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?
74 # What kinds of code fragments are allowed?
93 # What fields are required in teaching episode metadata?
94 TEACHING_METADATA_FIELDS = {
103 # What fields are required in break episode metadata?
104 BREAK_METADATA_FIELDS = {
110 # How long are lines allowed to be?
111 # Please keep this in sync with .editorconfig!
119 args.reporter = Reporter()
120 life_cycle = check_config(args.reporter, args.source_dir)
121 # pre-alpha lessons should report without error
122 if life_cycle == "pre-alpha":
123 args.permissive = True
124 check_source_rmd(args.reporter, args.source_dir, args.parser)
125 args.references = read_references(args.reporter, args.reference_path)
127 docs = read_all_markdown(args.source_dir, args.parser)
128 check_fileset(args.source_dir, args.reporter, list(docs.keys()))
129 check_unwanted_files(args.source_dir, args.reporter)
130 for filename in list(docs.keys()):
131 checker = create_checker(args, filename, docs[filename])
134 args.reporter.report()
135 if args.reporter.messages and not args.permissive:
140 """Parse command-line arguments."""
142 parser = ArgumentParser(description="""Check episode files in a lesson.""")
143 parser.add_argument('-l', '--linelen',
147 help='Check line lengths')
148 parser.add_argument('-p', '--parser',
151 help='path to Markdown parser')
152 parser.add_argument('-r', '--references',
154 dest='reference_path',
155 help='path to Markdown file of external references')
156 parser.add_argument('-s', '--source',
159 help='source directory')
160 parser.add_argument('-w', '--whitespace',
163 dest='trailing_whitespace',
164 help='Check for trailing whitespace')
165 parser.add_argument('--permissive',
169 help='Do not raise an error even if issues are detected')
171 args, extras = parser.parse_known_args()
172 require(args.parser is not None,
173 'Path to Markdown parser not provided')
175 'Unexpected trailing command-line arguments "{0}"'.format(extras))
180 def check_config(reporter, source_dir):
181 """Check configuration file."""
183 config_file = os.path.join(source_dir, '_config.yml')
184 config = load_yaml(config_file)
185 reporter.check_field(config_file, 'configuration',
186 config, 'kind', 'lesson')
187 reporter.check_field(config_file, 'configuration',
188 config, 'carpentry', ('swc', 'dc', 'lc', 'cp'))
189 reporter.check_field(config_file, 'configuration', config, 'title')
190 reporter.check_field(config_file, 'configuration', config, 'email')
193 {'values': {'root': '.', 'layout': 'page'}},
194 {'values': {'root': '..', 'layout': 'episode'}, 'scope': {'type': 'episodes', 'path': ''}},
195 {'values': {'root': '..', 'layout': 'page'}, 'scope': {'type': 'extras', 'path': ''}}
197 reporter.check(defaults in config.get('defaults', []),
199 '"root" not set to "." in configuration')
200 return config['life_cycle']
202 def check_source_rmd(reporter, source_dir, parser):
203 """Check that Rmd episode files include `source: Rmd`"""
205 episode_rmd_dir = [os.path.join(source_dir, d) for d in SOURCE_RMD_DIRS]
206 episode_rmd_files = [os.path.join(d, '*.Rmd') for d in episode_rmd_dir]
208 for pat in episode_rmd_files:
209 for f in glob.glob(pat):
210 data = read_markdown(parser, f)
211 dy = data['metadata']
213 reporter.check_field(f, 'episode_rmd',
216 def read_references(reporter, ref_path):
217 """Read shared file of reference links, returning dictionary of valid references
218 {symbolic_name : URL}
222 raise Warning("No filename has been provided.")
227 with open(ref_path, 'r', encoding='utf-8') as reader:
228 for (num, line) in enumerate(reader, 1):
230 if P_INTERNAL_INCLUDE_LINK.search(line): continue
232 m = P_INTERNAL_LINK_DEF.search(line)
234 message = '{}: {} not a valid reference: {}'
235 require(m, message.format(ref_path, num, line.rstrip()))
240 message = 'Empty reference at {0}:{1}'
241 require(name, message.format(ref_path, num))
243 unique_name = name not in result
244 unique_url = url not in urls_seen
246 reporter.check(unique_name,
248 'Duplicate reference name {0} at line {1}',
251 reporter.check(unique_url,
253 'Duplicate definition of URL {0} at line {1}',
262 def read_all_markdown(source_dir, parser):
263 """Read source files, returning
264 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
267 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
268 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
270 for pat in all_patterns:
271 for filename in glob.glob(pat):
272 data = read_markdown(parser, filename)
274 result[filename] = data
278 def check_fileset(source_dir, reporter, filenames_present):
279 """Are all required files present? Are extraneous files present?"""
281 # Check files with predictable names.
282 required = [os.path.join(source_dir, p) for p in REQUIRED_FILES]
283 missing = set(required) - set(filenames_present)
285 reporter.add(None, 'Missing required file {0}', m)
287 # Check episode files' names.
289 for filename in filenames_present:
290 if '_episodes' not in filename:
293 # split path to check episode name
294 base_name = os.path.basename(filename)
295 m = P_EPISODE_FILENAME.search(base_name)
297 seen.append(m.group(1))
300 None, 'Episode {0} has badly-formatted filename', filename)
302 # Check for duplicate episode numbers.
303 reporter.check(len(seen) == len(set(seen)),
305 'Duplicate episode numbers {0} vs {1}',
306 sorted(seen), sorted(set(seen)))
308 # Check that numbers are consecutive.
309 seen = sorted([int(s) for s in seen])
311 for i in range(len(seen) - 1):
312 clean = clean and ((seen[i+1] - seen[i]) == 1)
313 reporter.check(clean,
315 'Missing or non-consecutive episode numbers {0}',
319 def create_checker(args, filename, info):
320 """Create appropriate checker for file."""
322 for (pat, cls) in CHECKERS:
323 if pat.search(filename):
324 return cls(args, filename, **info)
325 return NotImplemented
328 """Base class for checking Markdown files."""
330 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
331 """Cache arguments for checking."""
334 self.reporter = self.args.reporter # for convenience
335 self.filename = filename
336 self.metadata = metadata
337 self.metadata_len = metadata_len
347 self.check_metadata()
348 self.check_line_lengths()
349 self.check_trailing_whitespace()
350 self.check_blockquote_classes()
351 self.check_codeblock_classes()
352 self.check_defined_link_references()
354 def check_metadata(self):
355 """Check the YAML metadata."""
357 self.reporter.check(self.metadata is not None,
359 'Missing metadata entirely')
361 if self.metadata and (self.layout is not None):
362 self.reporter.check_field(
363 self.filename, 'metadata', self.metadata, 'layout', self.layout)
365 def check_line_lengths(self):
366 """Check the raw text of the lesson body."""
368 if self.args.line_lengths:
369 over = [i for (i, l, n) in self.lines if (
370 n > MAX_LINE_LEN) and (not l.startswith('!'))]
371 self.reporter.check(not over,
373 'Line(s) too long: {0}',
374 ', '.join([str(i) for i in over]))
376 def check_trailing_whitespace(self):
377 """Check for whitespace at the ends of lines."""
379 if self.args.trailing_whitespace:
381 i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
382 self.reporter.check(not trailing,
384 'Line(s) end with whitespace: {0}',
385 ', '.join([str(i) for i in trailing]))
387 def check_blockquote_classes(self):
388 """Check that all blockquotes have known classes."""
390 for node in self.find_all(self.doc, {'type': 'blockquote'}):
391 cls = self.get_val(node, 'attr', 'class')
392 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
393 (self.filename, self.get_loc(node)),
394 'Unknown or missing blockquote type {0}',
397 def check_codeblock_classes(self):
398 """Check that all code blocks have known classes."""
400 for node in self.find_all(self.doc, {'type': 'codeblock'}):
401 cls = self.get_val(node, 'attr', 'class')
402 self.reporter.check(cls in KNOWN_CODEBLOCKS,
403 (self.filename, self.get_loc(node)),
404 'Unknown or missing code block type {0}',
407 def check_defined_link_references(self):
408 """Check that defined links resolve in the file.
410 Internally-defined links match the pattern [text][label].
414 for node in self.find_all(self.doc, {'type': 'text'}):
415 for match in P_INTERNAL_LINK_REF.findall(node['value']):
418 if link not in self.args.references:
419 result.add('"{0}"=>"{1}"'.format(text, link))
420 self.reporter.check(not result,
422 'Internally-defined links may be missing definitions: {0}',
423 ', '.join(sorted(result)))
425 def find_all(self, node, pattern, accum=None):
426 """Find all matches for a pattern."""
428 assert isinstance(pattern, dict), 'Patterns must be dictionaries'
431 if self.match(node, pattern):
433 for child in node.get('children', []):
434 self.find_all(child, pattern, accum)
437 def match(self, node, pattern):
438 """Does this node match the given pattern?"""
444 if isinstance(val, str):
447 elif isinstance(val, dict):
448 if not self.match(node[key], val):
453 def get_val(node, *chain):
454 """Get value one or more levels down."""
457 for selector in chain:
458 curr = curr.get(selector, None)
463 def get_loc(self, node):
464 """Convenience method to get node's line number."""
466 result = self.get_val(node, 'options', 'location')
467 if self.metadata_len is not None:
468 result += self.metadata_len
472 class CheckNonJekyll(CheckBase):
473 """Check a file that isn't translated by Jekyll."""
475 def check_metadata(self):
476 self.reporter.check(self.metadata is None,
478 'Unexpected metadata')
481 class CheckIndex(CheckBase):
482 """Check the main index page."""
484 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
485 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
486 self.layout = 'lesson'
488 def check_metadata(self):
489 super().check_metadata()
490 self.reporter.check(self.metadata.get('root', '') == '.',
492 'Root not set to "."')
495 class CheckEpisode(CheckBase):
496 """Check an episode page."""
499 """Run extra tests."""
502 self.check_reference_inclusion()
504 def check_metadata(self):
505 super().check_metadata()
507 if 'layout' in self.metadata:
508 if self.metadata['layout'] == 'break':
509 self.check_metadata_fields(BREAK_METADATA_FIELDS)
511 self.reporter.add(self.filename,
512 'Unknown episode layout "{0}"',
513 self.metadata['layout'])
515 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
517 def check_metadata_fields(self, expected):
518 """Check metadata fields."""
519 for (name, type_) in expected:
520 if name not in self.metadata:
521 self.reporter.add(self.filename,
522 'Missing metadata field {0}',
524 elif not isinstance(self.metadata[name], type_):
525 self.reporter.add(self.filename,
526 '"{0}" has wrong type in metadata ({1} instead of {2})',
527 name, type(self.metadata[name]), type_)
529 def check_reference_inclusion(self):
530 """Check that links file has been included."""
532 if not self.args.reference_path:
535 for (i, last_line, line_len) in reversed(self.lines):
540 'No non-empty lines in {0}'.format(self.filename))
542 include_filename = os.path.split(self.args.reference_path)[-1]
543 if include_filename not in last_line:
544 self.reporter.add(self.filename,
545 'episode does not include "{0}"',
549 class CheckReference(CheckBase):
550 """Check the reference page."""
552 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
553 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
554 self.layout = 'reference'
557 class CheckGeneric(CheckBase):
558 """Check a generic page."""
560 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
561 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
565 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
566 (re.compile(r'README\.md'), CheckNonJekyll),
567 (re.compile(r'index\.md'), CheckIndex),
568 (re.compile(r'reference\.md'), CheckReference),
569 (re.compile(os.path.join('_episodes', '*\.md')), CheckEpisode),
570 (re.compile(r'.*\.md'), CheckGeneric)
574 if __name__ == '__main__':