4 Check lesson files and their contents.
11 from optparse import OptionParser
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 # Required files: each entry is ('path': YAML_required).
22 # FIXME: We do not yet validate whether any files have the required
23 # YAML headers, but should in the future.
24 # The '%' is replaced with the source directory path for checking.
25 # Episodes are handled specially, and extra files in '_extras' are also handled
26 # specially. This list must include all the Markdown files listed in the
27 # 'bin/initialize' script.
30 '%/CONTRIBUTING.md': False,
33 '%/_extras/discuss.md': True,
34 '%/_extras/guide.md': True,
36 '%/reference.md': True,
40 # Episode filename pattern.
41 P_EPISODE_FILENAME = re.compile(r'/_episodes/(\d\d)-[-\w]+.md$')
43 # Pattern to match lines ending with whitespace.
44 P_TRAILING_WHITESPACE = re.compile(r'\s+$')
46 # Pattern to match figure references in HTML.
47 P_FIGURE_REFS = re.compile(r'<img[^>]+src="([^"]+)"[^>]*>')
49 # Pattern to match internally-defined Markdown links.
50 P_INTERNAL_LINK_REF = re.compile(r'\[([^\]]+)\]\[([^\]]+)\]')
52 # Pattern to match reference links (to resolve internally-defined references).
53 P_INTERNAL_LINK_DEF = re.compile(r'^\[([^\]]+)\]:\s*(.+)')
55 # What kinds of blockquotes are allowed?
69 # What kinds of code fragments are allowed?
84 # What fields are required in teaching episode metadata?
85 TEACHING_METADATA_FIELDS = {
94 # What fields are required in break episode metadata?
95 BREAK_METADATA_FIELDS = {
101 # How long are lines allowed to be?
109 args.reporter = Reporter()
110 check_config(args.reporter, args.source_dir)
111 args.references = read_references(args.reporter, args.reference_path)
113 docs = read_all_markdown(args.source_dir, args.parser)
114 check_fileset(args.source_dir, args.reporter, list(docs.keys()))
115 check_unwanted_files(args.source_dir, args.reporter)
116 for filename in list(docs.keys()):
117 checker = create_checker(args, filename, docs[filename])
120 args.reporter.report()
124 """Parse command-line arguments."""
126 parser = OptionParser()
127 parser.add_option('-l', '--linelen',
131 help='Check line lengths')
132 parser.add_option('-p', '--parser',
135 help='path to Markdown parser')
136 parser.add_option('-r', '--references',
138 dest='reference_path',
139 help='path to Markdown file of external references')
140 parser.add_option('-s', '--source',
143 help='source directory')
144 parser.add_option('-w', '--whitespace',
147 dest='trailing_whitespace',
148 help='Check for trailing whitespace')
150 args, extras = parser.parse_args()
151 require(args.parser is not None,
152 'Path to Markdown parser not provided')
154 'Unexpected trailing command-line arguments "{0}"'.format(extras))
159 def check_config(reporter, source_dir):
160 """Check configuration file."""
162 config_file = os.path.join(source_dir, '_config.yml')
163 config = load_yaml(config_file)
164 reporter.check_field(config_file, 'configuration',
165 config, 'kind', 'lesson')
166 reporter.check_field(config_file, 'configuration',
167 config, 'carpentry', ('swc', 'dc', 'lc'))
168 reporter.check_field(config_file, 'configuration', config, 'title')
169 reporter.check_field(config_file, 'configuration', config, 'email')
171 reporter.check({'values': {'root': '..'}} in config.get('defaults', []),
173 '"root" not set to ".." in configuration')
176 def read_references(reporter, ref_path):
177 """Read shared file of reference links, returning dictionary of valid references
178 {symbolic_name : URL}
184 with open(ref_path, 'r') as reader:
185 for (num, line) in enumerate(reader):
187 m = P_INTERNAL_LINK_DEF.search(line)
189 '{0}:{1} not valid reference:\n{2}'.format(ref_path, line_num, line.rstrip()))
193 'Empty reference at {0}:{1}'.format(ref_path, line_num))
194 reporter.check(name not in result,
196 'Duplicate reference {0} at line {1}',
198 reporter.check(url not in urls_seen,
200 'Duplicate definition of URL {0} at line {1}',
207 def read_all_markdown(source_dir, parser):
208 """Read source files, returning
209 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
212 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
213 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
215 for pat in all_patterns:
216 for filename in glob.glob(pat):
217 data = read_markdown(parser, filename)
219 result[filename] = data
223 def check_fileset(source_dir, reporter, filenames_present):
224 """Are all required files present? Are extraneous files present?"""
226 # Check files with predictable names.
227 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
228 missing = set(required) - set(filenames_present)
230 reporter.add(None, 'Missing required file {0}', m)
232 # Check episode files' names.
234 for filename in filenames_present:
235 if '_episodes' not in filename:
237 m = P_EPISODE_FILENAME.search(filename)
239 seen.append(m.group(1))
242 None, 'Episode {0} has badly-formatted filename', filename)
244 # Check for duplicate episode numbers.
245 reporter.check(len(seen) == len(set(seen)),
247 'Duplicate episode numbers {0} vs {1}',
248 sorted(seen), sorted(set(seen)))
250 # Check that numbers are consecutive.
251 seen = sorted([int(s) for s in seen])
253 for i in range(len(seen) - 1):
254 clean = clean and ((seen[i+1] - seen[i]) == 1)
255 reporter.check(clean,
257 'Missing or non-consecutive episode numbers {0}',
261 def create_checker(args, filename, info):
262 """Create appropriate checker for file."""
264 for (pat, cls) in CHECKERS:
265 if pat.search(filename):
266 return cls(args, filename, **info)
269 class CheckBase(object):
270 """Base class for checking Markdown files."""
272 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
273 """Cache arguments for checking."""
275 super(CheckBase, self).__init__()
277 self.reporter = self.args.reporter # for convenience
278 self.filename = filename
279 self.metadata = metadata
280 self.metadata_len = metadata_len
290 self.check_metadata()
291 self.check_line_lengths()
292 self.check_trailing_whitespace()
293 self.check_blockquote_classes()
294 self.check_codeblock_classes()
295 self.check_defined_link_references()
297 def check_metadata(self):
298 """Check the YAML metadata."""
300 self.reporter.check(self.metadata is not None,
302 'Missing metadata entirely')
304 if self.metadata and (self.layout is not None):
305 self.reporter.check_field(
306 self.filename, 'metadata', self.metadata, 'layout', self.layout)
308 def check_line_lengths(self):
309 """Check the raw text of the lesson body."""
311 if self.args.line_lengths:
312 over = [i for (i, l, n) in self.lines if (
313 n > MAX_LINE_LEN) and (not l.startswith('!'))]
314 self.reporter.check(not over,
316 'Line(s) are too long: {0}',
317 ', '.join([str(i) for i in over]))
319 def check_trailing_whitespace(self):
320 """Check for whitespace at the ends of lines."""
322 if self.args.trailing_whitespace:
324 i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
325 self.reporter.check(not trailing,
327 'Line(s) end with whitespace: {0}',
328 ', '.join([str(i) for i in trailing]))
330 def check_blockquote_classes(self):
331 """Check that all blockquotes have known classes."""
333 for node in self.find_all(self.doc, {'type': 'blockquote'}):
334 cls = self.get_val(node, 'attr', 'class')
335 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
336 (self.filename, self.get_loc(node)),
337 'Unknown or missing blockquote type {0}',
340 def check_codeblock_classes(self):
341 """Check that all code blocks have known classes."""
343 for node in self.find_all(self.doc, {'type': 'codeblock'}):
344 cls = self.get_val(node, 'attr', 'class')
345 self.reporter.check(cls in KNOWN_CODEBLOCKS,
346 (self.filename, self.get_loc(node)),
347 'Unknown or missing code block type {0}',
350 def check_defined_link_references(self):
351 """Check that defined links resolve in the file.
353 Internally-defined links match the pattern [text][label].
357 for node in self.find_all(self.doc, {'type': 'text'}):
358 for match in P_INTERNAL_LINK_REF.findall(node['value']):
361 if link not in self.args.references:
362 result.add('"{0}"=>"{1}"'.format(text, link))
363 self.reporter.check(not result,
365 'Internally-defined links may be missing definitions: {0}',
366 ', '.join(sorted(result)))
368 def find_all(self, node, pattern, accum=None):
369 """Find all matches for a pattern."""
371 assert isinstance(pattern, dict), 'Patterns must be dictionaries'
374 if self.match(node, pattern):
376 for child in node.get('children', []):
377 self.find_all(child, pattern, accum)
380 def match(self, node, pattern):
381 """Does this node match the given pattern?"""
387 if isinstance(val, str):
390 elif isinstance(val, dict):
391 if not self.match(node[key], val):
395 def get_val(self, node, *chain):
396 """Get value one or more levels down."""
399 for selector in chain:
400 curr = curr.get(selector, None)
405 def get_loc(self, node):
406 """Convenience method to get node's line number."""
408 result = self.get_val(node, 'options', 'location')
409 if self.metadata_len is not None:
410 result += self.metadata_len
414 class CheckNonJekyll(CheckBase):
415 """Check a file that isn't translated by Jekyll."""
417 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
418 super(CheckNonJekyll, self).__init__(
419 args, filename, metadata, metadata_len, text, lines, doc)
421 def check_metadata(self):
422 self.reporter.check(self.metadata is None,
424 'Unexpected metadata')
427 class CheckIndex(CheckBase):
428 """Check the main index page."""
430 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
431 super(CheckIndex, self).__init__(args, filename,
432 metadata, metadata_len, text, lines, doc)
433 self.layout = 'lesson'
435 def check_metadata(self):
436 super(CheckIndex, self).check_metadata()
437 self.reporter.check(self.metadata.get('root', '') == '.',
439 'Root not set to "."')
442 class CheckEpisode(CheckBase):
443 """Check an episode page."""
445 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
446 super(CheckEpisode, self).__init__(args, filename,
447 metadata, metadata_len, text, lines, doc)
450 """Run extra tests."""
452 super(CheckEpisode, self).check()
453 self.check_reference_inclusion()
455 def check_metadata(self):
456 super(CheckEpisode, self).check_metadata()
458 if 'layout' in self.metadata:
459 if self.metadata['layout'] == 'break':
460 self.check_metadata_fields(BREAK_METADATA_FIELDS)
462 self.reporter.add(self.filename,
463 'Unknown episode layout "{0}"',
464 self.metadata['layout'])
466 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
468 def check_metadata_fields(self, expected):
469 for (name, type_) in expected:
470 if name not in self.metadata:
471 self.reporter.add(self.filename,
472 'Missing metadata field {0}',
474 elif not isinstance(self.metadata[name], type_):
475 self.reporter.add(self.filename,
476 '"{0}" has wrong type in metadata ({1} instead of {2})',
477 name, type(self.metadata[name]), type_)
479 def check_reference_inclusion(self):
480 """Check that links file has been included."""
482 if not self.args.reference_path:
485 for (i, last_line, line_len) in reversed(self.lines):
490 'No non-empty lines in {0}'.format(self.filename))
492 include_filename = os.path.split(self.args.reference_path)[-1]
493 if include_filename not in last_line:
494 self.reporter.add(self.filename,
495 'episode does not include "{0}"',
499 class CheckReference(CheckBase):
500 """Check the reference page."""
502 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
503 super(CheckReference, self).__init__(
504 args, filename, metadata, metadata_len, text, lines, doc)
505 self.layout = 'reference'
508 class CheckGeneric(CheckBase):
509 """Check a generic page."""
511 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
512 super(CheckGeneric, self).__init__(args, filename,
513 metadata, metadata_len, text, lines, doc)
518 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
519 (re.compile(r'README\.md'), CheckNonJekyll),
520 (re.compile(r'index\.md'), CheckIndex),
521 (re.compile(r'reference\.md'), CheckReference),
522 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
523 (re.compile(r'.*\.md'), CheckGeneric)
527 if __name__ == '__main__':