4 Check lesson files and their contents.
7 from __future__ import print_function
13 from optparse import OptionParser
15 from util import Reporter, read_markdown, load_yaml, check_unwanted_files, require, IMAGE_FILE_SUFFIX
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 # Episode filename pattern.
42 P_EPISODE_FILENAME = re.compile(r'/_episodes/(\d\d)-[-\w]+.md$')
44 # Pattern to match lines ending with whitespace.
45 P_TRAILING_WHITESPACE = re.compile(r'\s+$')
47 # Pattern to match figure references in HTML.
48 P_FIGURE_REFS = re.compile(r'<img[^>]+src="([^"]+)"[^>]*>')
50 # Pattern to match internally-defined Markdown links.
51 P_INTERNALLY_DEFINED_LINK = re.compile(r'\[[^\]]+\]\[[^\]]+\]')
53 # What kinds of blockquotes are allowed?
67 # What kinds of code fragments are allowed?
80 # What fields are required in teaching episode metadata?
81 TEACHING_METADATA_FIELDS = {
90 # What fields are required in break episode metadata?
91 BREAK_METADATA_FIELDS = {
97 # How long are lines allowed to be?
104 args.reporter = Reporter()
105 check_config(args.reporter, args.source_dir)
106 docs = read_all_markdown(args.source_dir, args.parser)
107 check_fileset(args.source_dir, args.reporter, docs.keys())
108 check_unwanted_files(args.source_dir, args.reporter)
109 for filename in docs.keys():
110 checker = create_checker(args, filename, docs[filename])
112 check_figures(args.source_dir, args.reporter)
113 args.reporter.report()
117 """Parse command-line arguments."""
119 parser = OptionParser()
120 parser.add_option('-l', '--linelen',
124 help='Check line lengths')
125 parser.add_option('-p', '--parser',
128 help='path to Markdown parser')
129 parser.add_option('-s', '--source',
132 help='source directory')
133 parser.add_option('-w', '--whitespace',
136 dest='trailing_whitespace',
137 help='Check for trailing whitespace')
139 args, extras = parser.parse_args()
140 require(args.parser is not None,
141 'Path to Markdown parser not provided')
143 'Unexpected trailing command-line arguments "{0}"'.format(extras))
148 def check_config(reporter, source_dir):
149 """Check configuration file."""
151 config_file = os.path.join(source_dir, '_config.yml')
152 config = load_yaml(config_file)
153 reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
154 reporter.check_field(config_file, 'configuration', config, 'carpentry', ('swc', 'dc', 'lc'))
155 reporter.check_field(config_file, 'configuration', config, 'title')
156 reporter.check_field(config_file, 'configuration', config, 'contact')
158 reporter.check({'values': {'root': '..'}} in config.get('defaults', []),
160 '"root" not set to ".." in configuration')
163 def read_all_markdown(source_dir, parser):
164 """Read source files, returning
165 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
168 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
169 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
171 for pat in all_patterns:
172 for filename in glob.glob(pat):
173 data = read_markdown(parser, filename)
175 result[filename] = data
179 def check_fileset(source_dir, reporter, filenames_present):
180 """Are all required files present? Are extraneous files present?"""
182 # Check files with predictable names.
183 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
184 missing = set(required) - set(filenames_present)
186 reporter.add(None, 'Missing required file {0}', m)
188 # Check episode files' names.
190 for filename in filenames_present:
191 if '_episodes' not in filename:
193 m = P_EPISODE_FILENAME.search(filename)
195 seen.append(m.group(1))
197 reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
199 # Check for duplicate episode numbers.
200 reporter.check(len(seen) == len(set(seen)),
202 'Duplicate episode numbers {0} vs {1}',
203 sorted(seen), sorted(set(seen)))
205 # Check that numbers are consecutive.
206 seen = [int(s) for s in seen]
209 for i in range(len(seen) - 1):
210 clean = clean and ((seen[i+1] - seen[i]) == 1)
211 reporter.check(clean,
213 'Missing or non-consecutive episode numbers {0}',
217 def check_figures(source_dir, reporter):
218 """Check that all figures are present and referenced."""
222 all_figures_html = os.path.join(source_dir, '_includes', 'all_figures.html')
223 with open(all_figures_html, 'r') as reader:
225 figures = P_FIGURE_REFS.findall(text)
226 referenced = [os.path.split(f)[1] for f in figures if '/fig/' in f]
227 except FileNotFoundError as e:
228 reporter.add(all_figures_html,
232 # Get actual image files (ignore non-image files).
233 fig_dir_path = os.path.join(source_dir, 'fig')
234 actual = [f for f in os.listdir(fig_dir_path) if os.path.splitext(f)[1] in IMAGE_FILE_SUFFIX]
236 # Report differences.
237 unexpected = set(actual) - set(referenced)
238 reporter.check(not unexpected,
240 'Unexpected image files: {0}',
241 ', '.join(sorted(unexpected)))
242 missing = set(referenced) - set(actual)
243 reporter.check(not missing,
245 'Missing image files: {0}',
246 ', '.join(sorted(missing)))
249 def create_checker(args, filename, info):
250 """Create appropriate checker for file."""
252 for (pat, cls) in CHECKERS:
253 if pat.search(filename):
254 return cls(args, filename, **info)
257 class CheckBase(object):
258 """Base class for checking Markdown files."""
260 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
261 """Cache arguments for checking."""
263 super(CheckBase, self).__init__()
265 self.reporter = self.args.reporter # for convenience
266 self.filename = filename
267 self.metadata = metadata
268 self.metadata_len = metadata_len
277 """Run tests on metadata."""
279 self.check_metadata()
280 self.check_line_lengths()
281 self.check_trailing_whitespace()
282 self.check_blockquote_classes()
283 self.check_codeblock_classes()
284 self.check_defined_link_references()
287 def check_metadata(self):
288 """Check the YAML metadata."""
290 self.reporter.check(self.metadata is not None,
292 'Missing metadata entirely')
294 if self.metadata and (self.layout is not None):
295 self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
298 def check_line_lengths(self):
299 """Check the raw text of the lesson body."""
301 if self.args.line_lengths:
302 over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
303 self.reporter.check(not over,
305 'Line(s) are too long: {0}',
306 ', '.join([str(i) for i in over]))
309 def check_trailing_whitespace(self):
310 """Check for whitespace at the ends of lines."""
312 if self.args.trailing_whitespace:
313 trailing = [i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
314 self.reporter.check(not trailing,
316 'Line(s) end with whitespace: {0}',
317 ', '.join([str(i) for i in trailing]))
320 def check_blockquote_classes(self):
321 """Check that all blockquotes have known classes."""
323 for node in self.find_all(self.doc, {'type' : 'blockquote'}):
324 cls = self.get_val(node, 'attr', 'class')
325 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
326 (self.filename, self.get_loc(node)),
327 'Unknown or missing blockquote type {0}',
331 def check_codeblock_classes(self):
332 """Check that all code blocks have known classes."""
334 for node in self.find_all(self.doc, {'type' : 'codeblock'}):
335 cls = self.get_val(node, 'attr', 'class')
336 self.reporter.check(cls in KNOWN_CODEBLOCKS,
337 (self.filename, self.get_loc(node)),
338 'Unknown or missing code block type {0}',
342 def check_defined_link_references(self):
343 """Check that defined links resolve in the file.
345 Internally-defined links match the pattern [text][label]. If
346 the label contains '{{...}}', it is hopefully a references to
347 a configuration value - we should check that, but don't right
352 for node in self.find_all(self.doc, {'type' : 'text'}):
353 for match in P_INTERNALLY_DEFINED_LINK.findall(node['value']):
354 if '{{' not in match:
356 self.reporter.check(not result,
358 'Internally-defined links may be missing definitions: {0}',
359 ', '.join(sorted(result)))
362 def find_all(self, node, pattern, accum=None):
363 """Find all matches for a pattern."""
365 assert type(pattern) == dict, 'Patterns must be dictionaries'
368 if self.match(node, pattern):
370 for child in node.get('children', []):
371 self.find_all(child, pattern, accum)
375 def match(self, node, pattern):
376 """Does this node match the given pattern?"""
385 elif type(val) == dict:
386 if not self.match(node[key], val):
391 def get_val(self, node, *chain):
392 """Get value one or more levels down."""
395 for selector in chain:
396 curr = curr.get(selector, None)
402 def get_loc(self, node):
403 """Convenience method to get node's line number."""
405 result = self.get_val(node, 'options', 'location')
406 if self.metadata_len is not None:
407 result += self.metadata_len
411 class CheckNonJekyll(CheckBase):
412 """Check a file that isn't translated by Jekyll."""
414 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
415 super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
418 def check_metadata(self):
419 self.reporter.check(self.metadata is None,
421 'Unexpected metadata')
424 class CheckIndex(CheckBase):
425 """Check the main index page."""
427 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
428 super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
429 self.layout = 'lesson'
431 def check_metadata(self):
432 super(CheckIndex, self).check_metadata()
433 self.reporter.check(self.metadata.get('root', '') == '.',
435 'Root not set to "."')
438 class CheckEpisode(CheckBase):
439 """Check an episode page."""
441 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
442 super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
444 def check_metadata(self):
445 super(CheckEpisode, self).check_metadata()
447 if 'layout' in self.metadata:
448 if self.metadata['layout'] == 'break':
449 self.check_metadata_fields(BREAK_METADATA_FIELDS)
451 self.reporter.add(self.filename,
452 'Unknown episode layout "{0}"',
453 self.metadata['layout'])
455 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
458 def check_metadata_fields(self, expected):
459 for (name, type_) in expected:
460 if name not in self.metadata:
461 self.reporter.add(self.filename,
462 'Missing metadata field {0}',
464 elif type(self.metadata[name]) != type_:
465 self.reporter.add(self.filename,
466 '"{0}" has wrong type in metadata ({1} instead of {2})',
467 name, type(self.metadata[name]), type_)
470 class CheckReference(CheckBase):
471 """Check the reference page."""
473 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
474 super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
475 self.layout = 'reference'
478 class CheckGeneric(CheckBase):
479 """Check a generic page."""
481 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
482 super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
487 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
488 (re.compile(r'README\.md'), CheckNonJekyll),
489 (re.compile(r'index\.md'), CheckIndex),
490 (re.compile(r'reference\.md'), CheckReference),
491 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
492 (re.compile(r'.*\.md'), CheckGeneric)
496 if __name__ == '__main__':