
402 """Check that defined links resolve in the file.
404 Internally-defined links match the pattern [text][label].
408 for node in self.find_all(self.doc, {'type': 'text'}):
409 for match in P_INTERNAL_LINK_REF.findall(node['value']):
412 if link not in self.args.references:
413 result.add('"{0}"=>"{1}"'.format(text, link))
414 self.reporter.check(not result,
416 'Internally-defined links may be missing definitions: {0}',
417 ', '.join(sorted(result)))
419 def find_all(self, node, pattern, accum=None):
420 """Find all matches for a pattern."""
422 assert isinstance(pattern, dict), 'Patterns must be dictionaries'
425 if self.match(node, pattern):
427 for child in node.get('children', []):
428 self.find_all(child, pattern, accum)
431 def match(self, node, pattern):
432 """Does this node match the given pattern?"""
438 if isinstance(val, str):
441 elif isinstance(val, dict):
442 if not self.match(node[key], val):
447 def get_val(node, *chain):
448 """Get value one or more levels down."""
451 for selector in chain:
452 curr = curr.get(selector, None)
457 def get_loc(self, node):
458 """Convenience method to get node's line number."""
460 result = self.get_val(node, 'options', 'location')
461 if self.metadata_len is not None:
462 result += self.metadata_len
466 class CheckNonJekyll(CheckBase):
467 """Check a file that isn't translated by Jekyll."""
469 def check_metadata(self):
470 self.reporter.check(self.metadata is None,
472 'Unexpected metadata')
475 class CheckIndex(CheckBase):
476 """Check the main index page."""
478 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
479 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
480 self.layout = 'lesson'
482 def check_metadata(self):
483 super().check_metadata()
484 self.reporter.check(self.metadata.get('root', '') == '.',
486 'Root not set to "."')
489 class CheckEpisode(CheckBase):
490 """Check an episode page."""
493 """Run extra tests."""
496 self.check_reference_inclusion()
498 def check_metadata(self):
499 super().check_metadata()
501 if 'layout' in self.metadata:
502 if self.metadata['layout'] == 'break':
503 self.check_metadata_fields(BREAK_METADATA_FIELDS)
505 self.reporter.add(self.filename,
506 'Unknown episode layout "{0}"',
507 self.metadata['layout'])
509 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
511 def check_metadata_fields(self, expected):
512 """Check metadata fields."""
513 for (name, type_) in expected:
514 if name not in self.metadata:
515 self.reporter.add(self.filename,
516 'Missing metadata field {0}',
518 elif not isinstance(self.metadata[name], type_):
519 self.reporter.add(self.filename,
520 '"{0}" has wrong type in metadata ({1} instead of {2})',
521 name, type(self.metadata[name]), type_)
523 def check_reference_inclusion(self):
524 """Check that links file has been included."""
526 if not self.args.reference_path:
529 for (i, last_line, line_len) in reversed(self.lines):
534 'No non-empty lines in {0}'.format(self.filename))
536 include_filename = os.path.split(self.args.reference_path)[-1]
537 if include_filename not in last_line:
538 self.reporter.add(self.filename,
539 'episode does not include "{0}"',
543 class CheckReference(CheckBase):
544 """Check the reference page."""
546 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
547 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
548 self.layout = 'reference'
551 class CheckGeneric(CheckBase):
552 """Check a generic page."""
554 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
555 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
559 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
560 (re.compile(r'README\.md'), CheckNonJekyll),
561 (re.compile(r'index\.md'), CheckIndex),
562 (re.compile(r'reference\.md'), CheckReference),
563 (re.compile(os.path.join('_episodes', '*\.md')), CheckEpisode),
564 (re.compile(r'.*\.md'), CheckGeneric)
568 if __name__ == '__main__':