X-Git-Url: https://git.arvados.org/rnaseq-cwl-training.git/blobdiff_plain/570451a88f4071708bd73e83f0f6f42bdc6a5dfc..68c149376d66d135a98c41e382800f051aee50b3:/bin/lesson_check.py

diff --git a/bin/lesson_check.py b/bin/lesson_check.py
index 6de7714..7bc09a3 100755
--- a/bin/lesson_check.py
+++ b/bin/lesson_check.py
@@ -4,6 +4,7 @@
 Check lesson files and their contents.
 """
 
+from __future__ import print_function
 import sys
 import os
 import glob
@@ -11,7 +12,7 @@ import json
 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, require, IMAGE_FILE_SUFFIX
 
 __version__ = '0.2'
 
@@ -46,6 +47,9 @@ 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',
@@ -67,6 +71,7 @@ KNOWN_CODEBLOCKS = {
     'source',
     'bash',
     'make',
+    'matlab',
     'python',
     'r',
     'sql'
@@ -100,6 +105,7 @@ def main():
     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()
@@ -145,6 +151,9 @@ def check_config(reporter, source_dir):
     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 read_all_markdown(source_dir, parser):
@@ -216,9 +225,9 @@ def check_figures(source_dir, reporter):
                      'File not found')
         return
 
-    # Get actual files.
+    # 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 not f.startswith('.')]
+    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)
@@ -241,14 +250,6 @@ def create_checker(args, filename, info):
             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."""
 
@@ -276,6 +277,7 @@ class CheckBase(object):
         self.check_trailing_whitespace()
         self.check_blockquote_classes()
         self.check_codeblock_classes()
+        self.check_defined_link_references()
 
 
     def check_metadata(self):
@@ -333,6 +335,26 @@ class CheckBase(object):
                                 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."""