4 Check lesson files and their contents.
13 from optparse import OptionParser
15 from util import Reporter, read_markdown
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 # Required non-Markdown files.
42 NON_MARKDOWN_FILES = {
47 # Episode filename pattern.
48 P_EPISODE_FILENAME = re.compile(r'/_episodes/(\d\d)-[-\w]+.md$')
50 # What kinds of blockquotes are allowed?
64 # What kinds of code fragments are allowed?
76 # What fields are required in teaching episode metadata?
77 TEACHING_METADATA_FIELDS = {
86 # What fields are required in break episode metadata?
87 BREAK_METADATA_FIELDS = {
93 # How long are lines allowed to be?
100 args.reporter = Reporter(args)
102 check_non_markdown_files(args.source_dir, args.reporter)
103 docs = read_all_markdown(args, args.source_dir)
104 check_fileset(args.source_dir, args.reporter, docs.keys())
105 for filename in docs.keys():
106 checker = create_checker(args, filename, docs[filename])
108 args.reporter.report()
112 """Parse command-line arguments."""
114 parser = OptionParser()
115 parser.add_option('-l', '--linelen',
118 help='Check line lengths')
119 parser.add_option('-p', '--parser',
122 help='path to Markdown parser')
123 parser.add_option('-s', '--source',
126 help='source directory')
128 args, extras = parser.parse_args()
129 require(args.parser is not None,
130 'Path to Markdown parser not provided')
132 'Unexpected trailing command-line arguments "{0}"'.format(extras))
137 def check_config(args):
138 """Check configuration file."""
140 config_file = os.path.join(args.source_dir, '_config.yml')
141 with open(config_file, 'r') as reader:
142 config = yaml.load(reader)
144 args.reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
147 def check_non_markdown_files(source_dir, reporter):
148 """Check presence of non-Markdown files."""
150 for filename in NON_MARKDOWN_FILES:
151 path = os.path.join(source_dir, filename)
152 reporter.check(os.path.exists(path),
157 def read_all_markdown(args, source_dir):
158 """Read source files, returning
159 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
162 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
163 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
165 for pat in all_patterns:
166 for filename in glob.glob(pat):
167 data = read_markdown(args.parser, filename)
169 result[filename] = data
173 def check_fileset(source_dir, reporter, filenames_present):
174 """Are all required files present? Are extraneous files present?"""
176 # Check files with predictable names.
177 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
178 missing = set(required) - set(filenames_present)
180 reporter.add(None, 'Missing required file {0}', m)
182 # Check episode files' names.
184 for filename in filenames_present:
185 if '_episodes' not in filename:
187 m = P_EPISODE_FILENAME.search(filename)
189 seen.append(m.group(1))
191 reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
193 # Check for duplicate episode numbers.
194 reporter.check(len(seen) == len(set(seen)),
196 'Duplicate episode numbers {0} vs {1}',
197 sorted(seen), sorted(set(seen)))
199 # Check that numbers are consecutive.
200 seen = [int(s) for s in seen]
203 for i in range(len(seen) - 1):
204 clean = clean and ((seen[i+1] - seen[i]) == 1)
205 reporter.check(clean,
207 'Missing or non-consecutive episode numbers {0}',
211 def create_checker(args, filename, info):
212 """Create appropriate checker for file."""
214 for (pat, cls) in CHECKERS:
215 if pat.search(filename):
216 return cls(args, filename, **info)
219 def require(condition, message):
220 """Fail if condition not met."""
223 print(message, file=sys.stderr)
227 class CheckBase(object):
228 """Base class for checking Markdown files."""
230 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
231 """Cache arguments for checking."""
233 super(CheckBase, self).__init__()
235 self.reporter = self.args.reporter # for convenience
236 self.filename = filename
237 self.metadata = metadata
238 self.metadata_len = metadata_len
246 """Run tests on metadata."""
248 self.check_metadata()
250 self.check_blockquote_classes()
251 self.check_codeblock_classes()
254 def check_metadata(self):
255 """Check the YAML metadata."""
257 self.reporter.check(self.metadata is not None,
259 'Missing metadata entirely')
261 if self.metadata and (self.layout is not None):
262 self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
265 def check_text(self):
266 """Check the raw text of the lesson body."""
268 if self.args.line_len:
269 over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
270 self.reporter.check(not over,
272 'Line(s) are too long: {0}',
273 ', '.join([str(i) for i in over]))
276 def check_blockquote_classes(self):
277 """Check that all blockquotes have known classes."""
279 for node in self.find_all(self.doc, {'type' : 'blockquote'}):
280 cls = self.get_val(node, 'attr', 'class')
281 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
282 (self.filename, self.get_loc(node)),
283 'Unknown or missing blockquote type {0}',
287 def check_codeblock_classes(self):
288 """Check that all code blocks have known classes."""
290 for node in self.find_all(self.doc, {'type' : 'codeblock'}):
291 cls = self.get_val(node, 'attr', 'class')
292 self.reporter.check(cls in KNOWN_CODEBLOCKS,
293 (self.filename, self.get_loc(node)),
294 'Unknown or missing code block type {0}',
298 def find_all(self, node, pattern, accum=None):
299 """Find all matches for a pattern."""
301 assert type(pattern) == dict, 'Patterns must be dictionaries'
304 if self.match(node, pattern):
306 for child in node.get('children', []):
307 self.find_all(child, pattern, accum)
311 def match(self, node, pattern):
312 """Does this node match the given pattern?"""
321 elif type(val) == dict:
322 if not self.match(node[key], val):
327 def get_val(self, node, *chain):
328 """Get value one or more levels down."""
331 for selector in chain:
332 curr = curr.get(selector, None)
338 def get_loc(self, node):
339 """Convenience method to get node's line number."""
341 result = self.get_val(node, 'options', 'location')
342 if self.metadata_len is not None:
343 result += self.metadata_len
347 class CheckNonJekyll(CheckBase):
348 """Check a file that isn't translated by Jekyll."""
350 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
351 super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
354 def check_metadata(self):
355 self.reporter.check(self.metadata is None,
357 'Unexpected metadata')
360 class CheckIndex(CheckBase):
361 """Check the main index page."""
363 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
364 super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
365 self.layout = 'lesson'
368 class CheckEpisode(CheckBase):
369 """Check an episode page."""
371 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
372 super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
375 def check_metadata(self):
376 super(CheckEpisode, self).check_metadata()
378 if 'layout' in self.metadata:
379 if self.metadata['layout'] == 'break':
380 self.check_metadata_fields(BREAK_METADATA_FIELDS)
382 self.reporter.add(self.filename,
383 'Unknown episode layout "{0}"',
384 self.metadata['layout'])
386 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
389 def check_metadata_fields(self, expected):
390 for (name, type_) in expected:
391 if name not in self.metadata:
392 self.reporter.add(self.filename,
393 'Missing metadata field {0}',
395 elif type(self.metadata[name]) != type_:
396 self.reporter.add(self.filename,
397 '"{0}" has wrong type in metadata ({1} instead of {2})',
398 name, type(self.metadata[name]), type_)
401 class CheckReference(CheckBase):
402 """Check the reference page."""
404 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
405 super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
406 self.layout = 'reference'
409 class CheckGeneric(CheckBase):
410 """Check a generic page."""
412 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
413 super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
418 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
419 (re.compile(r'README\.md'), CheckNonJekyll),
420 (re.compile(r'index\.md'), CheckIndex),
421 (re.compile(r'reference\.md'), CheckReference),
422 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
423 (re.compile(r'.*\.md'), CheckGeneric)
427 if __name__ == '__main__':