-#!/usr/bin/env python
+#!/usr/bin/env python3
"""
Check lesson files and their contents.
"""
-import sys
+
import os
import glob
-import json
import re
-from optparse import OptionParser
+from argparse import ArgumentParser
-from util import Reporter, read_markdown, load_yaml, check_unwanted_files, require, IMAGE_FILE_SUFFIX
+from util import (Reporter, read_markdown, load_yaml, check_unwanted_files,
+ require)
-__version__ = '0.2'
+__version__ = '0.3'
# Where to look for source Markdown files.
SOURCE_DIRS = ['', '_episodes', '_extras']
# FIXME: We do not yet validate whether any files have the required
# YAML headers, but should in the future.
# The '%' is replaced with the source directory path for checking.
-# Episodes are handled specially, and extra files in '_extras' are also handled specially.
-# This list must include all the Markdown files listed in the 'bin/initialize' script.
+# Episodes are handled specially, and extra files in '_extras' are also handled
+# specially. This list must include all the Markdown files listed in the
+# 'bin/initialize' script.
REQUIRED_FILES = {
'%/CONDUCT.md': True,
'%/CONTRIBUTING.md': False,
'%/LICENSE.md': True,
'%/README.md': False,
'%/_extras/discuss.md': True,
- '%/_extras/figures.md': True,
'%/_extras/guide.md': True,
'%/index.md': True,
'%/reference.md': True,
# 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',
'error',
'output',
'source',
- 'bash',
- 'make',
- 'matlab',
- 'python',
- 'r',
- 'sql'
+ 'language-bash',
+ 'html',
+ 'language-make',
+ 'language-matlab',
+ 'language-python',
+ 'language-r',
+ 'language-shell',
+ 'language-sql'
}
# What fields are required in teaching episode metadata?
# How long are lines allowed to be?
MAX_LINE_LEN = 100
+
def main():
"""Main driver."""
args = parse_args()
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_fileset(args.source_dir, args.reporter, list(docs.keys()))
check_unwanted_files(args.source_dir, args.reporter)
- for filename in docs.keys():
+ for filename in list(docs.keys()):
checker = create_checker(args, filename, docs[filename])
checker.check()
- check_figures(args.source_dir, args.reporter)
+
args.reporter.report()
+ if args.reporter.messages and not args.permissive:
+ exit(1)
def parse_args():
"""Parse command-line arguments."""
- parser = OptionParser()
- parser.add_option('-l', '--linelen',
- default=False,
- 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('-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()
+ parser = ArgumentParser(description="""Check episode files in a lesson.""")
+ parser.add_argument('-l', '--linelen',
+ default=False,
+ action="store_true",
+ dest='line_lengths',
+ help='Check line lengths')
+ parser.add_argument('-p', '--parser',
+ default=None,
+ dest='parser',
+ help='path to Markdown parser')
+ parser.add_argument('-r', '--references',
+ default=None,
+ dest='reference_path',
+ help='path to Markdown file of external references')
+ parser.add_argument('-s', '--source',
+ default=os.curdir,
+ dest='source_dir',
+ help='source directory')
+ parser.add_argument('-w', '--whitespace',
+ default=False,
+ action="store_true",
+ dest='trailing_whitespace',
+ help='Check for trailing whitespace')
+ parser.add_argument('--permissive',
+ default=False,
+ action="store_true",
+ dest='permissive',
+ help='Do not raise an error even if issues are detected')
+
+ args, extras = parser.parse_known_args()
require(args.parser is not None,
'Path to Markdown parser not provided')
require(not extras,
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, '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, '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'])
+
+ reporter.check({'values': {'root': '..'}} in config.get('defaults', []),
+ 'configuration',
+ '"root" not set to ".." in configuration')
+
+
+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):
if m and m.group(1):
seen.append(m.group(1))
else:
- reporter.add(None, 'Episode {0} has badly-formatted filename', filename)
+ reporter.add(
+ None, 'Episode {0} has badly-formatted filename', filename)
# 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.sort()
+ seen = sorted([int(s) for s in seen])
clean = True
for i in range(len(seen) - 1):
clean = clean and ((seen[i+1] - seen[i]) == 1)
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."""
for (pat, cls) in CHECKERS:
if pat.search(filename):
return cls(args, filename, **info)
+ return NotImplemented
-
-class CheckBase(object):
+class CheckBase:
"""Base class for checking Markdown files."""
def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
"""Cache arguments for checking."""
- super(CheckBase, self).__init__()
self.args = args
- self.reporter = self.args.reporter # for convenience
+ self.reporter = self.args.reporter # for convenience
self.filename = filename
self.metadata = metadata
self.metadata_len = metadata_len
self.layout = None
-
def check(self):
- """Run tests on metadata."""
+ """Run tests."""
self.check_metadata()
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):
"""Check the YAML metadata."""
'Missing metadata entirely')
if self.metadata and (self.layout is not None):
- self.reporter.check_field(self.filename, 'metadata', self.metadata, 'layout', self.layout)
-
+ self.reporter.check_field(
+ self.filename, 'metadata', self.metadata, 'layout', self.layout)
def check_line_lengths(self):
"""Check the raw text of the lesson body."""
if self.args.line_lengths:
- over = [i for (i, l, n) in self.lines if (n > MAX_LINE_LEN) and (not l.startswith('!'))]
+ 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,
'Line(s) are too long: {0}',
', '.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)]
+ 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."""
- for node in self.find_all(self.doc, {'type' : 'blockquote'}):
+ for node in self.find_all(self.doc, {'type': 'blockquote'}):
cls = self.get_val(node, 'attr', 'class')
self.reporter.check(cls in KNOWN_BLOCKQUOTES,
(self.filename, self.get_loc(node)),
'Unknown or missing blockquote type {0}',
cls)
-
def check_codeblock_classes(self):
"""Check that all code blocks have known classes."""
- for node in self.find_all(self.doc, {'type' : 'codeblock'}):
+ for node in self.find_all(self.doc, {'type': 'codeblock'}):
cls = self.get_val(node, 'attr', 'class')
self.reporter.check(cls in KNOWN_CODEBLOCKS,
(self.filename, self.get_loc(node)),
'Unknown or missing code block type {0}',
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."""
- assert type(pattern) == dict, 'Patterns must be dictionaries'
+ assert isinstance(pattern, dict), 'Patterns must be dictionaries'
if accum is None:
accum = []
if self.match(node, pattern):
self.find_all(child, pattern, accum)
return accum
-
def match(self, node, pattern):
"""Does this node match the given pattern?"""
if key not in node:
return False
val = pattern[key]
- if type(val) == str:
+ if isinstance(val, str):
if node[key] != val:
return False
- elif type(val) == dict:
+ elif isinstance(val, dict):
if not self.match(node[key], val):
return False
return True
-
- def get_val(self, node, *chain):
+ @staticmethod
+ def get_val(node, *chain):
"""Get value one or more levels down."""
curr = node
break
return curr
-
def get_loc(self, node):
"""Convenience method to get node's line number."""
class CheckNonJekyll(CheckBase):
"""Check a file that isn't translated by Jekyll."""
- def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
- super(CheckNonJekyll, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
-
-
def check_metadata(self):
self.reporter.check(self.metadata is None,
self.filename,
"""Check the main index page."""
def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
- super(CheckIndex, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
+ super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
self.layout = 'lesson'
+ def check_metadata(self):
+ super().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().check()
+ self.check_reference_inclusion()
def check_metadata(self):
- super(CheckEpisode, self).check_metadata()
+ super().check_metadata()
if self.metadata:
if 'layout' in self.metadata:
if self.metadata['layout'] == 'break':
else:
self.check_metadata_fields(TEACHING_METADATA_FIELDS)
-
def check_metadata_fields(self, expected):
+ """Check metadata fields."""
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_:
+ elif not isinstance(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):
"""Check the reference page."""
def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
- super(CheckReference, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
+ super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
self.layout = 'reference'
"""Check a generic page."""
def __init__(self, args, filename, metadata, metadata_len, text, lines, doc):
- super(CheckGeneric, self).__init__(args, filename, metadata, metadata_len, text, lines, doc)
+ super().__init__(args, filename, metadata, metadata_len, text, lines, doc)
self.layout = 'page'
projects / rnaseq-cwl-training.git / blobdiff