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?
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}
213 raise Warning("No filename has been provided.")
218 with open(ref_path, 'r') as reader:
219 for (num, line) in enumerate(reader, 1):
221 if P_INTERNAL_INCLUDE_LINK.search(line): continue
223 m = P_INTERNAL_LINK_DEF.search(line)
225 message = '{}: {} not a valid reference: {}'
226 require(m, message.format(ref_path, num, line.rstrip()))
231 message = 'Empty reference at {0}:{1}'
232 require(name, message.format(ref_path, num))
234 unique_name = name not in result
235 unique_url = url not in urls_seen
237 reporter.check(unique_name,
239 'Duplicate reference name {0} at line {1}',
242 reporter.check(unique_url,
244 'Duplicate definition of URL {0} at line {1}',
253 def read_all_markdown(source_dir, parser):
254 """Read source files, returning
255 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
259 for d in SOURCE_DIRS:
260 dpath = os.path.join(source_dir, d)
262 pattern = os.path.join(dpath, '*.md')
263 for filename in glob.glob(pattern):
264 data = read_markdown(parser, filename)
266 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 = [os.path.join(source_dir, p) 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:
286 # split path to check episode name
287 base_name = os.path.basename(filename)
288 m = P_EPISODE_FILENAME.search(base_name)
290 seen.append(m.group(1))
293 None, 'Episode {0} has badly-formatted filename', filename)
295 # Check for duplicate episode numbers.
296 reporter.check(len(seen) == len(set(seen)),
298 'Duplicate episode numbers {0} vs {1}',
299 sorted(seen), sorted(set(seen)))
301 # Check that numbers are consecutive.
302 seen = sorted([int(s) for s in seen])
304 for i in range(len(seen) - 1):
305 clean = clean and ((seen[i+1] - seen[i]) == 1)
306 reporter.check(clean,
308 'Missing or non-consecutive episode numbers {0}',
312 def create_checker(args, filename, info):
313 """Create appropriate checker for file."""
315 for (pat, cls) in CHECKERS:
316 if pat.search(filename):
317 return cls(args, filename, **info)
318 return NotImplemented
321 """Base class for checking Markdown files."""
323 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
324 """Cache arguments for checking."""
327 self.reporter = self.args.reporter # for convenience
328 self.filename = filename
329 self.metadata = metadata
330 self.metadata_len = metadata_len
340 self.check_metadata()
341 self.check_line_lengths()
342 self.check_trailing_whitespace()
343 self.check_blockquote_classes()
344 self.check_codeblock_classes()
345 self.check_defined_link_references()
347 def check_metadata(self):
348 """Check the YAML metadata."""
350 self.reporter.check(self.metadata is not None,
352 'Missing metadata entirely')
354 if self.metadata and (self.layout is not None):
355 self.reporter.check_field(
356 self.filename, 'metadata', self.metadata, 'layout', self.layout)
358 def check_line_lengths(self):
359 """Check the raw text of the lesson body."""
361 if self.args.line_lengths:
362 over = [i for (i, l, n) in self.lines if (
363 n > MAX_LINE_LEN) and (not l.startswith('!'))]
364 self.reporter.check(not over,
366 'Line(s) too long: {0}',
367 ', '.join([str(i) for i in over]))
369 def check_trailing_whitespace(self):
370 """Check for whitespace at the ends of lines."""
372 if self.args.trailing_whitespace:
374 i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
375 self.reporter.check(not trailing,
377 'Line(s) end with whitespace: {0}',
378 ', '.join([str(i) for i in trailing]))
380 def check_blockquote_classes(self):
381 """Check that all blockquotes have known classes."""
383 for node in self.find_all(self.doc, {'type': 'blockquote'}):
384 cls = self.get_val(node, 'attr', 'class')
385 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
386 (self.filename, self.get_loc(node)),
387 'Unknown or missing blockquote type {0}',
390 def check_codeblock_classes(self):
391 """Check that all code blocks have known classes."""
393 for node in self.find_all(self.doc, {'type': 'codeblock'}):
394 cls = self.get_val(node, 'attr', 'class')
395 self.reporter.check(cls in KNOWN_CODEBLOCKS,
396 (self.filename, self.get_loc(node)),
397 'Unknown or missing code block type {0}',
400 def check_defined_link_references(self):
401 """Check that defined links resolve in the file.
403 Internally-defined links match the pattern [text][label].
407 for node in self.find_all(self.doc, {'type': 'text'}):
408 for match in P_INTERNAL_LINK_REF.findall(node['value']):
411 if link not in self.args.references:
412 result.add('"{0}"=>"{1}"'.format(text, link))
413 self.reporter.check(not result,
415 'Internally-defined links may be missing definitions: {0}',
416 ', '.join(sorted(result)))
418 def find_all(self, node, pattern, accum=None):
419 """Find all matches for a pattern."""
421 assert isinstance(pattern, dict), 'Patterns must be dictionaries'
424 if self.match(node, pattern):
426 for child in node.get('children', []):
427 self.find_all(child, pattern, accum)
430 def match(self, node, pattern):
431 """Does this node match the given pattern?"""
437 if isinstance(val, str):
440 elif isinstance(val, dict):
441 if not self.match(node[key], val):
446 def get_val(node, *chain):
447 """Get value one or more levels down."""
450 for selector in chain:
451 curr = curr.get(selector, None)
456 def get_loc(self, node):
457 """Convenience method to get node's line number."""
459 result = self.get_val(node, 'options', 'location')
460 if self.metadata_len is not None:
461 result += self.metadata_len
465 class CheckNonJekyll(CheckBase):
466 """Check a file that isn't translated by Jekyll."""
468 def check_metadata(self):
469 self.reporter.check(self.metadata is None,
471 'Unexpected metadata')
474 class CheckIndex(CheckBase):
475 """Check the main index page."""
477 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
478 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
479 self.layout = 'lesson'
481 def check_metadata(self):
482 super().check_metadata()
483 self.reporter.check(self.metadata.get('root', '') == '.',
485 'Root not set to "."')
488 class CheckEpisode(CheckBase):
489 """Check an episode page."""
492 """Run extra tests."""
495 self.check_reference_inclusion()
497 def check_metadata(self):
498 super().check_metadata()
500 if 'layout' in self.metadata:
501 if self.metadata['layout'] == 'break':
502 self.check_metadata_fields(BREAK_METADATA_FIELDS)
504 self.reporter.add(self.filename,
505 'Unknown episode layout "{0}"',
506 self.metadata['layout'])
508 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
510 def check_metadata_fields(self, expected):
511 """Check metadata fields."""
512 for (name, type_) in expected:
513 if name not in self.metadata:
514 self.reporter.add(self.filename,
515 'Missing metadata field {0}',
517 elif not isinstance(self.metadata[name], type_):
518 self.reporter.add(self.filename,
519 '"{0}" has wrong type in metadata ({1} instead of {2})',
520 name, type(self.metadata[name]), type_)
522 def check_reference_inclusion(self):
523 """Check that links file has been included."""
525 if not self.args.reference_path:
528 for (i, last_line, line_len) in reversed(self.lines):
533 'No non-empty lines in {0}'.format(self.filename))
535 include_filename = os.path.split(self.args.reference_path)[-1]
536 if include_filename not in last_line:
537 self.reporter.add(self.filename,
538 'episode does not include "{0}"',
542 class CheckReference(CheckBase):
543 """Check the reference page."""
545 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
546 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
547 self.layout = 'reference'
550 class CheckGeneric(CheckBase):
551 """Check a generic page."""
553 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
554 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
558 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
559 (re.compile(r'README\.md'), CheckNonJekyll),
560 (re.compile(r'index\.md'), CheckIndex),
561 (re.compile(r'reference\.md'), CheckReference),
562 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
563 (re.compile(r'.*\.md'), CheckGeneric)
567 if __name__ == '__main__':