Merge pull request #318 from gkreitz/markdown_image_fixes

Improve image handling in markdown statements

Author: pehrsoderman

problemtools/md2html.py

@@ -1,6 +1,7 @@
 #! /usr/bin/env python3
 # -*- coding: utf-8 -*-
 import argparse
+import hashlib
 import html
 import os
 from pathlib import Path
@@ -15,7 +16,7 @@
 
 
 def convert(problem_root: Path, options: argparse.Namespace, statement_file: Path) -> bool:
-    """Convert a Markdown statement to HTML
+    """Convert a Markdown statement to HTML. Writes output to current working directory.
 
     Args:
         problem: path to problem directory
@@ -85,7 +86,7 @@ def is_fn_id(s):
     allowed_classes = ('sample', 'problemheader', 'problembody', 'sampleinteractionwrite', 'sampleinteractionread')
 
     # Annoying: nh3 will ignore exceptions in attribute_filter
-    image_fail_reason: str | None = None
+    image_fail_reason: list[Exception] = []
 
     def attribute_filter(tag, attribute, value):
         if attribute == 'class' and value in allowed_classes:
@@ -103,10 +104,9 @@ def attribute_filter(tag, attribute, value):
                 statement_util.assert_image_is_valid(statement_dir, value)
             except Exception as e:
                 nonlocal image_fail_reason
-                image_fail_reason = str(e)
+                image_fail_reason.append(e)
                 return None
-            copy_image(statement_dir, value)
-            return value
+            return copy_image(statement_dir, value)
         return None
 
     statement_html = nh3.clean(
@@ -126,22 +126,25 @@ def attribute_filter(tag, attribute, value):
     )
 
     if image_fail_reason:
-        assert isinstance(image_fail_reason, str)
-        if 'Unsupported' in image_fail_reason:
-            raise ValueError(image_fail_reason)
-        raise FileNotFoundError(image_fail_reason)
+        # We don't have a great way to emit multiple errors from here, so just re-raise the first error
+        raise image_fail_reason[0]
 
     return statement_html
 
 
-def copy_image(statement_dir: Path, img_src: str) -> None:
-    """Copy image to output directory
+def copy_image(statement_dir: Path, img_src: str) -> str:
+    """Copy image to working directory (with new filename) and returns the new filename
 
     Args:
         statement_dir: the directory with problem statement files
         img_src: the image source as in the Markdown statement
     """
 
-    if os.path.isfile(img_src):  # already copied
-        return
-    shutil.copyfile(statement_dir / img_src, img_src)
+    # We rename to sha256 of contents, and preserve the suffix. This flattens
+    # the directory structure to a single folders in a simple way.
+    with open(statement_dir / img_src, 'rb') as f:
+        filename = hashlib.file_digest(f, 'sha256').hexdigest() + Path(img_src).suffix
+
+    if not os.path.isfile(filename):  # check if already copied
+        shutil.copyfile(statement_dir / img_src, filename)
+    return filename

problemtools/statement_util.py

@@ -7,6 +7,7 @@
 import tempfile
 from pathlib import Path
 from typing import Optional, List, Tuple
+from urllib.parse import urlparse
 
 from . import metadata
 from .formatversion import FormatVersion, get_format_version
@@ -106,10 +107,16 @@ def foreach_image(statement_path: Path, callback):
 
 def assert_image_is_valid(statement_dir: Path, img_src: str) -> None:
     """Check that the image exists and uses an allowed extension"""
-    extension = Path(img_src).suffix
+    img_path = Path(img_src)
+    extension = img_path.suffix
     # TODO: fix svg sanitization and allow svg
     if extension not in ALLOWED_IMAGE_EXTENSIONS:
         raise ValueError(f'Unsupported image extension {extension} for image {img_src}')
+    if img_path.is_absolute():
+        raise ValueError(f'Image path must be relative, but {img_src} is not.')
+    as_url = urlparse(img_src)
+    if as_url.scheme:
+        raise ValueError(f'Image path must not be an URL with a scheme, but {img_src} is.')
 
     source_file = statement_dir / img_src
     if not source_file.exists():

tests/problems/imgrequest2/statement/problem.en.md

@@ -1,3 +1,3 @@
 Make web request via image
 
-<img src="http:picsum.photos/400">
+<img src="https:open.kattis.com/images/site/header/logo-empty.png">