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'
+__version__ = '0.3'
# Where to look for source Markdown files.
SOURCE_DIRS = ['', '_episodes', '_extras']
# 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_INTERNAL_LINK_REF = re.compile(r'\[([^\]]+)\]\[([^\]]+)\]')
+
+# Pattern to match reference links (to resolve internally-defined references).
+P_INTERNAL_LINK_DEF = re.compile(r'^\[([^\]]+)\]:\s*(.+)')
+
# What kinds of blockquotes are allowed?
KNOWN_BLOCKQUOTES = {
'callout',
'source',
'bash',
'make',
+ 'matlab',
'python',
'r',
'sql'
}
-# What fields are required in episode metadata?
-EPISODE_METADATA_FIELDS = {
+# What fields are required in teaching episode metadata?
+TEACHING_METADATA_FIELDS = {
('title', str),
('teaching', int),
('exercises', int),
('keypoints', list)
}
+# What fields are required in break episode metadata?
+BREAK_METADATA_FIELDS = {
+ ('layout', str),
+ ('title', str),
+ ('break', int)
+}
+
# How long are lines allowed to be?
MAX_LINE_LEN = 100
"""Main driver."""
args = parse_args()
- args.reporter = Reporter(args)
- check_config(args)
- docs = read_all_markdown(args, args.source_dir)
+ args.reporter = Reporter()
+ check_config(args.reporter, args.source_dir)
+ args.references = read_references(args.reporter, args.reference_path)
+
+ 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,
dest='parser',
help='path to Markdown parser')
+ parser.add_option('-r', '--references',
+ default=None,
+ dest='reference_path',
+ help='path to Markdown file of external references')
parser.add_option('-s', '--source',
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)
+ 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', 'lc'))
+ reporter.check_field(config_file, 'configuration', config, 'title')
+ reporter.check_field(config_file, 'configuration', config, 'contact')
- args.reporter.check_field(config_file, 'configuration', config, 'kind', 'lesson')
+ reporter.check({'values': {'root': '..'}} in config.get('defaults', []),
+ 'configuration',
+ '"root" not set to ".." in configuration')
-def read_all_markdown(args, source_dir):
+def read_references(reporter, ref_path):
+ """Read shared file of reference links, returning dictionary of valid references
+ {symbolic_name : URL}
+ """
+
+ result = {}
+ urls_seen = set()
+ if ref_path:
+ with open(ref_path, 'r') as reader:
+ for (num, line) in enumerate(reader):
+ line_num = num + 1
+ m = P_INTERNAL_LINK_DEF.search(line)
+ require(m,
+ '{0}:{1} not valid reference:\n{2}'.format(ref_path, line_num, line.rstrip()))
+ name = m.group(1)
+ url = m.group(2)
+ require(name,
+ 'Empty reference at {0}:{1}'.format(ref_path, line_num))
+ reporter.check(name not in result,
+ ref_path,
+ 'Duplicate reference {0} at line {1}',
+ name, line_num)
+ reporter.check(url not in urls_seen,
+ ref_path,
+ 'Duplicate definition of URL {0} at line {1}',
+ url, line_num)
+ result[name] = url
+ urls_seen.add(url)
+ return result
+
+
+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
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."""
def check(self):
- """Run tests on metadata."""
+ """Run tests."""
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].
+ """
+
+ result = set()
+ for node in self.find_all(self.doc, {'type' : 'text'}):
+ for match in P_INTERNAL_LINK_REF.findall(node['value']):
+ text = match[0]
+ link = match[1]
+ if link not in self.args.references:
+ result.add('"{0}"=>"{1}"'.format(text, link))
+ 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(self):
+ """Run extra tests."""
+
+ super(CheckEpisode, self).check()
+ self.check_reference_inclusion()
+
+
def check_metadata(self):
super(CheckEpisode, self).check_metadata()
if self.metadata:
- for (name, type_) in EPISODE_METADATA_FIELDS:
- self.reporter.check(type(self.metadata.get(name, None)) == type_,
- self.filename,
- '"{0}" missing, empty, or has wrong type in metadata',
- name)
+ if 'layout' in self.metadata:
+ if self.metadata['layout'] == 'break':
+ self.check_metadata_fields(BREAK_METADATA_FIELDS)
+ else:
+ self.reporter.add(self.filename,
+ 'Unknown episode layout "{0}"',
+ self.metadata['layout'])
+ else:
+ self.check_metadata_fields(TEACHING_METADATA_FIELDS)
+
+
+ def check_metadata_fields(self, expected):
+ for (name, type_) in expected:
+ if name not in self.metadata:
+ self.reporter.add(self.filename,
+ 'Missing metadata field {0}',
+ name)
+ elif type(self.metadata[name]) != type_:
+ self.reporter.add(self.filename,
+ '"{0}" has wrong type in metadata ({1} instead of {2})',
+ name, type(self.metadata[name]), type_)
+
+
+ def check_reference_inclusion(self):
+ """Check that links file has been included."""
+
+ if not self.args.reference_path:
+ return
+
+ for (i, last_line, line_len) in reversed(self.lines):
+ if last_line:
+ break
+
+ require(last_line,
+ 'No non-empty lines in {0}'.format(self.filename))
+
+ include_filename = os.path.split(self.args.reference_path)[-1]
+ if include_filename not in last_line:
+ self.reporter.add(self.filename,
+ 'episode does not include "{0}"',
+ include_filename)
class CheckReference(CheckBase):