import re
from optparse import OptionParser
-from util import Reporter, read_markdown, load_yaml
+from util import Reporter, read_markdown, load_yaml, check_unwanted_files
__version__ = '0.2'
# Pattern to match lines ending with whitespace.
P_TRAILING_WHITESPACE = re.compile(r'\s+$')
+# Pattern to match figure references in HTML.
+P_FIGURE_REFS = re.compile(r'<img[^>]+src="([^"]+)"[^>]*>')
+
# What kinds of blockquotes are allowed?
KNOWN_BLOCKQUOTES = {
'callout',
check_config(args.reporter, args.source_dir)
docs = read_all_markdown(args.source_dir, args.parser)
check_fileset(args.source_dir, args.reporter, docs.keys())
+ check_unwanted_files(args.source_dir, args.reporter)
for filename in docs.keys():
checker = create_checker(args, filename, docs[filename])
checker.check()
+ check_figures(args.source_dir, args.reporter)
args.reporter.report()
config_file = os.path.join(source_dir, '_config.yml')
config = load_yaml(config_file)
reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
+ reporter.check_field(config_file, 'configuration', config, 'carpentry', ('swc', 'dc'))
+ reporter.check_field(config_file, 'configuration', config, 'title')
+ reporter.check_field(config_file, 'configuration', config, 'email')
+ reporter.check_field(config_file, 'configuration', config, 'repo')
+ reporter.check_field(config_file, 'configuration', config, 'root')
+ if ('repo' in config) and ('root' in config):
+ reporter.check(config['repo'].endswith(config['root']),
+ config_file,
+ 'Repository name "{0}" not consistent with root "{1}"',
+ config['repo'], config['root'])
def read_all_markdown(source_dir, parser):
seen)
+def check_figures(source_dir, reporter):
+ """Check that all figures are present and referenced."""
+
+ # Get references.
+ try:
+ all_figures_html = os.path.join(source_dir, '_includes', 'all_figures.html')
+ with open(all_figures_html, 'r') as reader:
+ text = reader.read()
+ figures = P_FIGURE_REFS.findall(text)
+ referenced = [os.path.split(f)[1] for f in figures if '/fig/' in f]
+ except FileNotFoundError as e:
+ reporter.add(all_figures_html,
+ 'File not found')
+ return
+
+ # Get actual files.
+ fig_dir_path = os.path.join(source_dir, 'fig')
+ actual = [f for f in os.listdir(fig_dir_path) if not f.startswith('.')]
+
+ # Report differences.
+ unexpected = set(actual) - set(referenced)
+ reporter.check(not unexpected,
+ None,
+ 'Unexpected image files: {0}',
+ ', '.join(sorted(unexpected)))
+ missing = set(referenced) - set(actual)
+ reporter.check(not missing,
+ None,
+ 'Missing image files: {0}',
+ ', '.join(sorted(missing)))
+
+
def create_checker(args, filename, info):
"""Create appropriate checker for file."""