4 Check lesson files and their contents.
12 from optparse import OptionParser
14 from util import Reporter, read_markdown, load_yaml
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 specially.
26 # This list must include all the Markdown files listed in the 'bin/initialize' script.
29 '%/CONTRIBUTING.md': False,
32 '%/_extras/discuss.md': True,
33 '%/_extras/figures.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 # What kinds of blockquotes are allowed?
60 # What kinds of code fragments are allowed?
72 # What fields are required in teaching episode metadata?
73 TEACHING_METADATA_FIELDS = {
82 # What fields are required in break episode metadata?
83 BREAK_METADATA_FIELDS = {
89 # How long are lines allowed to be?
96 args.reporter = Reporter()
97 check_config(args.reporter, args.source_dir)
98 docs = read_all_markdown(args.source_dir, args.parser)
99 check_fileset(args.source_dir, args.reporter, docs.keys())
100 for filename in docs.keys():
101 checker = create_checker(args, filename, docs[filename])
103 args.reporter.report()
107 """Parse command-line arguments."""
109 parser = OptionParser()
110 parser.add_option('-l', '--linelen',
114 help='Check line lengths')
115 parser.add_option('-p', '--parser',
118 help='path to Markdown parser')
119 parser.add_option('-s', '--source',
122 help='source directory')
123 parser.add_option('-w', '--whitespace',
126 dest='trailing_whitespace',
127 help='Check for trailing whitespace')
129 args, extras = parser.parse_args()
130 require(args.parser is not None,
131 'Path to Markdown parser not provided')
133 'Unexpected trailing command-line arguments "{0}"'.format(extras))
138 def check_config(reporter, source_dir):
139 """Check configuration file."""
141 config_file = os.path.join(source_dir, '_config.yml')
142 config = load_yaml(config_file)
143 reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
146 def read_all_markdown(source_dir, parser):
147 """Read source files, returning
148 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
151 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
152 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
154 for pat in all_patterns:
155 for filename in glob.glob(pat):
156 data = read_markdown(parser, filename)
158 result[filename] = data
162 def check_fileset(source_dir, reporter, filenames_present):
163 """Are all required files present? Are extraneous files present?"""
165 # Check files with predictable names.
166 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
167 missing = set(required) - set(filenames_present)
169 reporter.add(None, 'Missing required file {0}', m)
171 # Check episode files' names.
173 for filename in filenames_present:
174 if '_episodes' not in filename:
176 m = P_EPISODE_FILENAME.search(filename)
178 seen.append(m.group(1))
180 reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
182 # Check for duplicate episode numbers.
183 reporter.check(len(seen) == len(set(seen)),
185 'Duplicate episode numbers {0} vs {1}',
186 sorted(seen), sorted(set(seen)))
188 # Check that numbers are consecutive.
189 seen = [int(s) for s in seen]
192 for i in range(len(seen) - 1):
193 clean = clean and ((seen[i+1] - seen[i]) == 1)
194 reporter.check(clean,
196 'Missing or non-consecutive episode numbers {0}',
200 def create_checker(args, filename, info):
201 """Create appropriate checker for file."""
203 for (pat, cls) in CHECKERS:
204 if pat.search(filename):
205 return cls(args, filename, **info)
208 def require(condition, message):
209 """Fail if condition not met."""
212 print(message, file=sys.stderr)
216 class CheckBase(object):
217 """Base class for checking Markdown files."""
219 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
220 """Cache arguments for checking."""
222 super(CheckBase, self).__init__()
224 self.reporter = self.args.reporter # for convenience
225 self.filename = filename
226 self.metadata = metadata
227 self.metadata_len = metadata_len
236 """Run tests on metadata."""
238 self.check_metadata()
239 self.check_line_lengths()
240 self.check_trailing_whitespace()
241 self.check_blockquote_classes()
242 self.check_codeblock_classes()
245 def check_metadata(self):
246 """Check the YAML metadata."""
248 self.reporter.check(self.metadata is not None,
250 'Missing metadata entirely')
252 if self.metadata and (self.layout is not None):
253 self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
256 def check_line_lengths(self):
257 """Check the raw text of the lesson body."""
259 if self.args.line_lengths:
260 over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
261 self.reporter.check(not over,
263 'Line(s) are too long: {0}',
264 ', '.join([str(i) for i in over]))
267 def check_trailing_whitespace(self):
268 """Check for whitespace at the ends of lines."""
270 if self.args.trailing_whitespace:
271 trailing = [i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
272 self.reporter.check(not trailing,
274 'Line(s) end with whitespace: {0}',
275 ', '.join([str(i) for i in trailing]))
278 def check_blockquote_classes(self):
279 """Check that all blockquotes have known classes."""
281 for node in self.find_all(self.doc, {'type' : 'blockquote'}):
282 cls = self.get_val(node, 'attr', 'class')
283 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
284 (self.filename, self.get_loc(node)),
285 'Unknown or missing blockquote type {0}',
289 def check_codeblock_classes(self):
290 """Check that all code blocks have known classes."""
292 for node in self.find_all(self.doc, {'type' : 'codeblock'}):
293 cls = self.get_val(node, 'attr', 'class')
294 self.reporter.check(cls in KNOWN_CODEBLOCKS,
295 (self.filename, self.get_loc(node)),
296 'Unknown or missing code block type {0}',
300 def find_all(self, node, pattern, accum=None):
301 """Find all matches for a pattern."""
303 assert type(pattern) == dict, 'Patterns must be dictionaries'
306 if self.match(node, pattern):
308 for child in node.get('children', []):
309 self.find_all(child, pattern, accum)
313 def match(self, node, pattern):
314 """Does this node match the given pattern?"""
323 elif type(val) == dict:
324 if not self.match(node[key], val):
329 def get_val(self, node, *chain):
330 """Get value one or more levels down."""
333 for selector in chain:
334 curr = curr.get(selector, None)
340 def get_loc(self, node):
341 """Convenience method to get node's line number."""
343 result = self.get_val(node, 'options', 'location')
344 if self.metadata_len is not None:
345 result += self.metadata_len
349 class CheckNonJekyll(CheckBase):
350 """Check a file that isn't translated by Jekyll."""
352 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
353 super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
356 def check_metadata(self):
357 self.reporter.check(self.metadata is None,
359 'Unexpected metadata')
362 class CheckIndex(CheckBase):
363 """Check the main index page."""
365 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
366 super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
367 self.layout = 'lesson'
370 class CheckEpisode(CheckBase):
371 """Check an episode page."""
373 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
374 super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
376 def check_metadata(self):
377 super(CheckEpisode, self).check_metadata()
379 if 'layout' in self.metadata:
380 if self.metadata['layout'] == 'break':
381 self.check_metadata_fields(BREAK_METADATA_FIELDS)
383 self.reporter.add(self.filename,
384 'Unknown episode layout "{0}"',
385 self.metadata['layout'])
387 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
390 def check_metadata_fields(self, expected):
391 for (name, type_) in expected:
392 if name not in self.metadata:
393 self.reporter.add(self.filename,
394 'Missing metadata field {0}',
396 elif type(self.metadata[name]) != type_:
397 self.reporter.add(self.filename,
398 '"{0}" has wrong type in metadata ({1} instead of {2})',
399 name, type(self.metadata[name]), type_)
402 class CheckReference(CheckBase):
403 """Check the reference page."""
405 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
406 super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
407 self.layout = 'reference'
410 class CheckGeneric(CheckBase):
411 """Check a generic page."""
413 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
414 super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
419 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
420 (re.compile(r'README\.md'), CheckNonJekyll),
421 (re.compile(r'index\.md'), CheckIndex),
422 (re.compile(r'reference\.md'), CheckReference),
423 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
424 (re.compile(r'.*\.md'), CheckGeneric)
428 if __name__ == '__main__':