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?
92 # What fields are required in teaching episode metadata?
93 TEACHING_METADATA_FIELDS = {
102 # What fields are required in break episode metadata?
103 BREAK_METADATA_FIELDS = {
109 # How long are lines allowed to be?
110 # Please keep this in sync with .editorconfig!
118 args.reporter = Reporter()
119 check_config(args.reporter, args.source_dir)
120 check_source_rmd(args.reporter, args.source_dir, args.parser)
121 args.references = read_references(args.reporter, args.reference_path)
123 docs = read_all_markdown(args.source_dir, args.parser)
124 check_fileset(args.source_dir, args.reporter, list(docs.keys()))
125 check_unwanted_files(args.source_dir, args.reporter)
126 for filename in list(docs.keys()):
127 checker = create_checker(args, filename, docs[filename])
130 args.reporter.report()
131 if args.reporter.messages and not args.permissive:
136 """Parse command-line arguments."""
138 parser = ArgumentParser(description="""Check episode files in a lesson.""")
139 parser.add_argument('-l', '--linelen',
143 help='Check line lengths')
144 parser.add_argument('-p', '--parser',
147 help='path to Markdown parser')
148 parser.add_argument('-r', '--references',
150 dest='reference_path',
151 help='path to Markdown file of external references')
152 parser.add_argument('-s', '--source',
155 help='source directory')
156 parser.add_argument('-w', '--whitespace',
159 dest='trailing_whitespace',
160 help='Check for trailing whitespace')
161 parser.add_argument('--permissive',
165 help='Do not raise an error even if issues are detected')
167 args, extras = parser.parse_known_args()
168 require(args.parser is not None,
169 'Path to Markdown parser not provided')
171 'Unexpected trailing command-line arguments "{0}"'.format(extras))
176 def check_config(reporter, source_dir):
177 """Check configuration file."""
179 config_file = os.path.join(source_dir, '_config.yml')
180 config = load_yaml(config_file)
181 reporter.check_field(config_file, 'configuration',
182 config, 'kind', 'lesson')
183 reporter.check_field(config_file, 'configuration',
184 config, 'carpentry', ('swc', 'dc', 'lc', 'cp'))
185 reporter.check_field(config_file, 'configuration', config, 'title')
186 reporter.check_field(config_file, 'configuration', config, 'email')
189 {'values': {'root': '.', 'layout': 'page'}},
190 {'values': {'root': '..', 'layout': 'episode'}, 'scope': {'type': 'episodes', 'path': ''}},
191 {'values': {'root': '..', 'layout': 'page'}, 'scope': {'type': 'extras', 'path': ''}}
193 reporter.check(defaults in config.get('defaults', []),
195 '"root" not set to "." in configuration')
197 def check_source_rmd(reporter, source_dir, parser):
198 """Check that Rmd episode files include `source: Rmd`"""
200 episode_rmd_dir = [os.path.join(source_dir, d) for d in SOURCE_RMD_DIRS]
201 episode_rmd_files = [os.path.join(d, '*.Rmd') for d in episode_rmd_dir]
203 for pat in episode_rmd_files:
204 for f in glob.glob(pat):
205 data = read_markdown(parser, f)
206 dy = data['metadata']
208 reporter.check_field(f, 'episode_rmd',
211 def read_references(reporter, ref_path):
212 """Read shared file of reference links, returning dictionary of valid references
213 {symbolic_name : URL}
217 raise Warning("No filename has been provided.")
222 with open(ref_path, 'r', encoding='utf-8') as reader:
223 for (num, line) in enumerate(reader, 1):
225 if P_INTERNAL_INCLUDE_LINK.search(line): continue
227 m = P_INTERNAL_LINK_DEF.search(line)
229 message = '{}: {} not a valid reference: {}'
230 require(m, message.format(ref_path, num, line.rstrip()))
235 message = 'Empty reference at {0}:{1}'
236 require(name, message.format(ref_path, num))
238 unique_name = name not in result
239 unique_url = url not in urls_seen
241 reporter.check(unique_name,
243 'Duplicate reference name {0} at line {1}',
246 reporter.check(unique_url,
248 'Duplicate definition of URL {0} at line {1}',
257 def read_all_markdown(source_dir, parser):
258 """Read source files, returning
259 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
262 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
263 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
265 for pat in all_patterns:
266 for filename in glob.glob(pat):
267 data = read_markdown(parser, filename)
269 result[filename] = data
273 def check_fileset(source_dir, reporter, filenames_present):
274 """Are all required files present? Are extraneous files present?"""
276 # Check files with predictable names.
277 required = [os.path.join(source_dir, p) for p in REQUIRED_FILES]
278 missing = set(required) - set(filenames_present)
280 reporter.add(None, 'Missing required file {0}', m)
282 # Check episode files' names.
284 for filename in filenames_present:
285 if '_episodes' not in filename:
288 # split path to check episode name
289 base_name = os.path.basename(filename)
290 m = P_EPISODE_FILENAME.search(base_name)
292 seen.append(m.group(1))
295 None, 'Episode {0} has badly-formatted filename', filename)
297 # Check for duplicate episode numbers.
298 reporter.check(len(seen) == len(set(seen)),
300 'Duplicate episode numbers {0} vs {1}',
301 sorted(seen), sorted(set(seen)))
303 # Check that numbers are consecutive.
304 seen = sorted([int(s) for s in seen])
306 for i in range(len(seen) - 1):
307 clean = clean and ((seen[i+1] - seen[i]) == 1)
308 reporter.check(clean,
310 'Missing or non-consecutive episode numbers {0}',
314 def create_checker(args, filename, info):
315 """Create appropriate checker for file."""
317 for (pat, cls) in CHECKERS:
318 if pat.search(filename):
319 return cls(args, filename, **info)
320 return NotImplemented
323 """Base class for checking Markdown files."""
325 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
326 """Cache arguments for checking."""
329 self.reporter = self.args.reporter # for convenience
330 self.filename = filename
331 self.metadata = metadata
332 self.metadata_len = metadata_len
342 self.check_metadata()
343 self.check_line_lengths()
344 self.check_trailing_whitespace()
345 self.check_blockquote_classes()
346 self.check_codeblock_classes()
347 self.check_defined_link_references()
349 def check_metadata(self):
350 """Check the YAML metadata."""
352 self.reporter.check(self.metadata is not None,
354 'Missing metadata entirely')
356 if self.metadata and (self.layout is not None):
357 self.reporter.check_field(
358 self.filename, 'metadata', self.metadata, 'layout', self.layout)
360 def check_line_lengths(self):
361 """Check the raw text of the lesson body."""
363 if self.args.line_lengths:
364 over = [i for (i, l, n) in self.lines if (
365 n > MAX_LINE_LEN) and (not l.startswith('!'))]
366 self.reporter.check(not over,
368 'Line(s) too long: {0}',
369 ', '.join([str(i) for i in over]))
371 def check_trailing_whitespace(self):
372 """Check for whitespace at the ends of lines."""
374 if self.args.trailing_whitespace:
376 i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
377 self.reporter.check(not trailing,
379 'Line(s) end with whitespace: {0}',
380 ', '.join([str(i) for i in trailing]))
382 def check_blockquote_classes(self):
383 """Check that all blockquotes have known classes."""
385 for node in self.find_all(self.doc, {'type': 'blockquote'}):
386 cls = self.get_val(node, 'attr', 'class')
387 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
388 (self.filename, self.get_loc(node)),
389 'Unknown or missing blockquote type {0}',
392 def check_codeblock_classes(self):
393 """Check that all code blocks have known classes."""
395 for node in self.find_all(self.doc, {'type': 'codeblock'}):
396 cls = self.get_val(node, 'attr', 'class')
397 self.reporter.check(cls in KNOWN_CODEBLOCKS,
398 (self.filename, self.get_loc(node)),
399 'Unknown or missing code block type {0}',
402 def check_defined_link_references(self):
403 """Check that defined links resolve in the file.
405 Internally-defined links match the pattern [text][label].
409 for node in self.find_all(self.doc, {'type': 'text'}):
410 for match in P_INTERNAL_LINK_REF.findall(node['value']):
413 if link not in self.args.references:
414 result.add('"{0}"=>"{1}"'.format(text, link))
415 self.reporter.check(not result,
417 'Internally-defined links may be missing definitions: {0}',
418 ', '.join(sorted(result)))
420 def find_all(self, node, pattern, accum=None):
421 """Find all matches for a pattern."""
423 assert isinstance(pattern, dict), 'Patterns must be dictionaries'
426 if self.match(node, pattern):
428 for child in node.get('children', []):
429 self.find_all(child, pattern, accum)
432 def match(self, node, pattern):
433 """Does this node match the given pattern?"""
439 if isinstance(val, str):
442 elif isinstance(val, dict):
443 if not self.match(node[key], val):
448 def get_val(node, *chain):
449 """Get value one or more levels down."""
452 for selector in chain:
453 curr = curr.get(selector, None)
458 def get_loc(self, node):
459 """Convenience method to get node's line number."""
461 result = self.get_val(node, 'options', 'location')
462 if self.metadata_len is not None:
463 result += self.metadata_len
467 class CheckNonJekyll(CheckBase):
468 """Check a file that isn't translated by Jekyll."""
470 def check_metadata(self):
471 self.reporter.check(self.metadata is None,
473 'Unexpected metadata')
476 class CheckIndex(CheckBase):
477 """Check the main index page."""
479 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
480 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
481 self.layout = 'lesson'
483 def check_metadata(self):
484 super().check_metadata()
485 self.reporter.check(self.metadata.get('root', '') == '.',
487 'Root not set to "."')
490 class CheckEpisode(CheckBase):
491 """Check an episode page."""
494 """Run extra tests."""
497 self.check_reference_inclusion()
499 def check_metadata(self):
500 super().check_metadata()
502 if 'layout' in self.metadata:
503 if self.metadata['layout'] == 'break':
504 self.check_metadata_fields(BREAK_METADATA_FIELDS)
506 self.reporter.add(self.filename,
507 'Unknown episode layout "{0}"',
508 self.metadata['layout'])
510 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
512 def check_metadata_fields(self, expected):
513 """Check metadata fields."""
514 for (name, type_) in expected:
515 if name not in self.metadata:
516 self.reporter.add(self.filename,
517 'Missing metadata field {0}',
519 elif not isinstance(self.metadata[name], type_):
520 self.reporter.add(self.filename,
521 '"{0}" has wrong type in metadata ({1} instead of {2})',
522 name, type(self.metadata[name]), type_)
524 def check_reference_inclusion(self):
525 """Check that links file has been included."""
527 if not self.args.reference_path:
530 for (i, last_line, line_len) in reversed(self.lines):
535 'No non-empty lines in {0}'.format(self.filename))
537 include_filename = os.path.split(self.args.reference_path)[-1]
538 if include_filename not in last_line:
539 self.reporter.add(self.filename,
540 'episode does not include "{0}"',
544 class CheckReference(CheckBase):
545 """Check the reference page."""
547 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
548 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
549 self.layout = 'reference'
552 class CheckGeneric(CheckBase):
553 """Check a generic page."""
555 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
556 super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
560 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
561 (re.compile(r'README\.md'), CheckNonJekyll),
562 (re.compile(r'index\.md'), CheckIndex),
563 (re.compile(r'reference\.md'), CheckReference),
564 (re.compile(os.path.join('_episodes', '*\.md')), CheckEpisode),
565 (re.compile(r'.*\.md'), CheckGeneric)
569 if __name__ == '__main__':