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 check_config(args.reporter, args.source_dir)
121 check_source_rmd(args.reporter, args.source_dir, args.parser)
122 args.references = read_references(args.reporter, args.reference_path)
124 docs = read_all_markdown(args.source_dir, args.parser)
125 check_fileset(args.source_dir, args.reporter, list(docs.keys()))
126 check_unwanted_files(args.source_dir, args.reporter)
127 for filename in list(docs.keys()):
128 checker = create_checker(args, filename, docs[filename])
131 args.reporter.report()
132 if args.reporter.messages and not args.permissive:
137 """Parse command-line arguments."""
139 parser = ArgumentParser(description="""Check episode files in a lesson.""")
140 parser.add_argument('-l', '--linelen',
144 help='Check line lengths')
145 parser.add_argument('-p', '--parser',
148 help='path to Markdown parser')
149 parser.add_argument('-r', '--references',
151 dest='reference_path',
152 help='path to Markdown file of external references')
153 parser.add_argument('-s', '--source',
156 help='source directory')
157 parser.add_argument('-w', '--whitespace',
160 dest='trailing_whitespace',
161 help='Check for trailing whitespace')
162 parser.add_argument('--permissive',
166 help='Do not raise an error even if issues are detected')
168 args, extras = parser.parse_known_args()
169 require(args.parser is not None,
170 'Path to Markdown parser not provided')
172 'Unexpected trailing command-line arguments "{0}"'.format(extras))
177 def check_config(reporter, source_dir):
178 """Check configuration file."""
180 config_file = os.path.join(source_dir, '_config.yml')
181 config = load_yaml(config_file)
182 reporter.check_field(config_file, 'configuration',
183 config, 'kind', 'lesson')
184 reporter.check_field(config_file, 'configuration',
185 config, 'carpentry', ('swc', 'dc', 'lc', 'cp'))
186 reporter.check_field(config_file, 'configuration', config, 'title')
187 reporter.check_field(config_file, 'configuration', config, 'email')
190 {'values': {'root': '.', 'layout': 'page'}},
191 {'values': {'root': '..', 'layout': 'episode'}, 'scope': {'type': 'episodes', 'path': ''}},
192 {'values': {'root': '..', 'layout': 'page'}, 'scope': {'type': 'extras', 'path': ''}}
194 reporter.check(defaults in config.get('defaults', []),
196 '"root" not set to "." in configuration')
198 def check_source_rmd(reporter, source_dir, parser):
199 """Check that Rmd episode files include `source: Rmd`"""
201 episode_rmd_dir = [os.path.join(source_dir, d) for d in SOURCE_RMD_DIRS]
202 episode_rmd_files = [os.path.join(d, '*.Rmd') for d in episode_rmd_dir]
204 for pat in episode_rmd_files:
205 for f in glob.glob(pat):
206 data = read_markdown(parser, f)
207 dy = data['metadata']
209 reporter.check_field(f, 'episode_rmd',
212 def read_references(reporter, ref_path):
213 """Read shared file of reference links, returning dictionary of valid references
214 {symbolic_name : URL}
218 raise Warning("No filename has been provided.")
223 with open(ref_path, 'r', encoding='utf-8') as reader:
224 for (num, line) in enumerate(reader, 1):
226 if P_INTERNAL_INCLUDE_LINK.search(line): continue
228 m = P_INTERNAL_LINK_DEF.search(line)
230 message = '{}: {} not a valid reference: {}'
231 require(m, message.format(ref_path, num, line.rstrip()))
236 message = 'Empty reference at {0}:{1}'
237 require(name, message.format(ref_path, num))
239 unique_name = name not in result
240 unique_url = url not in urls_seen
242 reporter.check(unique_name,
244 'Duplicate reference name {0} at line {1}',
247 reporter.check(unique_url,
249 'Duplicate definition of URL {0} at line {1}',
258 def read_all_markdown(source_dir, parser):
259 """Read source files, returning
260 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
263 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
264 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
266 for pat in all_patterns:
267 for filename in glob.glob(pat):
268 data = read_markdown(parser, filename)
270 result[filename] = data
274 def check_fileset(source_dir, reporter, filenames_present):
275 """Are all required files present? Are extraneous files present?"""
277 # Check files with predictable names.
278 required = [os.path.join(source_dir, p) for p in REQUIRED_FILES]
279 missing = set(required) - set(filenames_present)
281 reporter.add(None, 'Missing required file {0}', m)
283 # Check episode files' names.
285 for filename in filenames_present:
286 if '_episodes' not in filename:
289 # split path to check episode name
290 base_name = os.path.basename(filename)
291 m = P_EPISODE_FILENAME.search(base_name)
293 seen.append(m.group(1))
296 None, 'Episode {0} has badly-formatted filename', filename)
298 # Check for duplicate episode numbers.
299 reporter.check(len(seen) == len(set(seen)),
301 'Duplicate episode numbers {0} vs {1}',
302 sorted(seen), sorted(set(seen)))
304 # Check that numbers are consecutive.
305 seen = sorted([int(s) for s in seen])
307 for i in range(len(seen) - 1):
308 clean = clean and ((seen[i+1] - seen[i]) == 1)
309 reporter.check(clean,
311 'Missing or non-consecutive episode numbers {0}',
315 def create_checker(args, filename, info):
316 """Create appropriate checker for file."""
318 for (pat, cls) in CHECKERS:
319 if pat.search(filename):
320 return cls(args, filename, **info)
321 return NotImplemented
324 """Base class for checking Markdown files."""
326 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
327 """Cache arguments for checking."""
330 self.reporter = self.args.reporter # for convenience
331 self.filename = filename
332 self.metadata = metadata
333 self.metadata_len = metadata_len
343 self.check_metadata()
344 self.check_line_lengths()
345 self.check_trailing_whitespace()
346 self.check_blockquote_classes()
347 self.check_codeblock_classes()
348 self.check_defined_link_references()
350 def check_metadata(self):
351 """Check the YAML metadata."""
353 self.reporter.check(self.metadata is not None,
355 'Missing metadata entirely')
357 if self.metadata and (self.layout is not None):
358 self.reporter.check_field(
359 self.filename, 'metadata', self.metadata, 'layout', self.layout)
361 def check_line_lengths(self):
362 """Check the raw text of the lesson body."""
364 if self.args.line_lengths:
365 over = [i for (i, l, n) in self.lines if (
366 n > MAX_LINE_LEN) and (not l.startswith('!'))]
367 self.reporter.check(not over,
369 'Line(s) too long: {0}',
370 ', '.join([str(i) for i in over]))
372 def check_trailing_whitespace(self):
373 """Check for whitespace at the ends of lines."""
375 if self.args.trailing_whitespace:
377 i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
378 self.reporter.check(not trailing,
380 'Line(s) end with whitespace: {0}',
381 ', '.join([str(i) for i in trailing]))
383 def check_blockquote_classes(self):
384 """Check that all blockquotes have known classes."""
386 for node in self.find_all(self.doc, {'type': 'blockquote'}):
387 cls = self.get_val(node, 'attr', 'class')
388 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
389 (self.filename, self.get_loc(node)),
390 'Unknown or missing blockquote type {0}',
393 def check_codeblock_classes(self):
394 """Check that all code blocks have known classes."""
396 for node in self.find_all(self.doc, {'type': 'codeblock'}):
397 cls = self.get_val(node, 'attr', 'class')
398 self.reporter.check(cls in KNOWN_CODEBLOCKS,
399 (self.filename, self.get_loc(node)),
400 'Unknown or missing code block type {0}',
403 def check_defined_link_references(self):
404 """Check that defined links resolve in the file.
406 Internally-defined links match the pattern [text][label].
410 for node in self.find_all(self.doc, {'type': 'text'}):
411 for match in P_INTERNAL_LINK_REF.findall(node['value']):
414 if link not in self.args.references:
415 result.add('"{0}"=>"{1}"'.format(text, link))
416 self.reporter.check(not result,
418 'Internally-defined links may be missing definitions: {0}',
419 ', '.join(sorted(result)))
421 def find_all(self, node, pattern, accum=None):
422 """Find all matches for a pattern."""
424 assert isinstance(pattern, dict), 'Patterns must be dictionaries'
427 if self.match(node, pattern):
429 for child in node.get('children', []):
430 self.find_all(child, pattern, accum)
433 def match(self, node, pattern):
434 """Does this node match the given pattern?"""
440 if isinstance(val, str):
443 elif isinstance(val, dict):
444 if not self.match(node[key], val):
449 def get_val(node, *chain):
450 """Get value one or more levels down."""
453 for selector in chain:
454 curr = curr.get(selector, None)
459 def get_loc(self, node):
460 """Convenience method to get node's line number."""
462 result = self.get_val(node, 'options', 'location')
463 if self.metadata_len is not None:
464 result += self.metadata_len
468 class CheckNonJekyll(CheckBase):
469 """Check a file that isn't translated by Jekyll."""
471 def check_metadata(self):
472 self.reporter.check(self.metadata is None,
474 'Unexpected metadata')
477 class CheckIndex(CheckBase):
478 """Check the main index page."""
480 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
481 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
482 self.layout = 'lesson'
484 def check_metadata(self):
485 super().check_metadata()
486 self.reporter.check(self.metadata.get('root', '') == '.',
488 'Root not set to "."')
491 class CheckEpisode(CheckBase):
492 """Check an episode page."""
495 """Run extra tests."""
498 self.check_reference_inclusion()
500 def check_metadata(self):
501 super().check_metadata()
503 if 'layout' in self.metadata:
504 if self.metadata['layout'] == 'break':
505 self.check_metadata_fields(BREAK_METADATA_FIELDS)
507 self.reporter.add(self.filename,
508 'Unknown episode layout "{0}"',
509 self.metadata['layout'])
511 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
513 def check_metadata_fields(self, expected):
514 """Check metadata fields."""
515 for (name, type_) in expected:
516 if name not in self.metadata:
517 self.reporter.add(self.filename,
518 'Missing metadata field {0}',
520 elif not isinstance(self.metadata[name], type_):
521 self.reporter.add(self.filename,
522 '"{0}" has wrong type in metadata ({1} instead of {2})',
523 name, type(self.metadata[name]), type_)
525 def check_reference_inclusion(self):
526 """Check that links file has been included."""
528 if not self.args.reference_path:
531 for (i, last_line, line_len) in reversed(self.lines):
536 'No non-empty lines in {0}'.format(self.filename))
538 include_filename = os.path.split(self.args.reference_path)[-1]
539 if include_filename not in last_line:
540 self.reporter.add(self.filename,
541 'episode does not include "{0}"',
545 class CheckReference(CheckBase):
546 """Check the reference page."""
548 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
549 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
550 self.layout = 'reference'
553 class CheckGeneric(CheckBase):
554 """Check a generic page."""
556 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
557 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
561 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
562 (re.compile(r'README\.md'), CheckNonJekyll),
563 (re.compile(r'index\.md'), CheckIndex),
564 (re.compile(r'reference\.md'), CheckReference),
565 (re.compile(os.path.join('_episodes', '*\.md')), CheckEpisode),
566 (re.compile(r'.*\.md'), CheckGeneric)
570 if __name__ == '__main__':