Update Mistune and add escape parameter

md2cf/confluence_renderer.py

@@ -60,7 +60,7 @@ def append(self, child):
         self.children.append(child)
 
 
-class ConfluenceRenderer(mistune.Renderer):
+class ConfluenceRenderer(mistune.HTMLRenderer):
     def __init__(
         self,
         strip_header=False,
@@ -81,14 +81,14 @@ def reinit(self):
         self.relative_links = list()
         self.title = None
 
-    def header(self, text, level, raw=None):
+    def heading(self, text, level, raw=None):
         if self.title is None and level == 1:
             self.title = text
             # Don't duplicate page title as a header
             if self.strip_header:
                 return ""
 
-        return super(ConfluenceRenderer, self).header(text, level, raw=raw)
+        return super(ConfluenceRenderer, self).heading(text, level, raw=raw)
 
     def structured_macro(self, name):
         return ConfluenceTag("structured-macro", attrib={"name": name})
@@ -103,8 +103,8 @@ def plain_text_body(self, text):
         body_tag.text = text
         return body_tag
 
-    def link(self, link, title, text):
-        parsed_link = urlparse(link)
+    def link(self, text, url, title=None):
+        parsed_link = urlparse(url)
         if (
             self.enable_relative_links
             and (not parsed_link.scheme and not parsed_link.netloc)
@@ -119,18 +119,18 @@ def link(self, link, title, text):
                     path=unquote(parsed_link.path),
                     replacement=replacement_link,
                     fragment=parsed_link.fragment,
-                    original=link,
-                    escaped_original=mistune.escape_link(link),
+                    original=url,
+                    escaped_original=mistune.escape_url(url),
                 )
             )
-            link = replacement_link
-        return super(ConfluenceRenderer, self).link(link, title, text)
+            url = replacement_link
+        return super(ConfluenceRenderer, self).link(text, url, title)
 
     def text(self, text):
         if self.remove_text_newlines:
             text = text.replace("\n", " ")
 
-        return super().text(text)
+        return text
 
     def block_code(self, code, lang=None):
         root_element = self.structured_macro("code")

md2cf/document.py

@@ -274,12 +274,15 @@ def parse_page(
     strip_header: bool = False,
     remove_text_newlines: bool = False,
     enable_relative_links: bool = False,
+    escape: bool = False,
+    allow_harmful_protocols: bool = False,
 ) -> Page:
     renderer = ConfluenceRenderer(
-        use_xhtml=True,
         strip_header=strip_header,
         remove_text_newlines=remove_text_newlines,
         enable_relative_links=enable_relative_links,
+        escape=escape,
+        allow_harmful_protocols=allow_harmful_protocols,
     )
     confluence_mistune = mistune.Markdown(renderer=renderer)
     confluence_content = confluence_mistune("".join(markdown_lines))

setup.py

@@ -23,7 +23,7 @@
     install_requires=[
         "rich-argparse==1.0.0",
         "rich==13.0.1",
-        "mistune==0.8.4",
+        "mistune==3.0.1",
         "chardet==5.1.0",
         "requests==2.31.0",
         "PyYAML==6.0.1",

test_package/functional/result.xml

@@ -1,27 +1,27 @@
 <h1>Markdown: Syntax</h1>
 <ul>
+<li><a href="#markdown-syntax">Markdown: Syntax</a><ul>
 <li><a href="#overview">Overview</a><ul>
 <li><a href="#philosophy">Philosophy</a></li>
-<li><a href="#html">Inline HTML</a></li>
-<li><a href="#autoescape">Automatic Escaping for Special Characters</a></li>
 </ul>
 </li>
-<li><a href="#block">Block Elements</a><ul>
-<li><a href="#p">Paragraphs and Line Breaks</a></li>
-<li><a href="#header">Headers</a></li>
-<li><a href="#blockquote">Blockquotes</a></li>
-<li><a href="#list">Lists</a></li>
-<li><a href="#precode">Code Blocks</a></li>
-<li><a href="#hr">Horizontal Rules</a></li>
+<li><a href="#block-elements">Block Elements</a><ul>
+<li><a href="#paragraphs-and-line-breaks">Paragraphs and Line Breaks</a></li>
+<li><a href="#headers">Headers</a></li>
+<li><a href="#blockquotes">Blockquotes</a></li>
+<li><a href="#lists">Lists</a></li>
+<li><a href="#code-blocks">Code Blocks</a></li>
 </ul>
 </li>
-<li><a href="#span">Span Elements</a><ul>
-<li><a href="#link">Links</a></li>
-<li><a href="#em">Emphasis</a></li>
+<li><a href="#span-elements">Span Elements</a><ul>
+<li><a href="#links">Links</a></li>
+<li><a href="#emphasis">Emphasis</a></li>
 <li><a href="#code">Code</a></li>
 </ul>
 </li>
 </ul>
+</li>
+</ul>
 <p><strong>Note:</strong> This document is itself written using Markdown; you
 can <a href="/projects/markdown/syntax.text">see the source for it by adding '.text' to the URL</a>.</p>
 <hr />
@@ -60,30 +60,37 @@ determines the header level.)</p>
 familiar with quoting passages of text in an email message, then you
 know how to create a blockquote in Markdown. It looks best if you hard
 wrap the text and put a <code>&gt;</code> before every line:</p>
-<blockquote><p>This is a blockquote with two paragraphs. Lorem ipsum dolor sit amet,
+<blockquote>
+<p>This is a blockquote with two paragraphs. Lorem ipsum dolor sit amet,
 consectetuer adipiscing elit. Aliquam hendrerit mi posuere lectus.
 Vestibulum enim wisi, viverra nec, fringilla in, laoreet vitae, risus.</p>
 <p>Donec sit amet nisl. Aliquam semper ipsum sit amet velit. Suspendisse
 id sem consectetuer libero luctus adipiscing.</p>
 </blockquote>
 <p>Markdown allows you to be lazy and only put the <code>&gt;</code> before the first
 line of a hard-wrapped paragraph:</p>
-<blockquote><p>This is a blockquote with two paragraphs. Lorem ipsum dolor sit amet,
+<blockquote>
+<p>This is a blockquote with two paragraphs. Lorem ipsum dolor sit amet,
 consectetuer adipiscing elit. Aliquam hendrerit mi posuere lectus.
 Vestibulum enim wisi, viverra nec, fringilla in, laoreet vitae, risus.</p>
+</blockquote>
+<blockquote>
 <p>Donec sit amet nisl. Aliquam semper ipsum sit amet velit. Suspendisse
 id sem consectetuer libero luctus adipiscing.</p>
 </blockquote>
 <p>Blockquotes can be nested (i.e. a blockquote-in-a-blockquote) by
 adding additional levels of <code>&gt;</code>:</p>
-<blockquote><p>This is the first level of quoting.</p>
-<blockquote><p>This is nested blockquote.</p>
+<blockquote>
+<p>This is the first level of quoting.</p>
+<blockquote>
+<p>This is nested blockquote.</p>
 </blockquote>
 <p>Back to the first level.</p>
 </blockquote>
 <p>Blockquotes can contain other Markdown elements, including headers, lists,
 and code blocks:</p>
-<blockquote><h2>This is a header.</h2>
+<blockquote>
+<h2>This is a header.</h2>
 <ol>
 <li>This is the first list item.</li>
 <li>This is the second list item.</li>
@@ -133,7 +140,7 @@ Markdown produces from the above list is:</p>
 <li>Parish</li>
 </ol>
 <p>or even:</p>
-<ol>
+<ol start="3">
 <li>Bird</li>
 <li>McHale</li>
 <li>Parish</li>
@@ -188,7 +195,8 @@ sit amet, consectetuer adipiscing elit.</p>
 delimiters need to be indented:</p>
 <ul>
 <li><p>A list item with a blockquote:</p>
-<blockquote><p>This is a blockquote
+<blockquote>
+<p>This is a blockquote
 inside a list item.</p>
 </blockquote>
 </li>
@@ -211,17 +219,13 @@ in both <code>&lt;pre&gt;</code> and <code>&lt;code&gt;</code> tags.</p>
 block by at least 4 spaces or 1 tab.</p>
 <p>This is a normal paragraph:</p>
 <ac:structured-macro ac:name="code"><ac:parameter ac:name="linenumbers">true</ac:parameter>
-<ac:plain-text-body><![CDATA[This is a code block.
-
-]]></ac:plain-text-body>
+<ac:plain-text-body><![CDATA[This is a code block.]]></ac:plain-text-body>
 </ac:structured-macro>
 <p>Here is an example of AppleScript:</p>
 <ac:structured-macro ac:name="code"><ac:parameter ac:name="linenumbers">true</ac:parameter>
 <ac:plain-text-body><![CDATA[tell application "Foo"
     beep
-end tell
-
-]]></ac:plain-text-body>
+end tell]]></ac:plain-text-body>
 </ac:structured-macro>
 <p>A code block continues until it reaches a line that is not indented
 (or the end of the article).</p>
@@ -233,17 +237,16 @@ ampersands and angle brackets. For example, this:</p>
 <ac:structured-macro ac:name="code"><ac:parameter ac:name="linenumbers">true</ac:parameter>
 <ac:plain-text-body><![CDATA[<div class="footer">
     &copy; 2004 Foo Corporation
-</div>
-
-]]></ac:plain-text-body>
+</div>]]></ac:plain-text-body>
 </ac:structured-macro>
 <p>Regular Markdown syntax is not processed within code blocks. E.g.,
 asterisks are just literal asterisks within a code block. This means
 it's also easy to use Markdown to write about Markdown's own syntax.</p>
 <ac:structured-macro ac:name="code"><ac:parameter ac:name="linenumbers">true</ac:parameter>
 <ac:plain-text-body><![CDATA[tell application "Foo"
     beep
-end tell]]></ac:plain-text-body>
+end tell
+]]></ac:plain-text-body>
 </ac:structured-macro>
 <h2>Span Elements</h2>
 <h3>Links</h3>

test_package/functional/test.md

@@ -1,20 +1,18 @@
 # Markdown: Syntax
 
-*   [Overview](#overview)
-    *   [Philosophy](#philosophy)
-    *   [Inline HTML](#html)
-    *   [Automatic Escaping for Special Characters](#autoescape)
-*   [Block Elements](#block)
-    *   [Paragraphs and Line Breaks](#p)
-    *   [Headers](#header)
-    *   [Blockquotes](#blockquote)
-    *   [Lists](#list)
-    *   [Code Blocks](#precode)
-    *   [Horizontal Rules](#hr)
-*   [Span Elements](#span)
-    *   [Links](#link)
-    *   [Emphasis](#em)
-    *   [Code](#code)
+- [Markdown: Syntax](#markdown-syntax)
+  - [Overview](#overview)
+    - [Philosophy](#philosophy)
+  - [Block Elements](#block-elements)
+    - [Paragraphs and Line Breaks](#paragraphs-and-line-breaks)
+    - [Headers](#headers)
+    - [Blockquotes](#blockquotes)
+    - [Lists](#lists)
+    - [Code Blocks](#code-blocks)
+  - [Span Elements](#span-elements)
+    - [Links](#links)
+    - [Emphasis](#emphasis)
+    - [Code](#code)
 
 
 **Note:** This document is itself written using Markdown; you

test_package/unit/test_renderer.py

@@ -116,7 +116,7 @@ def test_tag_render_with_child_and_text():
 
 def test_renderer_reinit():
     renderer = ConfluenceRenderer()
-    renderer.header("this is a title", 1)
+    renderer.heading("this is a title", 1)
     assert renderer.title is not None
 
     renderer.reinit()
@@ -157,7 +157,7 @@ def test_renderer_header_sets_title():
     test_header = "this is a header"
     renderer = ConfluenceRenderer()
 
-    renderer.header(test_header, 1)
+    renderer.heading(test_header, 1)
 
     assert renderer.title == test_header
 
@@ -166,7 +166,7 @@ def test_renderer_strips_header():
     test_header = "this is a header"
     renderer = ConfluenceRenderer(strip_header=True)
 
-    result = renderer.header(test_header, 1)
+    result = renderer.heading(test_header, 1)
 
     assert result == ""
 
@@ -175,7 +175,7 @@ def test_renderer_header_lower_level_does_not_set_title():
     test_header = "this is a header"
     renderer = ConfluenceRenderer()
 
-    renderer.header(test_header, 2)
+    renderer.heading(test_header, 2)
 
     assert renderer.title is None
 
@@ -185,8 +185,8 @@ def test_renderer_header_later_level_sets_title():
     test_header = "this is a header"
     renderer = ConfluenceRenderer()
 
-    renderer.header(test_lower_header, 2)
-    renderer.header(test_header, 1)
+    renderer.heading(test_lower_header, 2)
+    renderer.heading(test_header, 1)
 
     assert renderer.title is test_header
 
@@ -196,8 +196,8 @@ def test_renderer_header_only_sets_first_title():
     test_second_header = "this is another header"
     renderer = ConfluenceRenderer()
 
-    renderer.header(test_header, 1)
-    renderer.header(test_second_header, 1)
+    renderer.heading(test_header, 1)
+    renderer.heading(test_second_header, 1)
 
     assert renderer.title is test_header
 
@@ -274,7 +274,7 @@ def test_renderer_normal_link(relative_links):
     renderer = ConfluenceRenderer(enable_relative_links=relative_links)
 
     assert (
-        renderer.link(link="https://example.com", text="example link", title=None)
+        renderer.link(url="https://example.com", text="example link", title=None)
         == '<a href="https://example.com">example link</a>'
     )
 
@@ -284,7 +284,7 @@ def test_renderer_local_header_link(relative_links):
     renderer = ConfluenceRenderer(enable_relative_links=relative_links)
 
     assert (
-        renderer.link(link="#header-name", text="example link", title=None)
+        renderer.link(url="#header-name", text="example link", title=None)
         == '<a href="#header-name">example link</a>'
     )
 
@@ -296,7 +296,7 @@ def test_renderer_relative_link_enabled():
         r"<a href=\"md2cf-internal-link-([-a-z0-9]+)\">relative link</a>"
     )
     temporary_link = renderer.link(
-        link="document/../path/page.md", text="relative link", title=None
+        url="document/../path/page.md", text="relative link", title=None
     )
     assert relative_link_regex.match(temporary_link)
     assert len(renderer.relative_links) == 1
@@ -319,7 +319,7 @@ def test_renderer_relative_link_with_fragment_enabled():
         r"<a href=\"md2cf-internal-link-([-a-z0-9]+)\">relative link</a>"
     )
     temporary_link = renderer.link(
-        link="document/../path/page.md#header-name", text="relative link", title=None
+        url="document/../path/page.md#header-name", text="relative link", title=None
     )
     assert relative_link_regex.match(temporary_link)
     assert len(renderer.relative_links) == 1
@@ -339,7 +339,7 @@ def test_renderer_relative_link_disabled():
     renderer = ConfluenceRenderer(enable_relative_links=False)
 
     assert (
-        renderer.link(link="document/../path/page.md", text="relative link", title=None)
+        renderer.link(url="document/../path/page.md", text="relative link", title=None)
         == '<a href="document/../path/page.md">relative link</a>'
     )
     assert renderer.relative_links == []
@@ -350,7 +350,7 @@ def test_renderer_relative_link_with_fragment_disabled():
 
     assert (
         renderer.link(
-            link="document/../path/page.md#header-name",
+            url="document/../path/page.md#header-name",
             text="relative link",
             title=None,
         )