Check lesson files and their contents.
"""
+from __future__ import print_function
import sys
import os
import glob
import json
-import yaml
import re
from optparse import OptionParser
-from util import Reporter, read_markdown
+from util import Reporter, read_markdown, load_yaml, check_unwanted_files, require, IMAGE_FILE_SUFFIX
__version__ = '0.2'
'%/setup.md': True,
}
-# Required non-Markdown files.
-NON_MARKDOWN_FILES = {
- "AUTHORS",
- "CITATION"
-}
-
# Episode filename pattern.
P_EPISODE_FILENAME = re.compile(r'/_episodes/(\d\d)-[-\w]+.md$')
+# Pattern to match lines ending with whitespace.
+P_TRAILING_WHITESPACE = re.compile(r'\s+$')
+
+# Pattern to match figure references in HTML.
+P_FIGURE_REFS = re.compile(r'<img[^>]+src="([^"]+)"[^>]*>')
+
+# Pattern to match internally-defined Markdown links.
+P_INTERNALLY_DEFINED_LINK = re.compile(r'\[[^\]]+\]\[[^\]]+\]')
+
# What kinds of blockquotes are allowed?
KNOWN_BLOCKQUOTES = {
'callout',
'source',
'bash',
'make',
+ 'matlab',
'python',
'r',
'sql'
"""Main driver."""
args = parse_args()
- args.reporter = Reporter(args)
- check_config(args)
- check_non_markdown_files(args.source_dir, args.reporter)
- docs = read_all_markdown(args, args.source_dir)
+ args.reporter = Reporter()
+ 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()
parser = OptionParser()
parser.add_option('-l', '--linelen',
default=False,
- dest='line_len',
+ action="store_true",
+ dest='line_lengths',
help='Check line lengths')
parser.add_option('-p', '--parser',
default=None,
default=os.curdir,
dest='source_dir',
help='source directory')
+ parser.add_option('-w', '--whitespace',
+ default=False,
+ action="store_true",
+ dest='trailing_whitespace',
+ help='Check for trailing whitespace')
args, extras = parser.parse_args()
require(args.parser is not None,
return args
-def check_config(args):
+def check_config(reporter, source_dir):
"""Check configuration file."""
- config_file = os.path.join(args.source_dir, '_config.yml')
- with open(config_file, 'r') as reader:
- config = yaml.load(reader)
-
- args.reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
-
+ 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')
-def check_non_markdown_files(source_dir, reporter):
- """Check presence of non-Markdown files."""
+ reporter.check({'values': {'root': '..'}} in config.get('defaults', []),
+ 'configuration',
+ '"root" not set to ".." in configuration')
- for filename in NON_MARKDOWN_FILES:
- path = os.path.join(source_dir, filename)
- reporter.check(os.path.exists(path),
- filename,
- "File not found")
-
-def read_all_markdown(args, source_dir):
+def read_all_markdown(source_dir, parser):
"""Read source files, returning
{path : {'metadata':yaml, 'metadata_len':N, 'text':text, 'lines':[(i, line, len)], 'doc':doc}}
"""
result = {}
for pat in all_patterns:
for filename in glob.glob(pat):
- data = read_markdown(args.parser, filename)
+ data = read_markdown(parser, filename)
if data:
result[filename] = data
return result
# Check for duplicate episode numbers.
reporter.check(len(seen) == len(set(seen)),
- None,
- 'Duplicate episode numbers {0} vs {1}',
- sorted(seen), sorted(set(seen)))
+ None,
+ 'Duplicate episode numbers {0} vs {1}',
+ sorted(seen), sorted(set(seen)))
# Check that numbers are consecutive.
seen = [int(s) for s in seen]
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 image files (ignore non-image files).
+ fig_dir_path = os.path.join(source_dir, 'fig')
+ actual = [f for f in os.listdir(fig_dir_path) if os.path.splitext(f)[1] in IMAGE_FILE_SUFFIX]
+
+ # 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."""
return cls(args, filename, **info)
-def require(condition, message):
- """Fail if condition not met."""
-
- if not condition:
- print(message, file=sys.stderr)
- sys.exit(1)
-
-
class CheckBase(object):
"""Base class for checking Markdown files."""
self.text = text
self.lines = lines
self.doc = doc
+
self.layout = None
"""Run tests on metadata."""
self.check_metadata()
- self.check_text()
+ self.check_line_lengths()
+ self.check_trailing_whitespace()
self.check_blockquote_classes()
self.check_codeblock_classes()
+ self.check_defined_link_references()
def check_metadata(self):
self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
- def check_text(self):
+ def check_line_lengths(self):
"""Check the raw text of the lesson body."""
- if self.args.line_len:
+ if self.args.line_lengths:
over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
self.reporter.check(not over,
self.filename,
', '.join([str(i) for i in over]))
+ def check_trailing_whitespace(self):
+ """Check for whitespace at the ends of lines."""
+
+ if self.args.trailing_whitespace:
+ trailing = [i for (i, l, n) in self.lines if P_TRAILING_WHITESPACE.match(l)]
+ self.reporter.check(not trailing,
+ self.filename,
+ 'Line(s) end with whitespace: {0}',
+ ', '.join([str(i) for i in trailing]))
+
+
def check_blockquote_classes(self):
"""Check that all blockquotes have known classes."""
cls)
+ def check_defined_link_references(self):
+ """Check that defined links resolve in the file.
+
+ Internally-defined links match the pattern [text][label]. If
+ the label contains '{{...}}', it is hopefully a references to
+ a configuration value - we should check that, but don't right
+ now.
+ """
+
+ result = set()
+ for node in self.find_all(self.doc, {'type' : 'text'}):
+ for match in P_INTERNALLY_DEFINED_LINK.findall(node['value']):
+ if '{{' not in match:
+ result.add(match)
+ self.reporter.check(not result,
+ self.filename,
+ 'Internally-defined links may be missing definitions: {0}',
+ ', '.join(sorted(result)))
+
+
def find_all(self, node, pattern, accum=None):
"""Find all matches for a pattern."""
super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
self.layout = 'lesson'
+ def check_metadata(self):
+ super(CheckIndex, self).check_metadata()
+ self.reporter.check(self.metadata.get('root', '') == '.',
+ self.filename,
+ 'Root not set to "."')
+
class CheckEpisode(CheckBase):
"""Check an episode page."""
def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
super(CheckEpisode, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
-
def check_metadata(self):
super(CheckEpisode, self).check_metadata()
if self.metadata: