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'))
155 reporter.check_field(config_file, 'configuration', config, 'title')
156 reporter.check_field(config_file, 'configuration', config, 'email')
159 def read_all_markdown(source_dir, parser):
160 """Read source files, returning
161 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
164 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
165 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
167 for pat in all_patterns:
168 for filename in glob.glob(pat):
169 data = read_markdown(parser, filename)
171 result[filename] = data
175 def check_fileset(source_dir, reporter, filenames_present):
176 """Are all required files present? Are extraneous files present?"""
178 # Check files with predictable names.
179 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
180 missing = set(required) - set(filenames_present)
182 reporter.add(None, 'Missing required file {0}', m)
184 # Check episode files' names.
186 for filename in filenames_present:
187 if '_episodes' not in filename:
189 m = P_EPISODE_FILENAME.search(filename)
191 seen.append(m.group(1))
193 reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
195 # Check for duplicate episode numbers.
196 reporter.check(len(seen) == len(set(seen)),
198 'Duplicate episode numbers {0} vs {1}',
199 sorted(seen), sorted(set(seen)))
201 # Check that numbers are consecutive.
202 seen = [int(s) for s in seen]
205 for i in range(len(seen) - 1):
206 clean = clean and ((seen[i+1] - seen[i]) == 1)
207 reporter.check(clean,
209 'Missing or non-consecutive episode numbers {0}',
213 def check_figures(source_dir, reporter):
214 """Check that all figures are present and referenced."""
218 all_figures_html = os.path.join(source_dir, '_includes', 'all_figures.html')
219 with open(all_figures_html, 'r') as reader:
221 figures = P_FIGURE_REFS.findall(text)
222 referenced = [os.path.split(f)[1] for f in figures if '/fig/' in f]
223 except FileNotFoundError as e:
224 reporter.add(all_figures_html,
228 # Get actual image files (ignore non-image files).
229 fig_dir_path = os.path.join(source_dir, 'fig')
230 actual = [f for f in os.listdir(fig_dir_path) if os.path.splitext(f)[1] in IMAGE_FILE_SUFFIX]
232 # Report differences.
233 unexpected = set(actual) - set(referenced)
234 reporter.check(not unexpected,
236 'Unexpected image files: {0}',
237 ', '.join(sorted(unexpected)))
238 missing = set(referenced) - set(actual)
239 reporter.check(not missing,
241 'Missing image files: {0}',
242 ', '.join(sorted(missing)))
245 def create_checker(args, filename, info):
246 """Create appropriate checker for file."""
248 for (pat, cls) in CHECKERS:
249 if pat.search(filename):
250 return cls(args, filename, **info)
253 class CheckBase(object):
254 """Base class for checking Markdown files."""
256 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
257 """Cache arguments for checking."""
259 super(CheckBase, self).__init__()
261 self.reporter = self.args.reporter # for convenience
262 self.filename = filename
263 self.metadata = metadata
264 self.metadata_len = metadata_len
273 """Run tests on metadata."""
275 self.check_metadata()
276 self.check_line_lengths()
277 self.check_trailing_whitespace()
278 self.check_blockquote_classes()
279 self.check_codeblock_classes()
280 self.check_defined_link_references()
283 def check_metadata(self):
284 """Check the YAML metadata."""
286 self.reporter.check(self.metadata is not None,
288 'Missing metadata entirely')
290 if self.metadata and (self.layout is not None):
291 self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
294 def check_line_lengths(self):
295 """Check the raw text of the lesson body."""
297 if self.args.line_lengths:
298 over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
299 self.reporter.check(not over,
301 'Line(s) are too long: {0}',
302 ', '.join([str(i) for i in over]))
305 def check_trailing_whitespace(self):
306 """Check for whitespace at the ends of lines."""
308 if self.args.trailing_whitespace:
309 trailing = [i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
310 self.reporter.check(not trailing,
312 'Line(s) end with whitespace: {0}',
313 ', '.join([str(i) for i in trailing]))
316 def check_blockquote_classes(self):
317 """Check that all blockquotes have known classes."""
319 for node in self.find_all(self.doc, {'type' : 'blockquote'}):
320 cls = self.get_val(node, 'attr', 'class')
321 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
322 (self.filename, self.get_loc(node)),
323 'Unknown or missing blockquote type {0}',
327 def check_codeblock_classes(self):
328 """Check that all code blocks have known classes."""
330 for node in self.find_all(self.doc, {'type' : 'codeblock'}):
331 cls = self.get_val(node, 'attr', 'class')
332 self.reporter.check(cls in KNOWN_CODEBLOCKS,
333 (self.filename, self.get_loc(node)),
334 'Unknown or missing code block type {0}',
338 def check_defined_link_references(self):
339 """Check that defined links resolve in the file.
341 Internally-defined links match the pattern [text][label]. If
342 the label contains '{{...}}', it is hopefully a references to
343 a configuration value - we should check that, but don't right
348 for node in self.find_all(self.doc, {'type' : 'text'}):
349 for match in P_INTERNALLY_DEFINED_LINK.findall(node['value']):
350 if '{{' not in match:
352 self.reporter.check(not result,
354 'Internally-defined links may be missing definitions: {0}',
355 ', '.join(sorted(result)))
358 def find_all(self, node, pattern, accum=None):
359 """Find all matches for a pattern."""
361 assert type(pattern) == dict, 'Patterns must be dictionaries'
364 if self.match(node, pattern):
366 for child in node.get('children', []):
367 self.find_all(child, pattern, accum)
371 def match(self, node, pattern):
372 """Does this node match the given pattern?"""
381 elif type(val) == dict:
382 if not self.match(node[key], val):
387 def get_val(self, node, *chain):
388 """Get value one or more levels down."""
391 for selector in chain:
392 curr = curr.get(selector, None)
398 def get_loc(self, node):
399 """Convenience method to get node's line number."""
401 result = self.get_val(node, 'options', 'location')
402 if self.metadata_len is not None:
403 result += self.metadata_len
407 class CheckNonJekyll(CheckBase):
408 """Check a file that isn't translated by Jekyll."""
410 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
411 super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
414 def check_metadata(self):
415 self.reporter.check(self.metadata is None,
417 'Unexpected metadata')
420 class CheckIndex(CheckBase):
421 """Check the main index page."""
423 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
424 super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
425 self.layout = 'lesson'
428 class CheckEpisode(CheckBase):
429 """Check an episode page."""
431 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
432 super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
434 def check_metadata(self):
435 super(CheckEpisode, self).check_metadata()
437 if 'layout' in self.metadata:
438 if self.metadata['layout'] == 'break':
439 self.check_metadata_fields(BREAK_METADATA_FIELDS)
441 self.reporter.add(self.filename,
442 'Unknown episode layout "{0}"',
443 self.metadata['layout'])
445 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
448 def check_metadata_fields(self, expected):
449 for (name, type_) in expected:
450 if name not in self.metadata:
451 self.reporter.add(self.filename,
452 'Missing metadata field {0}',
454 elif type(self.metadata[name]) != type_:
455 self.reporter.add(self.filename,
456 '"{0}" has wrong type in metadata ({1} instead of {2})',
457 name, type(self.metadata[name]), type_)
460 class CheckReference(CheckBase):
461 """Check the reference page."""
463 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
464 super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
465 self.layout = 'reference'
468 class CheckGeneric(CheckBase):
469 """Check a generic page."""
471 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
472 super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
477 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
478 (re.compile(r'README\.md'), CheckNonJekyll),
479 (re.compile(r'index\.md'), CheckIndex),
480 (re.compile(r'reference\.md'), CheckReference),
481 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
482 (re.compile(r'.*\.md'), CheckGeneric)
486 if __name__ == '__main__':