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,
39 '%/reference.md': True,
43 # Episode filename pattern.
44 P_EPISODE_FILENAME = re.compile(r'/_episodes/(\d\d)-[-\w]+.md$')
46 # Pattern to match lines ending with whitespace.
47 P_TRAILING_WHITESPACE = re.compile(r'\s+$')
49 # Pattern to match figure references in HTML.
50 P_FIGURE_REFS = re.compile(r'<img[^>]+src="([^"]+)"[^>]*>')
52 # Pattern to match internally-defined Markdown links.
53 P_INTERNAL_LINK_REF = re.compile(r'\[([^\]]+)\]\[([^\]]+)\]')
55 # Pattern to match reference links (to resolve internally-defined references).
56 P_INTERNAL_LINK_DEF = re.compile(r'^\[([^\]]+)\]:\s*(.+)')
58 # Pattern to match {% include ... %} statements
59 P_INTERNAL_INCLUDE_LINK = re.compile(r'^{% include ([^ ]*) %}$')
61 # What kinds of blockquotes are allowed?
75 # What kinds of code fragments are allowed?
90 # What fields are required in teaching episode metadata?
91 TEACHING_METADATA_FIELDS = {
100 # What fields are required in break episode metadata?
101 BREAK_METADATA_FIELDS = {
107 # How long are lines allowed to be?
108 # Please keep this in sync with .editorconfig!
116 args.reporter = Reporter()
117 check_config(args.reporter, args.source_dir)
118 check_source_rmd(args.reporter, args.source_dir, args.parser)
119 args.references = read_references(args.reporter, args.reference_path)
121 docs = read_all_markdown(args.source_dir, args.parser)
122 check_fileset(args.source_dir, args.reporter, list(docs.keys()))
123 check_unwanted_files(args.source_dir, args.reporter)
124 for filename in list(docs.keys()):
125 checker = create_checker(args, filename, docs[filename])
128 args.reporter.report()
129 if args.reporter.messages and not args.permissive:
134 """Parse command-line arguments."""
136 parser = ArgumentParser(description="""Check episode files in a lesson.""")
137 parser.add_argument('-l', '--linelen',
141 help='Check line lengths')
142 parser.add_argument('-p', '--parser',
145 help='path to Markdown parser')
146 parser.add_argument('-r', '--references',
148 dest='reference_path',
149 help='path to Markdown file of external references')
150 parser.add_argument('-s', '--source',
153 help='source directory')
154 parser.add_argument('-w', '--whitespace',
157 dest='trailing_whitespace',
158 help='Check for trailing whitespace')
159 parser.add_argument('--permissive',
163 help='Do not raise an error even if issues are detected')
165 args, extras = parser.parse_known_args()
166 require(args.parser is not None,
167 'Path to Markdown parser not provided')
169 'Unexpected trailing command-line arguments "{0}"'.format(extras))
174 def check_config(reporter, source_dir):
175 """Check configuration file."""
177 config_file = os.path.join(source_dir, '_config.yml')
178 config = load_yaml(config_file)
179 reporter.check_field(config_file, 'configuration',
180 config, 'kind', 'lesson')
181 reporter.check_field(config_file, 'configuration',
182 config, 'carpentry', ('swc', 'dc', 'lc', 'cp'))
183 reporter.check_field(config_file, 'configuration', config, 'title')
184 reporter.check_field(config_file, 'configuration', config, 'email')
187 {'values': {'root': '.', 'layout': 'page'}},
188 {'values': {'root': '..', 'layout': 'episode'}, 'scope': {'type': 'episodes', 'path': ''}},
189 {'values': {'root': '..', 'layout': 'page'}, 'scope': {'type': 'extras', 'path': ''}}
191 reporter.check(defaults in config.get('defaults', []),
193 '"root" not set to "." in configuration')
195 def check_source_rmd(reporter, source_dir, parser):
196 """Check that Rmd episode files include `source: Rmd`"""
198 episode_rmd_dir = [os.path.join(source_dir, d) for d in SOURCE_RMD_DIRS]
199 episode_rmd_files = [os.path.join(d, '*.Rmd') for d in episode_rmd_dir]
201 for pat in episode_rmd_files:
202 for f in glob.glob(pat):
203 data = read_markdown(parser, f)
204 dy = data['metadata']
206 reporter.check_field(f, 'episode_rmd',
209 def read_references(reporter, ref_path):
210 """Read shared file of reference links, returning dictionary of valid references
211 {symbolic_name : URL}
215 raise Warning("No filename has been provided.")
220 with open(ref_path, 'r') as reader:
221 for (num, line) in enumerate(reader, 1):
223 if P_INTERNAL_INCLUDE_LINK.search(line): continue
225 m = P_INTERNAL_LINK_DEF.search(line)
227 message = '{}: {} not a valid reference: {}'
228 require(m, message.format(ref_path, num, line.rstrip()))
233 message = 'Empty reference at {0}:{1}'
234 require(name, message.format(ref_path, num))
236 unique_name = name not in result
237 unique_url = url not in urls_seen
239 reporter.check(unique_name,
241 'Duplicate reference name {0} at line {1}',
244 reporter.check(unique_url,
246 'Duplicate definition of URL {0} at line {1}',
255 def read_all_markdown(source_dir, parser):
256 """Read source files, returning
257 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
260 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
261 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
263 for pat in all_patterns:
264 for filename in glob.glob(pat):
265 data = read_markdown(parser, filename)
267 result[filename] = data
271 def check_fileset(source_dir, reporter, filenames_present):
272 """Are all required files present? Are extraneous files present?"""
274 # Check files with predictable names.
275 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
276 missing = set(required) - set(filenames_present)
278 reporter.add(None, 'Missing required file {0}', m)
280 # Check episode files' names.
282 for filename in filenames_present:
283 if '_episodes' not in filename:
285 m = P_EPISODE_FILENAME.search(filename)
287 seen.append(m.group(1))
290 None, 'Episode {0} has badly-formatted filename', filename)
292 # Check for duplicate episode numbers.
293 reporter.check(len(seen) == len(set(seen)),
295 'Duplicate episode numbers {0} vs {1}',
296 sorted(seen), sorted(set(seen)))
298 # Check that numbers are consecutive.
299 seen = sorted([int(s) for s in seen])
301 for i in range(len(seen) - 1):
302 clean = clean and ((seen[i+1] - seen[i]) == 1)
303 reporter.check(clean,
305 'Missing or non-consecutive episode numbers {0}',
309 def create_checker(args, filename, info):
310 """Create appropriate checker for file."""
312 for (pat, cls) in CHECKERS:
313 if pat.search(filename):
314 return cls(args, filename, **info)
315 return NotImplemented
318 """Base class for checking Markdown files."""
320 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
321 """Cache arguments for checking."""
324 self.reporter = self.args.reporter # for convenience
325 self.filename = filename
326 self.metadata = metadata
327 self.metadata_len = metadata_len
337 self.check_metadata()
338 self.check_line_lengths()
339 self.check_trailing_whitespace()
340 self.check_blockquote_classes()
341 self.check_codeblock_classes()
342 self.check_defined_link_references()
344 def check_metadata(self):
345 """Check the YAML metadata."""
347 self.reporter.check(self.metadata is not None,
349 'Missing metadata entirely')
351 if self.metadata and (self.layout is not None):
352 self.reporter.check_field(
353 self.filename, 'metadata', self.metadata, 'layout', self.layout)
355 def check_line_lengths(self):
356 """Check the raw text of the lesson body."""
358 if self.args.line_lengths:
359 over = [i for (i, l, n) in self.lines if (
360 n > MAX_LINE_LEN) and (not l.startswith('!'))]
361 self.reporter.check(not over,
363 'Line(s) too long: {0}',
364 ', '.join([str(i) for i in over]))
366 def check_trailing_whitespace(self):
367 """Check for whitespace at the ends of lines."""
369 if self.args.trailing_whitespace:
371 i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
372 self.reporter.check(not trailing,
374 'Line(s) end with whitespace: {0}',
375 ', '.join([str(i) for i in trailing]))
377 def check_blockquote_classes(self):
378 """Check that all blockquotes have known classes."""
380 for node in self.find_all(self.doc, {'type': 'blockquote'}):
381 cls = self.get_val(node, 'attr', 'class')
382 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
383 (self.filename, self.get_loc(node)),
384 'Unknown or missing blockquote type {0}',
387 def check_codeblock_classes(self):
388 """Check that all code blocks have known classes."""
390 for node in self.find_all(self.doc, {'type': 'codeblock'}):
391 cls = self.get_val(node, 'attr', 'class')
392 self.reporter.check(cls in KNOWN_CODEBLOCKS,
393 (self.filename, self.get_loc(node)),
394 'Unknown or missing code block type {0}',
397 def check_defined_link_references(self):
398 """Check that defined links resolve in the file.
400 Internally-defined links match the pattern [text][label].
404 for node in self.find_all(self.doc, {'type': 'text'}):
405 for match in P_INTERNAL_LINK_REF.findall(node['value']):
408 if link not in self.args.references:
409 result.add('"{0}"=>"{1}"'.format(text, link))
410 self.reporter.check(not result,
412 'Internally-defined links may be missing definitions: {0}',
413 ', '.join(sorted(result)))
415 def find_all(self, node, pattern, accum=None):
416 """Find all matches for a pattern."""
418 assert isinstance(pattern, dict), 'Patterns must be dictionaries'
421 if self.match(node, pattern):
423 for child in node.get('children', []):
424 self.find_all(child, pattern, accum)
427 def match(self, node, pattern):
428 """Does this node match the given pattern?"""
434 if isinstance(val, str):
437 elif isinstance(val, dict):
438 if not self.match(node[key], val):
443 def get_val(node, *chain):
444 """Get value one or more levels down."""
447 for selector in chain:
448 curr = curr.get(selector, None)
453 def get_loc(self, node):
454 """Convenience method to get node's line number."""
456 result = self.get_val(node, 'options', 'location')
457 if self.metadata_len is not None:
458 result += self.metadata_len
462 class CheckNonJekyll(CheckBase):
463 """Check a file that isn't translated by Jekyll."""
465 def check_metadata(self):
466 self.reporter.check(self.metadata is None,
468 'Unexpected metadata')
471 class CheckIndex(CheckBase):
472 """Check the main index page."""
474 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
475 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
476 self.layout = 'lesson'
478 def check_metadata(self):
479 super().check_metadata()
480 self.reporter.check(self.metadata.get('root', '') == '.',
482 'Root not set to "."')
485 class CheckEpisode(CheckBase):
486 """Check an episode page."""
489 """Run extra tests."""
492 self.check_reference_inclusion()
494 def check_metadata(self):
495 super().check_metadata()
497 if 'layout' in self.metadata:
498 if self.metadata['layout'] == 'break':
499 self.check_metadata_fields(BREAK_METADATA_FIELDS)
501 self.reporter.add(self.filename,
502 'Unknown episode layout "{0}"',
503 self.metadata['layout'])
505 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
507 def check_metadata_fields(self, expected):
508 """Check metadata fields."""
509 for (name, type_) in expected:
510 if name not in self.metadata:
511 self.reporter.add(self.filename,
512 'Missing metadata field {0}',
514 elif not isinstance(self.metadata[name], type_):
515 self.reporter.add(self.filename,
516 '"{0}" has wrong type in metadata ({1} instead of {2})',
517 name, type(self.metadata[name]), type_)
519 def check_reference_inclusion(self):
520 """Check that links file has been included."""
522 if not self.args.reference_path:
525 for (i, last_line, line_len) in reversed(self.lines):
530 'No non-empty lines in {0}'.format(self.filename))
532 include_filename = os.path.split(self.args.reference_path)[-1]
533 if include_filename not in last_line:
534 self.reporter.add(self.filename,
535 'episode does not include "{0}"',
539 class CheckReference(CheckBase):
540 """Check the reference page."""
542 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
543 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
544 self.layout = 'reference'
547 class CheckGeneric(CheckBase):
548 """Check a generic page."""
550 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
551 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
555 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
556 (re.compile(r'README\.md'), CheckNonJekyll),
557 (re.compile(r'index\.md'), CheckIndex),
558 (re.compile(r'reference\.md'), CheckReference),
559 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
560 (re.compile(r'.*\.md'), CheckGeneric)
564 if __name__ == '__main__':