4 Check lesson files and their contents.
11 from argparse import ArgumentParser
13 from util import (Reporter, read_markdown, load_yaml, check_unwanted_files,
18 # Where to look for source Markdown files.
19 SOURCE_DIRS = ['', '_episodes', '_extras']
21 # Where to look for source Rmd files.
22 SOURCE_RMD_DIRS = ['_episodes_rmd']
24 # Required files: each entry is ('path': YAML_required).
25 # FIXME: We do not yet validate whether any files have the required
26 # YAML headers, but should in the future.
27 # The '%' is replaced with the source directory path for checking.
28 # Episodes are handled specially, and extra files in '_extras' are also handled
29 # specially. This list must include all the Markdown files listed in the
30 # 'bin/initialize' script.
32 '%/CODE_OF_CONDUCT.md': True,
33 '%/CONTRIBUTING.md': False,
36 '%/_extras/discuss.md': True,
37 '%/_extras/guide.md': True,
38 '%/_config.yml': True,
40 '%/reference.md': True,
44 # Episode filename pattern.
45 P_EPISODE_FILENAME = re.compile(r'/_episodes/(\d\d)-[-\w]+.md$')
47 # Pattern to match lines ending with whitespace.
48 P_TRAILING_WHITESPACE = re.compile(r'\s+$')
50 # Pattern to match figure references in HTML.
51 P_FIGURE_REFS = re.compile(r'<img[^>]+src="([^"]+)"[^>]*>')
53 # Pattern to match internally-defined Markdown links.
54 P_INTERNAL_LINK_REF = re.compile(r'\[([^\]]+)\]\[([^\]]+)\]')
56 # Pattern to match reference links (to resolve internally-defined references).
57 P_INTERNAL_LINK_DEF = re.compile(r'^\[([^\]]+)\]:\s*(.+)')
59 # What kinds of blockquotes are allowed?
73 # What kinds of code fragments are allowed?
88 # What fields are required in teaching episode metadata?
89 TEACHING_METADATA_FIELDS = {
98 # What fields are required in break episode metadata?
99 BREAK_METADATA_FIELDS = {
105 # How long are lines allowed to be?
106 # Please keep this in sync with .editorconfig!
114 args.reporter = Reporter()
115 check_config(args.reporter, args.source_dir)
116 check_source_rmd(args.reporter, args.source_dir, args.parser)
117 args.references = read_references(args.reporter, args.reference_path)
119 docs = read_all_markdown(args.source_dir, args.parser)
120 check_fileset(args.source_dir, args.reporter, list(docs.keys()))
121 check_unwanted_files(args.source_dir, args.reporter)
122 for filename in list(docs.keys()):
123 checker = create_checker(args, filename, docs[filename])
126 args.reporter.report()
127 if args.reporter.messages and not args.permissive:
132 """Parse command-line arguments."""
134 parser = ArgumentParser(description="""Check episode files in a lesson.""")
135 parser.add_argument('-l', '--linelen',
139 help='Check line lengths')
140 parser.add_argument('-p', '--parser',
143 help='path to Markdown parser')
144 parser.add_argument('-r', '--references',
146 dest='reference_path',
147 help='path to Markdown file of external references')
148 parser.add_argument('-s', '--source',
151 help='source directory')
152 parser.add_argument('-w', '--whitespace',
155 dest='trailing_whitespace',
156 help='Check for trailing whitespace')
157 parser.add_argument('--permissive',
161 help='Do not raise an error even if issues are detected')
163 args, extras = parser.parse_known_args()
164 require(args.parser is not None,
165 'Path to Markdown parser not provided')
167 'Unexpected trailing command-line arguments "{0}"'.format(extras))
172 def check_config(reporter, source_dir):
173 """Check configuration file."""
175 config_file = os.path.join(source_dir, '_config.yml')
176 config = load_yaml(config_file)
177 reporter.check_field(config_file, 'configuration',
178 config, 'kind', 'lesson')
179 reporter.check_field(config_file, 'configuration',
180 config, 'carpentry', ('swc', 'dc', 'lc', 'cp'))
181 reporter.check_field(config_file, 'configuration', config, 'title')
182 reporter.check_field(config_file, 'configuration', config, 'email')
185 {'values': {'root': '.', 'layout': 'page'}},
186 {'values': {'root': '..', 'layout': 'episode'}, 'scope': {'type': 'episodes', 'path': ''}},
187 {'values': {'root': '..', 'layout': 'page'}, 'scope': {'type': 'extras', 'path': ''}}
189 reporter.check(defaults in config.get('defaults', []),
191 '"root" not set to "." in configuration')
193 def check_source_rmd(reporter, source_dir, parser):
194 """Check that Rmd episode files include `source: Rmd`"""
196 episode_rmd_dir = [os.path.join(source_dir, d) for d in SOURCE_RMD_DIRS]
197 episode_rmd_files = [os.path.join(d, '*.Rmd') for d in episode_rmd_dir]
199 for pat in episode_rmd_files:
200 for f in glob.glob(pat):
201 data = read_markdown(parser, f)
202 dy = data['metadata']
204 reporter.check_field(f, 'episode_rmd',
207 def read_references(reporter, ref_path):
208 """Read shared file of reference links, returning dictionary of valid references
209 {symbolic_name : URL}
215 with open(ref_path, 'r') as reader:
216 for (num, line) in enumerate(reader):
218 m = P_INTERNAL_LINK_DEF.search(line)
220 '{0}:{1} not valid reference:\n{2}'.format(ref_path, line_num, line.rstrip()))
224 'Empty reference at {0}:{1}'.format(ref_path, line_num))
225 reporter.check(name not in result,
227 'Duplicate reference {0} at line {1}',
229 reporter.check(url not in urls_seen,
231 'Duplicate definition of URL {0} at line {1}',
238 def read_all_markdown(source_dir, parser):
239 """Read source files, returning
240 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
243 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
244 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
246 for pat in all_patterns:
247 for filename in glob.glob(pat):
248 data = read_markdown(parser, filename)
250 result[filename] = data
254 def check_fileset(source_dir, reporter, filenames_present):
255 """Are all required files present? Are extraneous files present?"""
257 # Check files with predictable names.
258 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
259 missing = set(required) - set(filenames_present)
261 reporter.add(None, 'Missing required file {0}', m)
263 # Check episode files' names.
265 for filename in filenames_present:
266 if '_episodes' not in filename:
268 m = P_EPISODE_FILENAME.search(filename)
270 seen.append(m.group(1))
273 None, 'Episode {0} has badly-formatted filename', filename)
275 # Check for duplicate episode numbers.
276 reporter.check(len(seen) == len(set(seen)),
278 'Duplicate episode numbers {0} vs {1}',
279 sorted(seen), sorted(set(seen)))
281 # Check that numbers are consecutive.
282 seen = sorted([int(s) for s in seen])
284 for i in range(len(seen) - 1):
285 clean = clean and ((seen[i+1] - seen[i]) == 1)
286 reporter.check(clean,
288 'Missing or non-consecutive episode numbers {0}',
292 def create_checker(args, filename, info):
293 """Create appropriate checker for file."""
295 for (pat, cls) in CHECKERS:
296 if pat.search(filename):
297 return cls(args, filename, **info)
298 return NotImplemented
301 """Base class for checking Markdown files."""
303 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
304 """Cache arguments for checking."""
307 self.reporter = self.args.reporter # for convenience
308 self.filename = filename
309 self.metadata = metadata
310 self.metadata_len = metadata_len
320 self.check_metadata()
321 self.check_line_lengths()
322 self.check_trailing_whitespace()
323 self.check_blockquote_classes()
324 self.check_codeblock_classes()
325 self.check_defined_link_references()
327 def check_metadata(self):
328 """Check the YAML metadata."""
330 self.reporter.check(self.metadata is not None,
332 'Missing metadata entirely')
334 if self.metadata and (self.layout is not None):
335 self.reporter.check_field(
336 self.filename, 'metadata', self.metadata, 'layout', self.layout)
338 def check_line_lengths(self):
339 """Check the raw text of the lesson body."""
341 if self.args.line_lengths:
342 over = [i for (i, l, n) in self.lines if (
343 n > MAX_LINE_LEN) and (not l.startswith('!'))]
344 self.reporter.check(not over,
346 'Line(s) are too long: {0}',
347 ', '.join([str(i) for i in over]))
349 def check_trailing_whitespace(self):
350 """Check for whitespace at the ends of lines."""
352 if self.args.trailing_whitespace:
354 i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
355 self.reporter.check(not trailing,
357 'Line(s) end with whitespace: {0}',
358 ', '.join([str(i) for i in trailing]))
360 def check_blockquote_classes(self):
361 """Check that all blockquotes have known classes."""
363 for node in self.find_all(self.doc, {'type': 'blockquote'}):
364 cls = self.get_val(node, 'attr', 'class')
365 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
366 (self.filename, self.get_loc(node)),
367 'Unknown or missing blockquote type {0}',
370 def check_codeblock_classes(self):
371 """Check that all code blocks have known classes."""
373 for node in self.find_all(self.doc, {'type': 'codeblock'}):
374 cls = self.get_val(node, 'attr', 'class')
375 self.reporter.check(cls in KNOWN_CODEBLOCKS,
376 (self.filename, self.get_loc(node)),
377 'Unknown or missing code block type {0}',
380 def check_defined_link_references(self):
381 """Check that defined links resolve in the file.
383 Internally-defined links match the pattern [text][label].
387 for node in self.find_all(self.doc, {'type': 'text'}):
388 for match in P_INTERNAL_LINK_REF.findall(node['value']):
391 if link not in self.args.references:
392 result.add('"{0}"=>"{1}"'.format(text, link))
393 self.reporter.check(not result,
395 'Internally-defined links may be missing definitions: {0}',
396 ', '.join(sorted(result)))
398 def find_all(self, node, pattern, accum=None):
399 """Find all matches for a pattern."""
401 assert isinstance(pattern, dict), 'Patterns must be dictionaries'
404 if self.match(node, pattern):
406 for child in node.get('children', []):
407 self.find_all(child, pattern, accum)
410 def match(self, node, pattern):
411 """Does this node match the given pattern?"""
417 if isinstance(val, str):
420 elif isinstance(val, dict):
421 if not self.match(node[key], val):
426 def get_val(node, *chain):
427 """Get value one or more levels down."""
430 for selector in chain:
431 curr = curr.get(selector, None)
436 def get_loc(self, node):
437 """Convenience method to get node's line number."""
439 result = self.get_val(node, 'options', 'location')
440 if self.metadata_len is not None:
441 result += self.metadata_len
445 class CheckNonJekyll(CheckBase):
446 """Check a file that isn't translated by Jekyll."""
448 def check_metadata(self):
449 self.reporter.check(self.metadata is None,
451 'Unexpected metadata')
454 class CheckIndex(CheckBase):
455 """Check the main index page."""
457 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
458 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
459 self.layout = 'lesson'
461 def check_metadata(self):
462 super().check_metadata()
463 self.reporter.check(self.metadata.get('root', '') == '.',
465 'Root not set to "."')
468 class CheckEpisode(CheckBase):
469 """Check an episode page."""
472 """Run extra tests."""
475 self.check_reference_inclusion()
477 def check_metadata(self):
478 super().check_metadata()
480 if 'layout' in self.metadata:
481 if self.metadata['layout'] == 'break':
482 self.check_metadata_fields(BREAK_METADATA_FIELDS)
484 self.reporter.add(self.filename,
485 'Unknown episode layout "{0}"',
486 self.metadata['layout'])
488 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
490 def check_metadata_fields(self, expected):
491 """Check metadata fields."""
492 for (name, type_) in expected:
493 if name not in self.metadata:
494 self.reporter.add(self.filename,
495 'Missing metadata field {0}',
497 elif not isinstance(self.metadata[name], type_):
498 self.reporter.add(self.filename,
499 '"{0}" has wrong type in metadata ({1} instead of {2})',
500 name, type(self.metadata[name]), type_)
502 def check_reference_inclusion(self):
503 """Check that links file has been included."""
505 if not self.args.reference_path:
508 for (i, last_line, line_len) in reversed(self.lines):
513 'No non-empty lines in {0}'.format(self.filename))
515 include_filename = os.path.split(self.args.reference_path)[-1]
516 if include_filename not in last_line:
517 self.reporter.add(self.filename,
518 'episode does not include "{0}"',
522 class CheckReference(CheckBase):
523 """Check the reference page."""
525 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
526 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
527 self.layout = 'reference'
530 class CheckGeneric(CheckBase):
531 """Check a generic page."""
533 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
534 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
538 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
539 (re.compile(r'README\.md'), CheckNonJekyll),
540 (re.compile(r'index\.md'), CheckIndex),
541 (re.compile(r'reference\.md'), CheckReference),
542 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
543 (re.compile(r'aio\.md'), CheckNonJekyll),
544 (re.compile(r'.*\.md'), CheckGeneric)
548 if __name__ == '__main__':