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')
155 def read_all_markdown(source_dir, parser):
156 """Read source files, returning
157 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
160 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
161 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
163 for pat in all_patterns:
164 for filename in glob.glob(pat):
165 data = read_markdown(parser, filename)
167 result[filename] = data
171 def check_fileset(source_dir, reporter, filenames_present):
172 """Are all required files present? Are extraneous files present?"""
174 # Check files with predictable names.
175 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
176 missing = set(required) - set(filenames_present)
178 reporter.add(None, 'Missing required file {0}', m)
180 # Check episode files' names.
182 for filename in filenames_present:
183 if '_episodes' not in filename:
185 m = P_EPISODE_FILENAME.search(filename)
187 seen.append(m.group(1))
189 reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
191 # Check for duplicate episode numbers.
192 reporter.check(len(seen) == len(set(seen)),
194 'Duplicate episode numbers {0} vs {1}',
195 sorted(seen), sorted(set(seen)))
197 # Check that numbers are consecutive.
198 seen = [int(s) for s in seen]
201 for i in range(len(seen) - 1):
202 clean = clean and ((seen[i+1] - seen[i]) == 1)
203 reporter.check(clean,
205 'Missing or non-consecutive episode numbers {0}',
209 def check_figures(source_dir, reporter):
210 """Check that all figures are present and referenced."""
214 all_figures_html = os.path.join(source_dir, '_includes', 'all_figures.html')
215 with open(all_figures_html, 'r') as reader:
217 figures = P_FIGURE_REFS.findall(text)
218 referenced = [os.path.split(f)[1] for f in figures if '/fig/' in f]
219 except FileNotFoundError as e:
220 reporter.add(all_figures_html,
224 # Get actual image files (ignore non-image files).
225 fig_dir_path = os.path.join(source_dir, 'fig')
226 actual = [f for f in os.listdir(fig_dir_path) if os.path.splitext(f)[1] in IMAGE_FILE_SUFFIX]
228 # Report differences.
229 unexpected = set(actual) - set(referenced)
230 reporter.check(not unexpected,
232 'Unexpected image files: {0}',
233 ', '.join(sorted(unexpected)))
234 missing = set(referenced) - set(actual)
235 reporter.check(not missing,
237 'Missing image files: {0}',
238 ', '.join(sorted(missing)))
241 def create_checker(args, filename, info):
242 """Create appropriate checker for file."""
244 for (pat, cls) in CHECKERS:
245 if pat.search(filename):
246 return cls(args, filename, **info)
249 class CheckBase(object):
250 """Base class for checking Markdown files."""
252 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
253 """Cache arguments for checking."""
255 super(CheckBase, self).__init__()
257 self.reporter = self.args.reporter # for convenience
258 self.filename = filename
259 self.metadata = metadata
260 self.metadata_len = metadata_len
269 """Run tests on metadata."""
271 self.check_metadata()
272 self.check_line_lengths()
273 self.check_trailing_whitespace()
274 self.check_blockquote_classes()
275 self.check_codeblock_classes()
278 def check_metadata(self):
279 """Check the YAML metadata."""
281 self.reporter.check(self.metadata is not None,
283 'Missing metadata entirely')
285 if self.metadata and (self.layout is not None):
286 self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
289 def check_line_lengths(self):
290 """Check the raw text of the lesson body."""
292 if self.args.line_lengths:
293 over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
294 self.reporter.check(not over,
296 'Line(s) are too long: {0}',
297 ', '.join([str(i) for i in over]))
300 def check_trailing_whitespace(self):
301 """Check for whitespace at the ends of lines."""
303 if self.args.trailing_whitespace:
304 trailing = [i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
305 self.reporter.check(not trailing,
307 'Line(s) end with whitespace: {0}',
308 ', '.join([str(i) for i in trailing]))
311 def check_blockquote_classes(self):
312 """Check that all blockquotes have known classes."""
314 for node in self.find_all(self.doc, {'type' : 'blockquote'}):
315 cls = self.get_val(node, 'attr', 'class')
316 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
317 (self.filename, self.get_loc(node)),
318 'Unknown or missing blockquote type {0}',
322 def check_codeblock_classes(self):
323 """Check that all code blocks have known classes."""
325 for node in self.find_all(self.doc, {'type' : 'codeblock'}):
326 cls = self.get_val(node, 'attr', 'class')
327 self.reporter.check(cls in KNOWN_CODEBLOCKS,
328 (self.filename, self.get_loc(node)),
329 'Unknown or missing code block type {0}',
333 def find_all(self, node, pattern, accum=None):
334 """Find all matches for a pattern."""
336 assert type(pattern) == dict, 'Patterns must be dictionaries'
339 if self.match(node, pattern):
341 for child in node.get('children', []):
342 self.find_all(child, pattern, accum)
346 def match(self, node, pattern):
347 """Does this node match the given pattern?"""
356 elif type(val) == dict:
357 if not self.match(node[key], val):
362 def get_val(self, node, *chain):
363 """Get value one or more levels down."""
366 for selector in chain:
367 curr = curr.get(selector, None)
373 def get_loc(self, node):
374 """Convenience method to get node's line number."""
376 result = self.get_val(node, 'options', 'location')
377 if self.metadata_len is not None:
378 result += self.metadata_len
382 class CheckNonJekyll(CheckBase):
383 """Check a file that isn't translated by Jekyll."""
385 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
386 super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
389 def check_metadata(self):
390 self.reporter.check(self.metadata is None,
392 'Unexpected metadata')
395 class CheckIndex(CheckBase):
396 """Check the main index page."""
398 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
399 super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
400 self.layout = 'lesson'
403 class CheckEpisode(CheckBase):
404 """Check an episode page."""
406 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
407 super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
409 def check_metadata(self):
410 super(CheckEpisode, self).check_metadata()
412 if 'layout' in self.metadata:
413 if self.metadata['layout'] == 'break':
414 self.check_metadata_fields(BREAK_METADATA_FIELDS)
416 self.reporter.add(self.filename,
417 'Unknown episode layout "{0}"',
418 self.metadata['layout'])
420 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
423 def check_metadata_fields(self, expected):
424 for (name, type_) in expected:
425 if name not in self.metadata:
426 self.reporter.add(self.filename,
427 'Missing metadata field {0}',
429 elif type(self.metadata[name]) != type_:
430 self.reporter.add(self.filename,
431 '"{0}" has wrong type in metadata ({1} instead of {2})',
432 name, type(self.metadata[name]), type_)
435 class CheckReference(CheckBase):
436 """Check the reference page."""
438 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
439 super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
440 self.layout = 'reference'
443 class CheckGeneric(CheckBase):
444 """Check a generic page."""
446 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
447 super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
452 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
453 (re.compile(r'README\.md'), CheckNonJekyll),
454 (re.compile(r'index\.md'), CheckIndex),
455 (re.compile(r'reference\.md'), CheckReference),
456 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
457 (re.compile(r'.*\.md'), CheckGeneric)
461 if __name__ == '__main__':