4 Check lesson files and their contents.
7 from __future__ import print_function
13 from optparse import OptionParser
15 from util import Reporter, read_markdown, load_yaml, check_unwanted_files, require, IMAGE_FILE_SUFFIX
19 # Where to look for source Markdown files.
20 SOURCE_DIRS = ['', '_episodes', '_extras']
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 specially.
27 # This list must include all the Markdown files listed in the 'bin/initialize' script.
30 '%/CONTRIBUTING.md': False,
33 '%/_extras/discuss.md': True,
34 '%/_extras/figures.md': True,
35 '%/_extras/guide.md': True,
37 '%/reference.md': True,
41 # Episode filename pattern.
42 P_EPISODE_FILENAME = re.compile(r'/_episodes/(\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 # What kinds of blockquotes are allowed?
70 # What kinds of code fragments are allowed?
83 # What fields are required in teaching episode metadata?
84 TEACHING_METADATA_FIELDS = {
93 # What fields are required in break episode metadata?
94 BREAK_METADATA_FIELDS = {
100 # How long are lines allowed to be?
107 args.reporter = Reporter()
108 check_config(args.reporter, args.source_dir)
109 args.references = read_references(args.reporter, args.reference_path)
111 docs = read_all_markdown(args.source_dir, args.parser)
112 check_fileset(args.source_dir, args.reporter, docs.keys())
113 check_unwanted_files(args.source_dir, args.reporter)
114 for filename in docs.keys():
115 checker = create_checker(args, filename, docs[filename])
117 check_figures(args.source_dir, args.reporter)
119 args.reporter.report()
123 """Parse command-line arguments."""
125 parser = OptionParser()
126 parser.add_option('-l', '--linelen',
130 help='Check line lengths')
131 parser.add_option('-p', '--parser',
134 help='path to Markdown parser')
135 parser.add_option('-r', '--references',
137 dest='reference_path',
138 help='path to Markdown file of external references')
139 parser.add_option('-s', '--source',
142 help='source directory')
143 parser.add_option('-w', '--whitespace',
146 dest='trailing_whitespace',
147 help='Check for trailing whitespace')
149 args, extras = parser.parse_args()
150 require(args.parser is not None,
151 'Path to Markdown parser not provided')
153 'Unexpected trailing command-line arguments "{0}"'.format(extras))
158 def check_config(reporter, source_dir):
159 """Check configuration file."""
161 config_file = os.path.join(source_dir, '_config.yml')
162 config = load_yaml(config_file)
163 reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
164 reporter.check_field(config_file, 'configuration', config, 'carpentry', ('swc', 'dc', 'lc'))
165 reporter.check_field(config_file, 'configuration', config, 'title')
166 reporter.check_field(config_file, 'configuration', config, 'contact')
168 reporter.check({'values': {'root': '..'}} in config.get('defaults', []),
170 '"root" not set to ".." in configuration')
173 def read_references(reporter, ref_path):
174 """Read shared file of reference links, returning dictionary of valid references
175 {symbolic_name : URL}
181 with open(ref_path, 'r') as reader:
182 for (num, line) in enumerate(reader):
184 m = P_INTERNAL_LINK_DEF.search(line)
186 '{0}:{1} not valid reference:\n{2}'.format(ref_path, line_num, line.rstrip()))
190 'Empty reference at {0}:{1}'.format(ref_path, line_num))
191 reporter.check(name not in result,
193 'Duplicate reference {0} at line {1}',
195 reporter.check(url not in urls_seen,
197 'Duplicate definition of URL {0} at line {1}',
204 def read_all_markdown(source_dir, parser):
205 """Read source files, returning
206 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
209 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
210 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
212 for pat in all_patterns:
213 for filename in glob.glob(pat):
214 data = read_markdown(parser, filename)
216 result[filename] = data
220 def check_fileset(source_dir, reporter, filenames_present):
221 """Are all required files present? Are extraneous files present?"""
223 # Check files with predictable names.
224 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
225 missing = set(required) - set(filenames_present)
227 reporter.add(None, 'Missing required file {0}', m)
229 # Check episode files' names.
231 for filename in filenames_present:
232 if '_episodes' not in filename:
234 m = P_EPISODE_FILENAME.search(filename)
236 seen.append(m.group(1))
238 reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
240 # Check for duplicate episode numbers.
241 reporter.check(len(seen) == len(set(seen)),
243 'Duplicate episode numbers {0} vs {1}',
244 sorted(seen), sorted(set(seen)))
246 # Check that numbers are consecutive.
247 seen = [int(s) for s in seen]
250 for i in range(len(seen) - 1):
251 clean = clean and ((seen[i+1] - seen[i]) == 1)
252 reporter.check(clean,
254 'Missing or non-consecutive episode numbers {0}',
258 def check_figures(source_dir, reporter):
259 """Check that all figures are present and referenced."""
263 all_figures_html = os.path.join(source_dir, '_includes', 'all_figures.html')
264 with open(all_figures_html, 'r') as reader:
266 figures = P_FIGURE_REFS.findall(text)
267 referenced = [os.path.split(f)[1] for f in figures if '/fig/' in f]
268 except FileNotFoundError as e:
269 reporter.add(all_figures_html,
273 # Get actual image files (ignore non-image files).
274 fig_dir_path = os.path.join(source_dir, 'fig')
275 actual = [f for f in os.listdir(fig_dir_path) if os.path.splitext(f)[1] in IMAGE_FILE_SUFFIX]
277 # Report differences.
278 unexpected = set(actual) - set(referenced)
279 reporter.check(not unexpected,
281 'Unexpected image files: {0}',
282 ', '.join(sorted(unexpected)))
283 missing = set(referenced) - set(actual)
284 reporter.check(not missing,
286 'Missing image files: {0}',
287 ', '.join(sorted(missing)))
290 def create_checker(args, filename, info):
291 """Create appropriate checker for file."""
293 for (pat, cls) in CHECKERS:
294 if pat.search(filename):
295 return cls(args, filename, **info)
298 class CheckBase(object):
299 """Base class for checking Markdown files."""
301 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
302 """Cache arguments for checking."""
304 super(CheckBase, self).__init__()
306 self.reporter = self.args.reporter # for convenience
307 self.filename = filename
308 self.metadata = metadata
309 self.metadata_len = metadata_len
320 self.check_metadata()
321 self.check_line_lengths()
322 self.check_trailing_whitespace()
323 self.check_blockquote_classes()
324 self.check_codeblock_classes()
325 self.check_defined_link_references()
328 def check_metadata(self):
329 """Check the YAML metadata."""
331 self.reporter.check(self.metadata is not None,
333 'Missing metadata entirely')
335 if self.metadata and (self.layout is not None):
336 self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
339 def check_line_lengths(self):
340 """Check the raw text of the lesson body."""
342 if self.args.line_lengths:
343 over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
344 self.reporter.check(not over,
346 'Line(s) are too long: {0}',
347 ', '.join([str(i) for i in over]))
350 def check_trailing_whitespace(self):
351 """Check for whitespace at the ends of lines."""
353 if self.args.trailing_whitespace:
354 trailing = [i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
355 self.reporter.check(not trailing,
357 'Line(s) end with whitespace: {0}',
358 ', '.join([str(i) for i in trailing]))
361 def check_blockquote_classes(self):
362 """Check that all blockquotes have known classes."""
364 for node in self.find_all(self.doc, {'type' : 'blockquote'}):
365 cls = self.get_val(node, 'attr', 'class')
366 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
367 (self.filename, self.get_loc(node)),
368 'Unknown or missing blockquote type {0}',
372 def check_codeblock_classes(self):
373 """Check that all code blocks have known classes."""
375 for node in self.find_all(self.doc, {'type' : 'codeblock'}):
376 cls = self.get_val(node, 'attr', 'class')
377 self.reporter.check(cls in KNOWN_CODEBLOCKS,
378 (self.filename, self.get_loc(node)),
379 'Unknown or missing code block type {0}',
383 def check_defined_link_references(self):
384 """Check that defined links resolve in the file.
386 Internally-defined links match the pattern [text][label].
390 for node in self.find_all(self.doc, {'type' : 'text'}):
391 for match in P_INTERNAL_LINK_REF.findall(node['value']):
394 if link not in self.args.references:
395 result.add('"{0}"=>"{1}"'.format(text, link))
396 self.reporter.check(not result,
398 'Internally-defined links may be missing definitions: {0}',
399 ', '.join(sorted(result)))
402 def find_all(self, node, pattern, accum=None):
403 """Find all matches for a pattern."""
405 assert type(pattern) == dict, 'Patterns must be dictionaries'
408 if self.match(node, pattern):
410 for child in node.get('children', []):
411 self.find_all(child, pattern, accum)
415 def match(self, node, pattern):
416 """Does this node match the given pattern?"""
425 elif type(val) == dict:
426 if not self.match(node[key], val):
431 def get_val(self, node, *chain):
432 """Get value one or more levels down."""
435 for selector in chain:
436 curr = curr.get(selector, None)
442 def get_loc(self, node):
443 """Convenience method to get node's line number."""
445 result = self.get_val(node, 'options', 'location')
446 if self.metadata_len is not None:
447 result += self.metadata_len
451 class CheckNonJekyll(CheckBase):
452 """Check a file that isn't translated by Jekyll."""
454 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
455 super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
458 def check_metadata(self):
459 self.reporter.check(self.metadata is None,
461 'Unexpected metadata')
464 class CheckIndex(CheckBase):
465 """Check the main index page."""
467 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
468 super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
469 self.layout = 'lesson'
471 def check_metadata(self):
472 super(CheckIndex, self).check_metadata()
473 self.reporter.check(self.metadata.get('root', '') == '.',
475 'Root not set to "."')
478 class CheckEpisode(CheckBase):
479 """Check an episode page."""
481 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
482 super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
486 """Run extra tests."""
488 super(CheckEpisode, self).check()
489 self.check_reference_inclusion()
492 def check_metadata(self):
493 super(CheckEpisode, self).check_metadata()
495 if 'layout' in self.metadata:
496 if self.metadata['layout'] == 'break':
497 self.check_metadata_fields(BREAK_METADATA_FIELDS)
499 self.reporter.add(self.filename,
500 'Unknown episode layout "{0}"',
501 self.metadata['layout'])
503 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
506 def check_metadata_fields(self, expected):
507 for (name, type_) in expected:
508 if name not in self.metadata:
509 self.reporter.add(self.filename,
510 'Missing metadata field {0}',
512 elif type(self.metadata[name]) != type_:
513 self.reporter.add(self.filename,
514 '"{0}" has wrong type in metadata ({1} instead of {2})',
515 name, type(self.metadata[name]), type_)
518 def check_reference_inclusion(self):
519 """Check that links file has been included."""
521 if not self.args.reference_path:
524 for (i, last_line, line_len) in reversed(self.lines):
529 'No non-empty lines in {0}'.format(self.filename))
531 include_filename = os.path.split(self.args.reference_path)[-1]
532 if include_filename not in last_line:
533 self.reporter.add(self.filename,
534 'episode does not include "{0}"',
538 class CheckReference(CheckBase):
539 """Check the reference page."""
541 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
542 super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
543 self.layout = 'reference'
546 class CheckGeneric(CheckBase):
547 """Check a generic page."""
549 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
550 super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
555 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
556 (re.compile(r'README\.md'), CheckNonJekyll),
557 (re.compile(r'index\.md'), CheckIndex),
558 (re.compile(r'reference\.md'), CheckReference),
559 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
560 (re.compile(r'.*\.md'), CheckGeneric)
564 if __name__ == '__main__':