4 Check lesson files and their contents.
12 from optparse import OptionParser
14 from util import Reporter, read_markdown, load_yaml, check_unwanted_files, require, IMAGE_FILE_SUFFIX
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 # Pattern to match figure references in HTML.
47 P_FIGURE_REFS = re.compile(r'<img[^>]+src="([^"]+)"[^>]*>')
49 # What kinds of blockquotes are allowed?
63 # 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()
101 check_config(args.reporter, args.source_dir)
102 docs = read_all_markdown(args.source_dir, args.parser)
103 check_fileset(args.source_dir, args.reporter, docs.keys())
104 check_unwanted_files(args.source_dir, args.reporter)
105 for filename in docs.keys():
106 checker = create_checker(args, filename, docs[filename])
108 check_figures(args.source_dir, args.reporter)
109 args.reporter.report()
113 """Parse command-line arguments."""
115 parser = OptionParser()
116 parser.add_option('-l', '--linelen',
120 help='Check line lengths')
121 parser.add_option('-p', '--parser',
124 help='path to Markdown parser')
125 parser.add_option('-s', '--source',
128 help='source directory')
129 parser.add_option('-w', '--whitespace',
132 dest='trailing_whitespace',
133 help='Check for trailing whitespace')
135 args, extras = parser.parse_args()
136 require(args.parser is not None,
137 'Path to Markdown parser not provided')
139 'Unexpected trailing command-line arguments "{0}"'.format(extras))
144 def check_config(reporter, source_dir):
145 """Check configuration file."""
147 config_file = os.path.join(source_dir, '_config.yml')
148 config = load_yaml(config_file)
149 reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
150 reporter.check_field(config_file, 'configuration', config, 'carpentry', ('swc', 'dc'))
151 reporter.check_field(config_file, 'configuration', config, 'title')
152 reporter.check_field(config_file, 'configuration', config, 'email')
153 reporter.check_field(config_file, 'configuration', config, 'repo')
154 reporter.check_field(config_file, 'configuration', config, 'root')
155 if ('repo' in config) and ('root' in config):
156 reporter.check(config['repo'].endswith(config['root']),
158 'Repository name "{0}" not consistent with root "{1}"',
159 config['repo'], config['root'])
162 def read_all_markdown(source_dir, parser):
163 """Read source files, returning
164 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
167 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
168 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
170 for pat in all_patterns:
171 for filename in glob.glob(pat):
172 data = read_markdown(parser, filename)
174 result[filename] = data
178 def check_fileset(source_dir, reporter, filenames_present):
179 """Are all required files present? Are extraneous files present?"""
181 # Check files with predictable names.
182 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
183 missing = set(required) - set(filenames_present)
185 reporter.add(None, 'Missing required file {0}', m)
187 # Check episode files' names.
189 for filename in filenames_present:
190 if '_episodes' not in filename:
192 m = P_EPISODE_FILENAME.search(filename)
194 seen.append(m.group(1))
196 reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
198 # Check for duplicate episode numbers.
199 reporter.check(len(seen) == len(set(seen)),
201 'Duplicate episode numbers {0} vs {1}',
202 sorted(seen), sorted(set(seen)))
204 # Check that numbers are consecutive.
205 seen = [int(s) for s in seen]
208 for i in range(len(seen) - 1):
209 clean = clean and ((seen[i+1] - seen[i]) == 1)
210 reporter.check(clean,
212 'Missing or non-consecutive episode numbers {0}',
216 def check_figures(source_dir, reporter):
217 """Check that all figures are present and referenced."""
221 all_figures_html = os.path.join(source_dir, '_includes', 'all_figures.html')
222 with open(all_figures_html, 'r') as reader:
224 figures = P_FIGURE_REFS.findall(text)
225 referenced = [os.path.split(f)[1] for f in figures if '/fig/' in f]
226 except FileNotFoundError as e:
227 reporter.add(all_figures_html,
231 # Get actual image files (ignore non-image files).
232 fig_dir_path = os.path.join(source_dir, 'fig')
233 actual = [f for f in os.listdir(fig_dir_path) if os.path.splitext(f)[1] in IMAGE_FILE_SUFFIX]
235 # Report differences.
236 unexpected = set(actual) - set(referenced)
237 reporter.check(not unexpected,
239 'Unexpected image files: {0}',
240 ', '.join(sorted(unexpected)))
241 missing = set(referenced) - set(actual)
242 reporter.check(not missing,
244 'Missing image files: {0}',
245 ', '.join(sorted(missing)))
248 def create_checker(args, filename, info):
249 """Create appropriate checker for file."""
251 for (pat, cls) in CHECKERS:
252 if pat.search(filename):
253 return cls(args, filename, **info)
256 class CheckBase(object):
257 """Base class for checking Markdown files."""
259 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
260 """Cache arguments for checking."""
262 super(CheckBase, self).__init__()
264 self.reporter = self.args.reporter # for convenience
265 self.filename = filename
266 self.metadata = metadata
267 self.metadata_len = metadata_len
276 """Run tests on metadata."""
278 self.check_metadata()
279 self.check_line_lengths()
280 self.check_trailing_whitespace()
281 self.check_blockquote_classes()
282 self.check_codeblock_classes()
285 def check_metadata(self):
286 """Check the YAML metadata."""
288 self.reporter.check(self.metadata is not None,
290 'Missing metadata entirely')
292 if self.metadata and (self.layout is not None):
293 self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
296 def check_line_lengths(self):
297 """Check the raw text of the lesson body."""
299 if self.args.line_lengths:
300 over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
301 self.reporter.check(not over,
303 'Line(s) are too long: {0}',
304 ', '.join([str(i) for i in over]))
307 def check_trailing_whitespace(self):
308 """Check for whitespace at the ends of lines."""
310 if self.args.trailing_whitespace:
311 trailing = [i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
312 self.reporter.check(not trailing,
314 'Line(s) end with whitespace: {0}',
315 ', '.join([str(i) for i in trailing]))
318 def check_blockquote_classes(self):
319 """Check that all blockquotes have known classes."""
321 for node in self.find_all(self.doc, {'type' : 'blockquote'}):
322 cls = self.get_val(node, 'attr', 'class')
323 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
324 (self.filename, self.get_loc(node)),
325 'Unknown or missing blockquote type {0}',
329 def check_codeblock_classes(self):
330 """Check that all code blocks have known classes."""
332 for node in self.find_all(self.doc, {'type' : 'codeblock'}):
333 cls = self.get_val(node, 'attr', 'class')
334 self.reporter.check(cls in KNOWN_CODEBLOCKS,
335 (self.filename, self.get_loc(node)),
336 'Unknown or missing code block type {0}',
340 def find_all(self, node, pattern, accum=None):
341 """Find all matches for a pattern."""
343 assert type(pattern) == dict, 'Patterns must be dictionaries'
346 if self.match(node, pattern):
348 for child in node.get('children', []):
349 self.find_all(child, pattern, accum)
353 def match(self, node, pattern):
354 """Does this node match the given pattern?"""
363 elif type(val) == dict:
364 if not self.match(node[key], val):
369 def get_val(self, node, *chain):
370 """Get value one or more levels down."""
373 for selector in chain:
374 curr = curr.get(selector, None)
380 def get_loc(self, node):
381 """Convenience method to get node's line number."""
383 result = self.get_val(node, 'options', 'location')
384 if self.metadata_len is not None:
385 result += self.metadata_len
389 class CheckNonJekyll(CheckBase):
390 """Check a file that isn't translated by Jekyll."""
392 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
393 super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
396 def check_metadata(self):
397 self.reporter.check(self.metadata is None,
399 'Unexpected metadata')
402 class CheckIndex(CheckBase):
403 """Check the main index page."""
405 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
406 super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
407 self.layout = 'lesson'
410 class CheckEpisode(CheckBase):
411 """Check an episode page."""
413 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
414 super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
416 def check_metadata(self):
417 super(CheckEpisode, self).check_metadata()
419 if 'layout' in self.metadata:
420 if self.metadata['layout'] == 'break':
421 self.check_metadata_fields(BREAK_METADATA_FIELDS)
423 self.reporter.add(self.filename,
424 'Unknown episode layout "{0}"',
425 self.metadata['layout'])
427 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
430 def check_metadata_fields(self, expected):
431 for (name, type_) in expected:
432 if name not in self.metadata:
433 self.reporter.add(self.filename,
434 'Missing metadata field {0}',
436 elif type(self.metadata[name]) != type_:
437 self.reporter.add(self.filename,
438 '"{0}" has wrong type in metadata ({1} instead of {2})',
439 name, type(self.metadata[name]), type_)
442 class CheckReference(CheckBase):
443 """Check the reference page."""
445 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
446 super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
447 self.layout = 'reference'
450 class CheckGeneric(CheckBase):
451 """Check a generic page."""
453 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
454 super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
459 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
460 (re.compile(r'README\.md'), CheckNonJekyll),
461 (re.compile(r'index\.md'), CheckIndex),
462 (re.compile(r'reference\.md'), CheckReference),
463 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
464 (re.compile(r'.*\.md'), CheckGeneric)
468 if __name__ == '__main__':