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 os.path.join('_extras', 'discuss.md'): True,
37 os.path.join('_extras', 'guide.md'): True,
43 # Episode filename pattern.
44 P_EPISODE_FILENAME = re.compile(r'(\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)
120 docs = read_all_markdown(args.source_dir, args.parser)
121 check_fileset(args.source_dir, args.reporter, list(docs.keys()))
122 check_unwanted_files(args.source_dir, args.reporter)
123 for filename in list(docs.keys()):
124 checker = create_checker(args, filename, docs[filename])
127 args.reporter.report()
128 if args.reporter.messages and not args.permissive:
133 """Parse command-line arguments."""
135 parser = ArgumentParser(description="""Check episode files in a lesson.""")
136 parser.add_argument('-l', '--linelen',
140 help='Check line lengths')
141 parser.add_argument('-p', '--parser',
144 help='path to Markdown parser')
145 parser.add_argument('-r', '--references',
147 dest='reference_path',
148 help='path to Markdown file of external references')
149 parser.add_argument('-s', '--source',
152 help='source directory')
153 parser.add_argument('-w', '--whitespace',
156 dest='trailing_whitespace',
157 help='Check for trailing whitespace')
158 parser.add_argument('--permissive',
162 help='Do not raise an error even if issues are detected')
164 args, extras = parser.parse_known_args()
165 require(args.parser is not None,
166 'Path to Markdown parser not provided')
168 'Unexpected trailing command-line arguments "{0}"'.format(extras))
173 def check_config(reporter, source_dir):
174 """Check configuration file."""
176 config_file = os.path.join(source_dir, '_config.yml')
177 config = load_yaml(config_file)
178 reporter.check_field(config_file, 'configuration',
179 config, 'kind', 'lesson')
180 reporter.check_field(config_file, 'configuration',
181 config, 'carpentry', ('swc', 'dc', 'lc', 'cp'))
182 reporter.check_field(config_file, 'configuration', config, 'title')
183 reporter.check_field(config_file, 'configuration', config, 'email')
186 {'values': {'root': '.', 'layout': 'page'}},
187 {'values': {'root': '..', 'layout': 'episode'}, 'scope': {'type': 'episodes', 'path': ''}},
188 {'values': {'root': '..', 'layout': 'page'}, 'scope': {'type': 'extras', 'path': ''}}
190 reporter.check(defaults in config.get('defaults', []),
192 '"root" not set to "." in configuration')
194 def check_source_rmd(reporter, source_dir, parser):
195 """Check that Rmd episode files include `source: Rmd`"""
197 episode_rmd_dir = [os.path.join(source_dir, d) for d in SOURCE_RMD_DIRS]
198 episode_rmd_files = [os.path.join(d, '*.Rmd') for d in episode_rmd_dir]
200 for pat in episode_rmd_files:
201 for f in glob.glob(pat):
202 data = read_markdown(parser, f)
203 dy = data['metadata']
205 reporter.check_field(f, 'episode_rmd',
208 def read_references(reporter, ref_path):
209 """Read shared file of reference links, returning dictionary of valid references
210 {symbolic_name : URL}
214 raise Warning("No filename has been provided.")
219 with open(ref_path, 'r') as reader:
220 for (num, line) in enumerate(reader, 1):
222 if P_INTERNAL_INCLUDE_LINK.search(line): continue
224 m = P_INTERNAL_LINK_DEF.search(line)
226 message = '{}: {} not a valid reference: {}'
227 require(m, message.format(ref_path, num, line.rstrip()))
232 message = 'Empty reference at {0}:{1}'
233 require(name, message.format(ref_path, num))
235 unique_name = name not in result
236 unique_url = url not in urls_seen
238 reporter.check(unique_name,
240 'Duplicate reference name {0} at line {1}',
243 reporter.check(unique_url,
245 'Duplicate definition of URL {0} at line {1}',
254 def read_all_markdown(source_dir, parser):
255 """Read source files, returning
256 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
260 for d in SOURCE_DIRS:
261 dpath = os.path.join(source_dir, d)
263 pattern = os.path.join(dpath, '*.md')
264 for filename in glob.glob(pattern):
265 data = read_markdown(parser, filename)
267 result[filename] = data
271 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
273 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
276 for pat in all_patterns:
278 for filename in glob.glob(pat):
279 data = read_markdown(parser, filename)
282 result[filename] = data
286 def check_fileset(source_dir, reporter, filenames_present):
287 """Are all required files present? Are extraneous files present?"""
289 # Check files with predictable names.
290 required = [os.path.join(source_dir, p) for p in REQUIRED_FILES]
291 missing = set(required) - set(filenames_present)
293 reporter.add(None, 'Missing required file {0}', m)
295 # Check episode files' names.
297 for filepath in filenames_present:
298 if '_episodes' not in filepath:
301 # split path to check episode name
302 fname = os.path.basename(filepath)
303 m = P_EPISODE_FILENAME.search(fname)
305 seen.append(m.group(1))
308 None, 'Episode {0} has badly-formatted filename', filepath)
310 # Check for duplicate episode numbers.
311 reporter.check(len(seen) == len(set(seen)),
313 'Duplicate episode numbers {0} vs {1}',
314 sorted(seen), sorted(set(seen)))
316 # Check that numbers are consecutive.
317 seen = sorted([int(s) for s in seen])
319 for i in range(len(seen) - 1):
320 clean = clean and ((seen[i+1] - seen[i]) == 1)
321 reporter.check(clean,
323 'Missing or non-consecutive episode numbers {0}',
327 def create_checker(args, filename, info):
328 """Create appropriate checker for file."""
330 for (pat, cls) in CHECKERS:
331 if pat.search(filename):
332 return cls(args, filename, **info)
333 return NotImplemented
336 """Base class for checking Markdown files."""
338 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
339 """Cache arguments for checking."""
342 self.reporter = self.args.reporter # for convenience
343 self.filename = filename
344 self.metadata = metadata
345 self.metadata_len = metadata_len
355 self.check_metadata()
356 self.check_line_lengths()
357 self.check_trailing_whitespace()
358 self.check_blockquote_classes()
359 self.check_codeblock_classes()
360 self.check_defined_link_references()
362 def check_metadata(self):
363 """Check the YAML metadata."""
365 self.reporter.check(self.metadata is not None,
367 'Missing metadata entirely')
369 if self.metadata and (self.layout is not None):
370 self.reporter.check_field(
371 self.filename, 'metadata', self.metadata, 'layout', self.layout)
373 def check_line_lengths(self):
374 """Check the raw text of the lesson body."""
376 if self.args.line_lengths:
377 over = [i for (i, l, n) in self.lines if (
378 n > MAX_LINE_LEN) and (not l.startswith('!'))]
379 self.reporter.check(not over,
381 'Line(s) too long: {0}',
382 ', '.join([str(i) for i in over]))
384 def check_trailing_whitespace(self):
385 """Check for whitespace at the ends of lines."""
387 if self.args.trailing_whitespace:
389 i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
390 self.reporter.check(not trailing,
392 'Line(s) end with whitespace: {0}',
393 ', '.join([str(i) for i in trailing]))
395 def check_blockquote_classes(self):
396 """Check that all blockquotes have known classes."""
398 for node in self.find_all(self.doc, {'type': 'blockquote'}):
399 cls = self.get_val(node, 'attr', 'class')
400 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
401 (self.filename, self.get_loc(node)),
402 'Unknown or missing blockquote type {0}',
405 def check_codeblock_classes(self):
406 """Check that all code blocks have known classes."""
408 for node in self.find_all(self.doc, {'type': 'codeblock'}):
409 cls = self.get_val(node, 'attr', 'class')
410 self.reporter.check(cls in KNOWN_CODEBLOCKS,
411 (self.filename, self.get_loc(node)),
412 'Unknown or missing code block type {0}',
415 def check_defined_link_references(self):
416 """Check that defined links resolve in the file.
418 Internally-defined links match the pattern [text][label].
422 for node in self.find_all(self.doc, {'type': 'text'}):
423 for match in P_INTERNAL_LINK_REF.findall(node['value']):
426 if link not in self.args.references:
427 result.add('"{0}"=>"{1}"'.format(text, link))
428 self.reporter.check(not result,
430 'Internally-defined links may be missing definitions: {0}',
431 ', '.join(sorted(result)))
433 def find_all(self, node, pattern, accum=None):
434 """Find all matches for a pattern."""
436 assert isinstance(pattern, dict), 'Patterns must be dictionaries'
439 if self.match(node, pattern):
441 for child in node.get('children', []):
442 self.find_all(child, pattern, accum)
445 def match(self, node, pattern):
446 """Does this node match the given pattern?"""
452 if isinstance(val, str):
455 elif isinstance(val, dict):
456 if not self.match(node[key], val):
461 def get_val(node, *chain):
462 """Get value one or more levels down."""
465 for selector in chain:
466 curr = curr.get(selector, None)
471 def get_loc(self, node):
472 """Convenience method to get node's line number."""
474 result = self.get_val(node, 'options', 'location')
475 if self.metadata_len is not None:
476 result += self.metadata_len
480 class CheckNonJekyll(CheckBase):
481 """Check a file that isn't translated by Jekyll."""
483 def check_metadata(self):
484 self.reporter.check(self.metadata is None,
486 'Unexpected metadata')
489 class CheckIndex(CheckBase):
490 """Check the main index page."""
492 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
493 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
494 self.layout = 'lesson'
496 def check_metadata(self):
497 super().check_metadata()
498 self.reporter.check(self.metadata.get('root', '') == '.',
500 'Root not set to "."')
503 class CheckEpisode(CheckBase):
504 """Check an episode page."""
507 """Run extra tests."""
510 self.check_reference_inclusion()
512 def check_metadata(self):
513 super().check_metadata()
515 if 'layout' in self.metadata:
516 if self.metadata['layout'] == 'break':
517 self.check_metadata_fields(BREAK_METADATA_FIELDS)
519 self.reporter.add(self.filename,
520 'Unknown episode layout "{0}"',
521 self.metadata['layout'])
523 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
525 def check_metadata_fields(self, expected):
526 """Check metadata fields."""
527 for (name, type_) in expected:
528 if name not in self.metadata:
529 self.reporter.add(self.filename,
530 'Missing metadata field {0}',
532 elif not isinstance(self.metadata[name], type_):
533 self.reporter.add(self.filename,
534 '"{0}" has wrong type in metadata ({1} instead of {2})',
535 name, type(self.metadata[name]), type_)
537 def check_reference_inclusion(self):
538 """Check that links file has been included."""
540 if not self.args.reference_path:
543 for (i, last_line, line_len) in reversed(self.lines):
548 'No non-empty lines in {0}'.format(self.filename))
550 include_filename = os.path.split(self.args.reference_path)[-1]
551 if include_filename not in last_line:
552 self.reporter.add(self.filename,
553 'episode does not include "{0}"',
557 class CheckReference(CheckBase):
558 """Check the reference page."""
560 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
561 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
562 self.layout = 'reference'
565 class CheckGeneric(CheckBase):
566 """Check a generic page."""
568 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
569 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
573 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
574 (re.compile(r'README\.md'), CheckNonJekyll),
575 (re.compile(r'index\.md'), CheckIndex),
576 (re.compile(r'reference\.md'), CheckReference),
577 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
578 (re.compile(r'.*\.md'), CheckGeneric)
582 if __name__ == '__main__':