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 # What kinds of blockquotes are allowed?
64 # What kinds of code fragments are allowed?
77 # What fields are required in teaching episode metadata?
78 TEACHING_METADATA_FIELDS = {
87 # What fields are required in break episode metadata?
88 BREAK_METADATA_FIELDS = {
94 # How long are lines allowed to be?
101 args.reporter = Reporter()
102 check_config(args.reporter, args.source_dir)
103 docs = read_all_markdown(args.source_dir, args.parser)
104 check_fileset(args.source_dir, args.reporter, docs.keys())
105 check_unwanted_files(args.source_dir, args.reporter)
106 for filename in docs.keys():
107 checker = create_checker(args, filename, docs[filename])
109 check_figures(args.source_dir, args.reporter)
110 args.reporter.report()
114 """Parse command-line arguments."""
116 parser = OptionParser()
117 parser.add_option('-l', '--linelen',
121 help='Check line lengths')
122 parser.add_option('-p', '--parser',
125 help='path to Markdown parser')
126 parser.add_option('-s', '--source',
129 help='source directory')
130 parser.add_option('-w', '--whitespace',
133 dest='trailing_whitespace',
134 help='Check for trailing whitespace')
136 args, extras = parser.parse_args()
137 require(args.parser is not None,
138 'Path to Markdown parser not provided')
140 'Unexpected trailing command-line arguments "{0}"'.format(extras))
145 def check_config(reporter, source_dir):
146 """Check configuration file."""
148 config_file = os.path.join(source_dir, '_config.yml')
149 config = load_yaml(config_file)
150 reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
151 reporter.check_field(config_file, 'configuration', config, 'carpentry', ('swc', 'dc'))
152 reporter.check_field(config_file, 'configuration', config, 'title')
153 reporter.check_field(config_file, 'configuration', config, 'email')
156 def read_all_markdown(source_dir, parser):
157 """Read source files, returning
158 {path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
161 all_dirs = [os.path.join(source_dir, d) for d in SOURCE_DIRS]
162 all_patterns = [os.path.join(d, '*.md') for d in all_dirs]
164 for pat in all_patterns:
165 for filename in glob.glob(pat):
166 data = read_markdown(parser, filename)
168 result[filename] = data
172 def check_fileset(source_dir, reporter, filenames_present):
173 """Are all required files present? Are extraneous files present?"""
175 # Check files with predictable names.
176 required = [p.replace('%', source_dir) for p in REQUIRED_FILES]
177 missing = set(required) - set(filenames_present)
179 reporter.add(None, 'Missing required file {0}', m)
181 # Check episode files' names.
183 for filename in filenames_present:
184 if '_episodes' not in filename:
186 m = P_EPISODE_FILENAME.search(filename)
188 seen.append(m.group(1))
190 reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
192 # Check for duplicate episode numbers.
193 reporter.check(len(seen) == len(set(seen)),
195 'Duplicate episode numbers {0} vs {1}',
196 sorted(seen), sorted(set(seen)))
198 # Check that numbers are consecutive.
199 seen = [int(s) for s in seen]
202 for i in range(len(seen) - 1):
203 clean = clean and ((seen[i+1] - seen[i]) == 1)
204 reporter.check(clean,
206 'Missing or non-consecutive episode numbers {0}',
210 def check_figures(source_dir, reporter):
211 """Check that all figures are present and referenced."""
215 all_figures_html = os.path.join(source_dir, '_includes', 'all_figures.html')
216 with open(all_figures_html, 'r') as reader:
218 figures = P_FIGURE_REFS.findall(text)
219 referenced = [os.path.split(f)[1] for f in figures if '/fig/' in f]
220 except FileNotFoundError as e:
221 reporter.add(all_figures_html,
225 # Get actual image files (ignore non-image files).
226 fig_dir_path = os.path.join(source_dir, 'fig')
227 actual = [f for f in os.listdir(fig_dir_path) if os.path.splitext(f)[1] in IMAGE_FILE_SUFFIX]
229 # Report differences.
230 unexpected = set(actual) - set(referenced)
231 reporter.check(not unexpected,
233 'Unexpected image files: {0}',
234 ', '.join(sorted(unexpected)))
235 missing = set(referenced) - set(actual)
236 reporter.check(not missing,
238 'Missing image files: {0}',
239 ', '.join(sorted(missing)))
242 def create_checker(args, filename, info):
243 """Create appropriate checker for file."""
245 for (pat, cls) in CHECKERS:
246 if pat.search(filename):
247 return cls(args, filename, **info)
250 class CheckBase(object):
251 """Base class for checking Markdown files."""
253 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
254 """Cache arguments for checking."""
256 super(CheckBase, self).__init__()
258 self.reporter = self.args.reporter # for convenience
259 self.filename = filename
260 self.metadata = metadata
261 self.metadata_len = metadata_len
270 """Run tests on metadata."""
272 self.check_metadata()
273 self.check_line_lengths()
274 self.check_trailing_whitespace()
275 self.check_blockquote_classes()
276 self.check_codeblock_classes()
279 def check_metadata(self):
280 """Check the YAML metadata."""
282 self.reporter.check(self.metadata is not None,
284 'Missing metadata entirely')
286 if self.metadata and (self.layout is not None):
287 self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
290 def check_line_lengths(self):
291 """Check the raw text of the lesson body."""
293 if self.args.line_lengths:
294 over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
295 self.reporter.check(not over,
297 'Line(s) are too long: {0}',
298 ', '.join([str(i) for i in over]))
301 def check_trailing_whitespace(self):
302 """Check for whitespace at the ends of lines."""
304 if self.args.trailing_whitespace:
305 trailing = [i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
306 self.reporter.check(not trailing,
308 'Line(s) end with whitespace: {0}',
309 ', '.join([str(i) for i in trailing]))
312 def check_blockquote_classes(self):
313 """Check that all blockquotes have known classes."""
315 for node in self.find_all(self.doc, {'type' : 'blockquote'}):
316 cls = self.get_val(node, 'attr', 'class')
317 self.reporter.check(cls in KNOWN_BLOCKQUOTES,
318 (self.filename, self.get_loc(node)),
319 'Unknown or missing blockquote type {0}',
323 def check_codeblock_classes(self):
324 """Check that all code blocks have known classes."""
326 for node in self.find_all(self.doc, {'type' : 'codeblock'}):
327 cls = self.get_val(node, 'attr', 'class')
328 self.reporter.check(cls in KNOWN_CODEBLOCKS,
329 (self.filename, self.get_loc(node)),
330 'Unknown or missing code block type {0}',
334 def find_all(self, node, pattern, accum=None):
335 """Find all matches for a pattern."""
337 assert type(pattern) == dict, 'Patterns must be dictionaries'
340 if self.match(node, pattern):
342 for child in node.get('children', []):
343 self.find_all(child, pattern, accum)
347 def match(self, node, pattern):
348 """Does this node match the given pattern?"""
357 elif type(val) == dict:
358 if not self.match(node[key], val):
363 def get_val(self, node, *chain):
364 """Get value one or more levels down."""
367 for selector in chain:
368 curr = curr.get(selector, None)
374 def get_loc(self, node):
375 """Convenience method to get node's line number."""
377 result = self.get_val(node, 'options', 'location')
378 if self.metadata_len is not None:
379 result += self.metadata_len
383 class CheckNonJekyll(CheckBase):
384 """Check a file that isn't translated by Jekyll."""
386 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
387 super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
390 def check_metadata(self):
391 self.reporter.check(self.metadata is None,
393 'Unexpected metadata')
396 class CheckIndex(CheckBase):
397 """Check the main index page."""
399 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
400 super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
401 self.layout = 'lesson'
404 class CheckEpisode(CheckBase):
405 """Check an episode page."""
407 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
408 super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
410 def check_metadata(self):
411 super(CheckEpisode, self).check_metadata()
413 if 'layout' in self.metadata:
414 if self.metadata['layout'] == 'break':
415 self.check_metadata_fields(BREAK_METADATA_FIELDS)
417 self.reporter.add(self.filename,
418 'Unknown episode layout "{0}"',
419 self.metadata['layout'])
421 self.check_metadata_fields(TEACHING_METADATA_FIELDS)
424 def check_metadata_fields(self, expected):
425 for (name, type_) in expected:
426 if name not in self.metadata:
427 self.reporter.add(self.filename,
428 'Missing metadata field {0}',
430 elif type(self.metadata[name]) != type_:
431 self.reporter.add(self.filename,
432 '"{0}" has wrong type in metadata ({1} instead of {2})',
433 name, type(self.metadata[name]), type_)
436 class CheckReference(CheckBase):
437 """Check the reference page."""
439 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
440 super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
441 self.layout = 'reference'
444 class CheckGeneric(CheckBase):
445 """Check a generic page."""
447 def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
448 super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
453 (re.compile(r'CONTRIBUTING\.md'), CheckNonJekyll),
454 (re.compile(r'README\.md'), CheckNonJekyll),
455 (re.compile(r'index\.md'), CheckIndex),
456 (re.compile(r'reference\.md'), CheckReference),
457 (re.compile(r'_episodes/.*\.md'), CheckEpisode),
458 (re.compile(r'.*\.md'), CheckGeneric)
462 if __name__ == '__main__':