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?
82 # What fields are required in teaching episode metadata?
83 TEACHING_METADATA_FIELDS = {
92 # What fields are required in break episode metadata?
93 BREAK_METADATA_FIELDS = {
99 # How long are lines allowed to be?
100 # Please keep this in sync with .editorconfig!
108 args.reporter = Reporter()
109 life_cycle = check_config(args.reporter, args.source_dir)
110 # pre-alpha lessons should report without error
111 if life_cycle == "pre-alpha":
112 args.permissive = True
113 check_source_rmd(args.reporter, args.source_dir, args.parser)
114 args.references = read_references(args.reporter, args.reference_path)
116 docs = read_all_markdown(args.source_dir, args.parser)
117 check_fileset(args.source_dir, args.reporter, list(docs.keys()))
118 check_unwanted_files(args.source_dir, args.reporter)
119 for filename in list(docs.keys()):
120 checker = create_checker(args, filename, docs[filename])
123 args.reporter.report()
124 if args.reporter.messages and not args.permissive:
129 """Parse command-line arguments."""
131 parser = ArgumentParser(description="""Check episode files in a lesson.""")
132 parser.add_argument('-l', '--linelen',
136 help='Check line lengths')
137 parser.add_argument('-p', '--parser',
140 help='path to Markdown parser')
141 parser.add_argument('-r', '--references',
143 dest='reference_path',
144 help='path to Markdown file of external references')
145 parser.add_argument('-s', '--source',
148 help='source directory')
149 parser.add_argument('-w', '--whitespace',
152 dest='trailing_whitespace',
153 help='Check for trailing whitespace')
154 parser.add_argument('--permissive',
158 help='Do not raise an error even if issues are detected')
160 args, extras = parser.parse_known_args()
161 require(args.parser is not None,
162 'Path to Markdown parser not provided')
164 'Unexpected trailing command-line arguments "{0}"'.format(extras))
169 def check_config(reporter, source_dir):
170 """Check configuration file."""
172 config_file = os.path.join(source_dir, '_config.yml')
173 config = load_yaml(config_file)
174 reporter.check_field(config_file, 'configuration',
175 config, 'kind', 'lesson')
176 reporter.check_field(config_file, 'configuration',
177 config, 'carpentry', ('swc', 'dc', 'lc', 'cp'))
178 reporter.check_field(config_file, 'configuration', config, 'title')
179 reporter.check_field(config_file, 'configuration', config, 'email')
182 {'values': {'root': '.', 'layout': 'page'}},
183 {'values': {'root': '..', 'layout': 'episode'}, 'scope': {'type': 'episodes', 'path': ''}},
184 {'values': {'root': '..', 'layout': 'page'}, 'scope': {'type': 'extras', 'path': ''}}
186 reporter.check(defaults in config.get('defaults', []),
188 '"root" not set to "." in configuration')
189 return config['life_cycle']
191 def check_source_rmd(reporter, source_dir, parser):
192 """Check that Rmd episode files include `source: Rmd`"""
194 episode_rmd_dir = [os.path.join(source_dir, d) for d in SOURCE_RMD_DIRS]
195 episode_rmd_files = [os.path.join(d, '*.Rmd') for d in episode_rmd_dir]
197 for pat in episode_rmd_files:
198 for f in glob.glob(pat):
199 data = read_markdown(parser, f)
200 dy = data['metadata']
202 reporter.check_field(f, 'episode_rmd',
205 def read_references(reporter, ref_path):
206 """Read shared file of reference links, returning dictionary of valid references
207 {symbolic_name : URL}
211 raise Warning("No filename has been provided.")
216 with open(ref_path, 'r', encoding='utf-8') as reader:
217 for (num, line) in enumerate(reader, 1):
219 if P_INTERNAL_INCLUDE_LINK.search(line): continue
221 m = P_INTERNAL_LINK_DEF.search(line)
223 message = '{}: {} not a valid reference: {}'
224 require(m, message.format(ref_path, num, line.rstrip()))
229 message = 'Empty reference at {0}:{1}'
230 require(name, message.format(ref_path, num))
232 unique_name = name not in result
233 unique_url = url not in urls_seen
235 reporter.check(unique_name,
237 'Duplicate reference name {0} at line {1}',
240 reporter.check(unique_url,
242 'Duplicate definition of URL {0} at line {1}',
251 def read_all_markdown(source_dir, parser):
252 """Read source files, returning
253 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
256 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
257 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
259 for pat in all_patterns:
260 for filename in glob.glob(pat):
261 data = read_markdown(parser, filename)
263 result[filename] = data
267 def check_fileset(source_dir, reporter, filenames_present):
268 """Are all required files present? Are extraneous files present?"""
270 # Check files with predictable names.
271 required = [os.path.join(source_dir, p) for p in REQUIRED_FILES]
272 missing = set(required) - set(filenames_present)
274 reporter.add(None, 'Missing required file {0}', m)
276 # Check episode files' names.
278 for filename in filenames_present:
279 if '_episodes' not in filename:
282 # split path to check episode name
283 base_name = os.path.basename(filename)
284 m = P_EPISODE_FILENAME.search(base_name)
286 seen.append(m.group(1))
289 None, 'Episode {0} has badly-formatted filename', filename)
291 # Check for duplicate episode numbers.
292 reporter.check(len(seen) == len(set(seen)),
294 'Duplicate episode numbers {0} vs {1}',
295 sorted(seen), sorted(set(seen)))
297 # Check that numbers are consecutive.
298 seen = sorted([int(s) for s in seen])
300 for i in range(len(seen) - 1):
301 clean = clean and ((seen[i+1] - seen[i]) == 1)
302 reporter.check(clean,
304 'Missing or non-consecutive episode numbers {0}',
308 def create_checker(args, filename, info):
309 """Create appropriate checker for file."""
311 for (pat, cls) in CHECKERS:
312 if pat.search(filename):
313 return cls(args, filename, **info)
314 return NotImplemented
317 """Base class for checking Markdown files."""
319 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
320 """Cache arguments for checking."""
323 self.reporter = self.args.reporter # for convenience
324 self.filename = filename
325 self.metadata = metadata
326 self.metadata_len = metadata_len
336 self.check_metadata()
337 self.check_line_lengths()
338 self.check_trailing_whitespace()
339 self.check_blockquote_classes()
340 self.check_codeblock_classes()
341 self.check_defined_link_references()
343 def check_metadata(self):
344 """Check the YAML metadata."""
346 self.reporter.check(self.metadata is not None,
348 'Missing metadata entirely')
350 if self.metadata and (self.layout is not None):
351 self.reporter.check_field(
352 self.filename, 'metadata', self.metadata, 'layout', self.layout)
354 def check_line_lengths(self):
355 """Check the raw text of the lesson body."""
357 if self.args.line_lengths:
358 over = [i for (i, l, n) in self.lines if (
359 n > MAX_LINE_LEN) and (not l.startswith('!'))]
360 self.reporter.check(not over,
362 'Line(s) too long: {0}',
363 ', '.join([str(i) for i in over]))
365 def check_trailing_whitespace(self):
366 """Check for whitespace at the ends of lines."""
368 if self.args.trailing_whitespace:
370 i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
371 self.reporter.check(not trailing,
373 'Line(s) end with whitespace: {0}',
374 ', '.join([str(i) for i in trailing]))
376 def check_blockquote_classes(self):
377 """Check that all blockquotes have known classes."""
379 for node in self.find_all(self.doc, {'type': 'blockquote'}):
380 cls = self.get_val(node, 'attr', 'class')
381 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
382 (self.filename, self.get_loc(node)),
383 'Unknown or missing blockquote type {0}',
386 def check_codeblock_classes(self):
387 """Check that all code blocks have known classes."""
389 for node in self.find_all(self.doc, {'type': 'codeblock'}):
390 cls = self.get_val(node, 'attr', 'class')
391 self.reporter.check(cls in KNOWN_CODEBLOCKS or cls.startswith('language-'),
392 (self.filename, self.get_loc(node)),
393 'Unknown or missing code block type {0}',
396 def check_defined_link_references(self):
397 """Check that defined links resolve in the file.
399 Internally-defined links match the pattern [text][label].
403 for node in self.find_all(self.doc, {'type': 'text'}):
404 for match in P_INTERNAL_LINK_REF.findall(node['value']):
407 if link not in self.args.references:
408 result.add('"{0}"=>"{1}"'.format(text, link))
409 self.reporter.check(not result,
411 'Internally-defined links may be missing definitions: {0}',
412 ', '.join(sorted(result)))
414 def find_all(self, node, pattern, accum=None):
415 """Find all matches for a pattern."""
417 assert isinstance(pattern, dict), 'Patterns must be dictionaries'
420 if self.match(node, pattern):
422 for child in node.get('children', []):
423 self.find_all(child, pattern, accum)
426 def match(self, node, pattern):
427 """Does this node match the given pattern?"""
433 if isinstance(val, str):
436 elif isinstance(val, dict):
437 if not self.match(node[key], val):
442 def get_val(node, *chain):
443 """Get value one or more levels down."""
446 for selector in chain:
447 curr = curr.get(selector, None)
452 def get_loc(self, node):
453 """Convenience method to get node's line number."""
455 result = self.get_val(node, 'options', 'location')
456 if self.metadata_len is not None:
457 result += self.metadata_len
461 class CheckNonJekyll(CheckBase):
462 """Check a file that isn't translated by Jekyll."""
464 def check_metadata(self):
465 self.reporter.check(self.metadata is None,
467 'Unexpected metadata')
470 class CheckIndex(CheckBase):
471 """Check the main index page."""
473 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
474 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
475 self.layout = 'lesson'
477 def check_metadata(self):
478 super().check_metadata()
479 self.reporter.check(self.metadata.get('root', '') == '.',
481 'Root not set to "."')
484 class CheckEpisode(CheckBase):
485 """Check an episode page."""
488 """Run extra tests."""
491 self.check_reference_inclusion()
493 def check_metadata(self):
494 super().check_metadata()
496 if 'layout' in self.metadata:
497 if self.metadata['layout'] == 'break':
498 self.check_metadata_fields(BREAK_METADATA_FIELDS)
500 self.reporter.add(self.filename,
501 'Unknown episode layout "{0}"',
502 self.metadata['layout'])
504 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
506 def check_metadata_fields(self, expected):
507 """Check metadata fields."""
508 for (name, type_) in expected:
509 if name not in self.metadata:
510 self.reporter.add(self.filename,
511 'Missing metadata field {0}',
513 elif not isinstance(self.metadata[name], type_):
514 self.reporter.add(self.filename,
515 '"{0}" has wrong type in metadata ({1} instead of {2})',
516 name, type(self.metadata[name]), type_)
518 def check_reference_inclusion(self):
519 """Check that links file has been included."""
521 if not self.args.reference_path:
524 for (i, last_line, line_len) in reversed(self.lines):
529 'No non-empty lines in {0}'.format(self.filename))
531 include_filename = os.path.split(self.args.reference_path)[-1]
532 if include_filename not in last_line:
533 self.reporter.add(self.filename,
534 'episode does not include "{0}"',
538 class CheckReference(CheckBase):
539 """Check the reference page."""
541 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
542 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
543 self.layout = 'reference'
546 class CheckGeneric(CheckBase):
547 """Check a generic page."""
549 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
550 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
554 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
555 (re.compile(r'README\.md'), CheckNonJekyll),
556 (re.compile(r'index\.md'), CheckIndex),
557 (re.compile(r'reference\.md'), CheckReference),
558 (re.compile(os.path.join('_episodes', '*\.md')), CheckEpisode),
559 (re.compile(r'.*\.md'), CheckGeneric)
563 if __name__ == '__main__':