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?
85 # What fields are required in teaching episode metadata?
86 TEACHING_METADATA_FIELDS = {
95 # What fields are required in break episode metadata?
96 BREAK_METADATA_FIELDS = {
102 # 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, docs.keys())
115 check_unwanted_files(args.source_dir, args.reporter)
116 for filename in docs.keys():
117 checker = create_checker(args, filename, docs[filename])
119 check_figures(args.source_dir, args.reporter)
121 args.reporter.report()
125 """Parse command-line arguments."""
127 parser = OptionParser()
128 parser.add_option('-l', '--linelen',
132 help='Check line lengths')
133 parser.add_option('-p', '--parser',
136 help='path to Markdown parser')
137 parser.add_option('-r', '--references',
139 dest='reference_path',
140 help='path to Markdown file of external references')
141 parser.add_option('-s', '--source',
144 help='source directory')
145 parser.add_option('-w', '--whitespace',
148 dest='trailing_whitespace',
149 help='Check for trailing whitespace')
151 args, extras = parser.parse_args()
152 require(args.parser is not None,
153 'Path to Markdown parser not provided')
155 'Unexpected trailing command-line arguments "{0}"'.format(extras))
160 def check_config(reporter, source_dir):
161 """Check configuration file."""
163 config_file = os.path.join(source_dir, '_config.yml')
164 config = load_yaml(config_file)
165 reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
166 reporter.check_field(config_file, 'configuration', config, 'carpentry', ('swc', 'dc', 'lc'))
167 reporter.check_field(config_file, 'configuration', config, 'title')
168 reporter.check_field(config_file, 'configuration', config, 'email')
170 reporter.check({'values': {'root': '..'}} in config.get('defaults', []),
172 '"root" not set to ".." in configuration')
175 def read_references(reporter, ref_path):
176 """Read shared file of reference links, returning dictionary of valid references
177 {symbolic_name : URL}
183 with open(ref_path, 'r') as reader:
184 for (num, line) in enumerate(reader):
186 m = P_INTERNAL_LINK_DEF.search(line)
188 '{0}:{1} not valid reference:\n{2}'.format(ref_path, line_num, line.rstrip()))
192 'Empty reference at {0}:{1}'.format(ref_path, line_num))
193 reporter.check(name not in result,
195 'Duplicate reference {0} at line {1}',
197 reporter.check(url not in urls_seen,
199 'Duplicate definition of URL {0} at line {1}',
206 def read_all_markdown(source_dir, parser):
207 """Read source files, returning
208 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
211 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
212 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
214 for pat in all_patterns:
215 for filename in glob.glob(pat):
216 data = read_markdown(parser, filename)
218 result[filename] = data
222 def check_fileset(source_dir, reporter, filenames_present):
223 """Are all required files present? Are extraneous files present?"""
225 # Check files with predictable names.
226 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
227 missing = set(required) - set(filenames_present)
229 reporter.add(None, 'Missing required file {0}', m)
231 # Check episode files' names.
233 for filename in filenames_present:
234 if '_episodes' not in filename:
236 m = P_EPISODE_FILENAME.search(filename)
238 seen.append(m.group(1))
240 reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
242 # Check for duplicate episode numbers.
243 reporter.check(len(seen) == len(set(seen)),
245 'Duplicate episode numbers {0} vs {1}',
246 sorted(seen), sorted(set(seen)))
248 # Check that numbers are consecutive.
249 seen = [int(s) for s in seen]
252 for i in range(len(seen) - 1):
253 clean = clean and ((seen[i+1] - seen[i]) == 1)
254 reporter.check(clean,
256 'Missing or non-consecutive episode numbers {0}',
260 def check_figures(source_dir, reporter):
261 """Check that all figures are present and referenced."""
265 all_figures_html = os.path.join(source_dir, '_includes', 'all_figures.html')
266 with open(all_figures_html, 'r') as reader:
268 figures = P_FIGURE_REFS.findall(text)
269 referenced = [os.path.split(f)[1] for f in figures if '/fig/' in f]
270 except FileNotFoundError as e:
271 reporter.add(all_figures_html,
275 # Get actual image files (ignore non-image files).
276 fig_dir_path = os.path.join(source_dir, 'fig')
277 actual = [f for f in os.listdir(fig_dir_path) if os.path.splitext(f)[1] in IMAGE_FILE_SUFFIX]
279 # Report differences.
280 unexpected = set(actual) - set(referenced)
281 reporter.check(not unexpected,
283 'Unexpected image files: {0}',
284 ', '.join(sorted(unexpected)))
285 missing = set(referenced) - set(actual)
286 reporter.check(not missing,
288 'Missing image files: {0}',
289 ', '.join(sorted(missing)))
292 def create_checker(args, filename, info):
293 """Create appropriate checker for file."""
295 for (pat, cls) in CHECKERS:
296 if pat.search(filename):
297 return cls(args, filename, **info)
300 class CheckBase(object):
301 """Base class for checking Markdown files."""
303 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
304 """Cache arguments for checking."""
306 super(CheckBase, self).__init__()
308 self.reporter = self.args.reporter # for convenience
309 self.filename = filename
310 self.metadata = metadata
311 self.metadata_len = metadata_len
322 self.check_metadata()
323 self.check_line_lengths()
324 self.check_trailing_whitespace()
325 self.check_blockquote_classes()
326 self.check_codeblock_classes()
327 self.check_defined_link_references()
330 def check_metadata(self):
331 """Check the YAML metadata."""
333 self.reporter.check(self.metadata is not None,
335 'Missing metadata entirely')
337 if self.metadata and (self.layout is not None):
338 self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
341 def check_line_lengths(self):
342 """Check the raw text of the lesson body."""
344 if self.args.line_lengths:
345 over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
346 self.reporter.check(not over,
348 'Line(s) are too long: {0}',
349 ', '.join([str(i) for i in over]))
352 def check_trailing_whitespace(self):
353 """Check for whitespace at the ends of lines."""
355 if self.args.trailing_whitespace:
356 trailing = [i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
357 self.reporter.check(not trailing,
359 'Line(s) end with whitespace: {0}',
360 ', '.join([str(i) for i in trailing]))
363 def check_blockquote_classes(self):
364 """Check that all blockquotes have known classes."""
366 for node in self.find_all(self.doc, {'type' : 'blockquote'}):
367 cls = self.get_val(node, 'attr', 'class')
368 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
369 (self.filename, self.get_loc(node)),
370 'Unknown or missing blockquote type {0}',
374 def check_codeblock_classes(self):
375 """Check that all code blocks have known classes."""
377 for node in self.find_all(self.doc, {'type' : 'codeblock'}):
378 cls = self.get_val(node, 'attr', 'class')
379 self.reporter.check(cls in KNOWN_CODEBLOCKS,
380 (self.filename, self.get_loc(node)),
381 'Unknown or missing code block type {0}',
385 def check_defined_link_references(self):
386 """Check that defined links resolve in the file.
388 Internally-defined links match the pattern [text][label].
392 for node in self.find_all(self.doc, {'type' : 'text'}):
393 for match in P_INTERNAL_LINK_REF.findall(node['value']):
396 if link not in self.args.references:
397 result.add('"{0}"=>"{1}"'.format(text, link))
398 self.reporter.check(not result,
400 'Internally-defined links may be missing definitions: {0}',
401 ', '.join(sorted(result)))
404 def find_all(self, node, pattern, accum=None):
405 """Find all matches for a pattern."""
407 assert type(pattern) == dict, 'Patterns must be dictionaries'
410 if self.match(node, pattern):
412 for child in node.get('children', []):
413 self.find_all(child, pattern, accum)
417 def match(self, node, pattern):
418 """Does this node match the given pattern?"""
427 elif type(val) == dict:
428 if not self.match(node[key], val):
433 def get_val(self, node, *chain):
434 """Get value one or more levels down."""
437 for selector in chain:
438 curr = curr.get(selector, None)
444 def get_loc(self, node):
445 """Convenience method to get node's line number."""
447 result = self.get_val(node, 'options', 'location')
448 if self.metadata_len is not None:
449 result += self.metadata_len
453 class CheckNonJekyll(CheckBase):
454 """Check a file that isn't translated by Jekyll."""
456 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
457 super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
460 def check_metadata(self):
461 self.reporter.check(self.metadata is None,
463 'Unexpected metadata')
466 class CheckIndex(CheckBase):
467 """Check the main index page."""
469 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
470 super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
471 self.layout = 'lesson'
473 def check_metadata(self):
474 super(CheckIndex, self).check_metadata()
475 self.reporter.check(self.metadata.get('root', '') == '.',
477 'Root not set to "."')
480 class CheckEpisode(CheckBase):
481 """Check an episode page."""
483 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
484 super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
488 """Run extra tests."""
490 super(CheckEpisode, self).check()
491 self.check_reference_inclusion()
494 def check_metadata(self):
495 super(CheckEpisode, self).check_metadata()
497 if 'layout' in self.metadata:
498 if self.metadata['layout'] == 'break':
499 self.check_metadata_fields(BREAK_METADATA_FIELDS)
501 self.reporter.add(self.filename,
502 'Unknown episode layout "{0}"',
503 self.metadata['layout'])
505 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
508 def check_metadata_fields(self, expected):
509 for (name, type_) in expected:
510 if name not in self.metadata:
511 self.reporter.add(self.filename,
512 'Missing metadata field {0}',
514 elif type(self.metadata[name]) != type_:
515 self.reporter.add(self.filename,
516 '"{0}" has wrong type in metadata ({1} instead of {2})',
517 name, type(self.metadata[name]), type_)
520 def check_reference_inclusion(self):
521 """Check that links file has been included."""
523 if not self.args.reference_path:
526 for (i, last_line, line_len) in reversed(self.lines):
531 'No non-empty lines in {0}'.format(self.filename))
533 include_filename = os.path.split(self.args.reference_path)[-1]
534 if include_filename not in last_line:
535 self.reporter.add(self.filename,
536 'episode does not include "{0}"',
540 class CheckReference(CheckBase):
541 """Check the reference page."""
543 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
544 super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
545 self.layout = 'reference'
548 class CheckGeneric(CheckBase):
549 """Check a generic page."""
551 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
552 super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
557 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
558 (re.compile(r'README\.md'), CheckNonJekyll),
559 (re.compile(r'index\.md'), CheckIndex),
560 (re.compile(r'reference\.md'), CheckReference),
561 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
562 (re.compile(r'.*\.md'), CheckGeneric)
566 if __name__ == '__main__':