Add base_url option to get absolute URLs

README.rst

@@ -64,6 +64,11 @@ autolinks
   A boolean indicating whether the "automatic link" style should be used when
   a ``a`` tag's contents match its href. Defaults to ``True``.
 
+base_url
+  A base URL to use for resolving relative URLs. When specified, relative URLs
+  in the HTML will be converted to absolute URLs using this base. Defaults to
+  no base URL.
+
 default_title
   A boolean to enable setting the title of a link to its href, if no title is
   given. Defaults to ``False``.

markdownify/__init__.py

@@ -2,6 +2,7 @@
 from textwrap import fill
 import re
 import six
+import urllib.parse
 
 
 # General-purpose regex patterns
@@ -176,6 +177,7 @@ def _next_block_content_sibling(el):
 class MarkdownConverter(object):
     class DefaultOptions:
         autolinks = True
+        base_url = ''
         bs4_options = 'html.parser'
         bullets = '*+-'  # An iterable of bullet types.
         code_language = ''
@@ -435,6 +437,20 @@ def underline(self, text, pad_char):
         text = (text or '').rstrip()
         return '\n\n%s\n%s\n\n' % (text, pad_char * len(text)) if text else ''
 
+    def _make_absolute(self, url):
+        """Convert a URL to absolute using base_url if it's not already absolute."""
+        base_url = self.options['base_url']
+
+        # Check if URLs to join actually exist
+        if not url or not base_url:
+            return url
+
+        # Check if URL is already absolute
+        if urllib.parse.urlparse(url).netloc:
+            return url
+
+        return urllib.parse.urljoin(base_url, url)
+
     def convert_a(self, el, text, parent_tags):
         if '_noformat' in parent_tags:
             return text
@@ -453,6 +469,7 @@ def convert_a(self, el, text, parent_tags):
         if self.options['default_title'] and not title:
             title = href
         title_part = ' "%s"' % title.replace('"', r'\"') if title else ''
+        href = self._make_absolute(href)
         return '%s[%s](%s%s)%s' % (prefix, text, href, title_part, suffix) if href else text
 
     convert_b = abstract_inline_conversion(lambda self: 2 * self.options['strong_em_symbol'])
@@ -588,6 +605,7 @@ def convert_img(self, el, text, parent_tags):
                 and el.parent.name not in self.options['keep_inline_images_in']):
             return alt
 
+        src = self._make_absolute(src)
         return '![%s](%s%s)' % (alt, src, title_part)
 
     def convert_video(self, el, text, parent_tags):
@@ -600,6 +618,8 @@ def convert_video(self, el, text, parent_tags):
             if sources:
                 src = sources[0].attrs.get('src', None) or ''
         poster = el.attrs.get('poster', None) or ''
+        src = self._make_absolute(src)
+        poster = self._make_absolute(poster)
         if src and poster:
             return '[![%s](%s)](%s)' % (text, poster, src)
         if src:

markdownify/__init__.pyi

@@ -18,6 +18,7 @@ STRIP_ONE: str
 def markdownify(
     html: str,
     autolinks: bool = ...,
+    base_url: str = ...,
     bs4_options: str = ...,
     bullets: str = ...,
     code_language: str = ...,
@@ -46,6 +47,7 @@ class MarkdownConverter:
     def __init__(
         self,
         autolinks: bool = ...,
+        base_url: str = ...,
         bs4_options: str = ...,
         bullets: str = ...,
         code_language: str = ...,

tests/test_conversions.py

@@ -374,3 +374,10 @@ def test_spaces():
     assert md(' <ol> <li> x </li> <li> y </li> </ol> ') == '\n\n1. x\n2. y\n'
     assert md(' <ul> <li> x </li> <li> y </li> </ol> ') == '\n\n* x\n* y\n'
     assert md('test <pre> foo </pre> bar') == 'test\n\n```\n foo\n```\n\nbar'
+
+
+def test_a_with_base_url():
+    assert md('<a href="/path">Link</a>', base_url='https://example.com') == '[Link](https://example.com/path)'
+    assert md('<a href="../other">Link</a>', base_url='https://example.com/page/') == '[Link](https://example.com/other)'
+    assert md('<a href="https://other.com">Link</a>', base_url='https://example.com') == '[Link](https://other.com)'
+    assert md('<a href="/path">Link</a>') == '[Link](/path)'