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 # What kinds of blockquotes are allowed?
57 # What kinds of code fragments are allowed?
69 # What fields are required in teaching episode metadata?
70 TEACHING_METADATA_FIELDS = {
79 # What fields are required in break episode metadata?
80 BREAK_METADATA_FIELDS = {
86 # How long are lines allowed to be?
93 args.reporter = Reporter()
94 check_config(args.reporter, args.source_dir)
95 docs = read_all_markdown(args.source_dir, args.parser)
96 check_fileset(args.source_dir, args.reporter, docs.keys())
97 for filename in docs.keys():
98 checker = create_checker(args, filename, docs[filename])
100 args.reporter.report()
104 """Parse command-line arguments."""
106 parser = OptionParser()
107 parser.add_option('-l', '--linelen',
110 help='Check line lengths')
111 parser.add_option('-p', '--parser',
114 help='path to Markdown parser')
115 parser.add_option('-s', '--source',
118 help='source directory')
119 parser.add_option('-w', '--whitespace',
121 dest='trailing_whitespace',
122 help='Check for trailing whitespace')
124 args, extras = parser.parse_args()
125 require(args.parser is not None,
126 'Path to Markdown parser not provided')
128 'Unexpected trailing command-line arguments "{0}"'.format(extras))
133 def check_config(reporter, source_dir):
134 """Check configuration file."""
136 config_file = os.path.join(source_dir, '_config.yml')
137 config = load_yaml(config_file)
138 reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
141 def read_all_markdown(source_dir, parser):
142 """Read source files, returning
143 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
146 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
147 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
149 for pat in all_patterns:
150 for filename in glob.glob(pat):
151 data = read_markdown(parser, filename)
153 result[filename] = data
157 def check_fileset(source_dir, reporter, filenames_present):
158 """Are all required files present? Are extraneous files present?"""
160 # Check files with predictable names.
161 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
162 missing = set(required) - set(filenames_present)
164 reporter.add(None, 'Missing required file {0}', m)
166 # Check episode files' names.
168 for filename in filenames_present:
169 if '_episodes' not in filename:
171 m = P_EPISODE_FILENAME.search(filename)
173 seen.append(m.group(1))
175 reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
177 # Check for duplicate episode numbers.
178 reporter.check(len(seen) == len(set(seen)),
180 'Duplicate episode numbers {0} vs {1}',
181 sorted(seen), sorted(set(seen)))
183 # Check that numbers are consecutive.
184 seen = [int(s) for s in seen]
187 for i in range(len(seen) - 1):
188 clean = clean and ((seen[i+1] - seen[i]) == 1)
189 reporter.check(clean,
191 'Missing or non-consecutive episode numbers {0}',
195 def create_checker(args, filename, info):
196 """Create appropriate checker for file."""
198 for (pat, cls) in CHECKERS:
199 if pat.search(filename):
200 return cls(args, filename, **info)
203 def require(condition, message):
204 """Fail if condition not met."""
207 print(message, file=sys.stderr)
211 class CheckBase(object):
212 """Base class for checking Markdown files."""
214 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
215 """Cache arguments for checking."""
217 super(CheckBase, self).__init__()
219 self.reporter = self.args.reporter # for convenience
220 self.filename = filename
221 self.metadata = metadata
222 self.metadata_len = metadata_len
231 """Run tests on metadata."""
233 self.check_metadata()
234 self.check_line_lengths()
235 self.check_trailing_whitespace()
236 self.check_blockquote_classes()
237 self.check_codeblock_classes()
240 def check_metadata(self):
241 """Check the YAML metadata."""
243 self.reporter.check(self.metadata is not None,
245 'Missing metadata entirely')
247 if self.metadata and (self.layout is not None):
248 self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
251 def check_line_lengths(self):
252 """Check the raw text of the lesson body."""
254 if self.args.line_lengths:
255 over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
256 self.reporter.check(not over,
258 'Line(s) are too long: {0}',
259 ', '.join([str(i) for i in over]))
262 def check_trailing_whitespace(self):
263 """Check for whitespace at the ends of lines."""
265 if self.args.trailing_whitespace:
266 trailing = [i for (i, l, n) in self.lines if l.endswidth(' ')]
267 self.reporter.check(not trailing,
269 'Line(s) end with whitespace: {0}',
270 ', '.join([str[i] for i in over]))
273 def check_blockquote_classes(self):
274 """Check that all blockquotes have known classes."""
276 for node in self.find_all(self.doc, {'type' : 'blockquote'}):
277 cls = self.get_val(node, 'attr', 'class')
278 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
279 (self.filename, self.get_loc(node)),
280 'Unknown or missing blockquote type {0}',
284 def check_codeblock_classes(self):
285 """Check that all code blocks have known classes."""
287 for node in self.find_all(self.doc, {'type' : 'codeblock'}):
288 cls = self.get_val(node, 'attr', 'class')
289 self.reporter.check(cls in KNOWN_CODEBLOCKS,
290 (self.filename, self.get_loc(node)),
291 'Unknown or missing code block type {0}',
295 def find_all(self, node, pattern, accum=None):
296 """Find all matches for a pattern."""
298 assert type(pattern) == dict, 'Patterns must be dictionaries'
301 if self.match(node, pattern):
303 for child in node.get('children', []):
304 self.find_all(child, pattern, accum)
308 def match(self, node, pattern):
309 """Does this node match the given pattern?"""
318 elif type(val) == dict:
319 if not self.match(node[key], val):
324 def get_val(self, node, *chain):
325 """Get value one or more levels down."""
328 for selector in chain:
329 curr = curr.get(selector, None)
335 def get_loc(self, node):
336 """Convenience method to get node's line number."""
338 result = self.get_val(node, 'options', 'location')
339 if self.metadata_len is not None:
340 result += self.metadata_len
344 class CheckNonJekyll(CheckBase):
345 """Check a file that isn't translated by Jekyll."""
347 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
348 super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
351 def check_metadata(self):
352 self.reporter.check(self.metadata is None,
354 'Unexpected metadata')
357 class CheckIndex(CheckBase):
358 """Check the main index page."""
360 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
361 super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
362 self.layout = 'lesson'
365 class CheckEpisode(CheckBase):
366 """Check an episode page."""
368 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
369 super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
371 def check_metadata(self):
372 super(CheckEpisode, self).check_metadata()
374 if 'layout' in self.metadata:
375 if self.metadata['layout'] == 'break':
376 self.check_metadata_fields(BREAK_METADATA_FIELDS)
378 self.reporter.add(self.filename,
379 'Unknown episode layout "{0}"',
380 self.metadata['layout'])
382 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
385 def check_metadata_fields(self, expected):
386 for (name, type_) in expected:
387 if name not in self.metadata:
388 self.reporter.add(self.filename,
389 'Missing metadata field {0}',
391 elif type(self.metadata[name]) != type_:
392 self.reporter.add(self.filename,
393 '"{0}" has wrong type in metadata ({1} instead of {2})',
394 name, type(self.metadata[name]), type_)
397 class CheckReference(CheckBase):
398 """Check the reference page."""
400 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
401 super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
402 self.layout = 'reference'
405 class CheckGeneric(CheckBase):
406 """Check a generic page."""
408 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
409 super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
414 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
415 (re.compile(r'README\.md'), CheckNonJekyll),
416 (re.compile(r'index\.md'), CheckIndex),
417 (re.compile(r'reference\.md'), CheckReference),
418 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
419 (re.compile(r'.*\.md'), CheckGeneric)
423 if __name__ == '__main__':