moved mardkown stuff into separate module

sg/markdown.py

@@ -0,0 +1,100 @@
+from datetime import datetime
+import logging
+from urllib.parse import urlsplit, urlunsplit
+
+from markdown import Markdown
+from markdown.extensions import Extension
+from markdown.treeprocessors import Treeprocessor
+
+
+logger = logging.getLogger(__name__)
+
+
+def markdown_factory():
+    """Create a Markdown instance.
+
+    This method exists only to ensure we use the same Markdown instance
+    for tests as for the actual thing.
+
+    Returns
+    -------
+    markdown.Markdown
+
+    """
+    md = Markdown(
+        extensions=[
+            'meta', 'fenced_code', 'codehilite',
+            MarkdownLinkExtension()
+        ],
+        output_format='html5',
+    )
+    return md
+
+
+def convert_markdown(md, markdown):
+    """Convert markdown into html and extract meta data.
+
+    Parameters
+    ----------
+    md : markdown.Markdown instance
+    markdown : str
+
+    Returns
+    -------
+    str, dict :
+        html and metadata
+
+    """
+    md.reset()
+    content = md.convert(markdown)
+    meta = md.Meta
+
+    # markdowns metadata consists as list of strings -- one item per
+    # line. let's convert into single strings.
+    for key, value in meta.items():
+        value = '\n'.join(value)
+        meta[key] = value
+
+    # convert known metadata
+    # date: datetime
+    if 'date' in meta:
+        meta['date'] = datetime.fromisoformat(meta['date'])
+    # tags: list[str]
+    if 'tags' in meta:
+        tags = meta['tags'].split(',')
+        tags = [t.strip() for t in tags]
+        meta['tags'] = tags
+
+    return content, meta
+
+
+class MarkdownLinkTreeprocessor(Treeprocessor):
+    """Converts relative links to .md files to .html
+
+    """
+
+    def run(self, root):
+        for element in root.iter():
+            if element.tag == 'a':
+                url = element.get('href')
+                converted = self.convert(url)
+                element.set('href', converted)
+        return root
+
+    def convert(self, url):
+        scheme, netloc, path, query, fragment = urlsplit(url)
+        #logger.debug(f'{url} -> scheme: {scheme} netloc: {netloc} path: {path} query: {query} fragment: {fragment}')
+        if (scheme or netloc or not path):
+            return url
+        if path.endswith('.md'):
+            path = path[:-3] + '.html'
+
+        url = urlunsplit((scheme, netloc, path, query, fragment))
+        return url
+
+
+class MarkdownLinkExtension(Extension):
+    def extendMarkdown(self, md):
+        md.treeprocessors.register(
+                MarkdownLinkTreeprocessor(md), 'mdlink', 0,
+        )

sg/sg.py

@@ -12,15 +12,12 @@ import argparse
 import os
 import shutil
 import logging
-from datetime import datetime
-from urllib.parse import urlsplit, urlunsplit
 
-from markdown import Markdown
-from markdown.extensions import Extension
-from markdown.treeprocessors import Treeprocessor
 from jinja2 import Environment, ChoiceLoader, FileSystemLoader, PackageLoader
 import feedgenerator
 
+from sg.markdown import markdown_factory, convert_markdown
+
 logger = logging.getLogger(__name__)
 logging.basicConfig(
         level=logging.DEBUG,
@@ -89,27 +86,6 @@ def build(args):
     convert_to_html(convertibles, args.input_dir, args.output_dir)
 
 
-def markdown_factory():
-    """Create a Markdown instance.
-
-    This method exists only to ensure we use the same Markdown instance
-    for tests as for the actual thing.
-
-    Returns
-    -------
-    markdown.Markdown
-
-    """
-    md = Markdown(
-        extensions=[
-            'meta', 'fenced_code', 'codehilite',
-            MarkdownLinkExtension()
-        ],
-        output_format='html5',
-    )
-    return md
-
-
 def convert_to_html(convertibles, input_dir, output_dir):
 
     env = Environment(
@@ -192,74 +168,6 @@ def convert_to_html(convertibles, input_dir, output_dir):
 
 
 
-def convert_markdown(md, markdown):
-    """Convert markdown into html and extract meta data.
-
-    Parameters
-    ----------
-    md : markdown.Markdown instance
-    markdown : str
-
-    Returns
-    -------
-    str, dict :
-        html and metadata
-
-    """
-    md.reset()
-    content = md.convert(markdown)
-    meta = md.Meta
-
-    # markdowns metadata consists as list of strings -- one item per
-    # line. let's convert into single strings.
-    for key, value in meta.items():
-        value = '\n'.join(value)
-        meta[key] = value
-
-    # convert known metadata
-    # date: datetime
-    if 'date' in meta:
-        meta['date'] = datetime.fromisoformat(meta['date'])
-    # tags: list[str]
-    if 'tags' in meta:
-        tags = meta['tags'].split(',')
-        tags = [t.strip() for t in tags]
-        meta['tags'] = tags
-
-    return content, meta
-
-
-class MarkdownLinkTreeprocessor(Treeprocessor):
-    """Converts relative links to .md files to .html
-
-    """
-
-    def run(self, root):
-        for element in root.iter():
-            if element.tag == 'a':
-                url = element.get('href')
-                converted = self.convert(url)
-                element.set('href', converted)
-        return root
-
-    def convert(self, url):
-        scheme, netloc, path, query, fragment = urlsplit(url)
-        logger.debug(f'{url} -> scheme: {scheme} netloc: {netloc} path: {path} query: {query} fragment: {fragment}')
-        if (scheme or netloc or not path):
-            return url
-        if path.endswith('.md'):
-            path = path[:-3] + '.html'
-
-        url = urlunsplit((scheme, netloc, path, query, fragment))
-        return url
-
-
-class MarkdownLinkExtension(Extension):
-    def extendMarkdown(self, md):
-        md.treeprocessors.register(
-                MarkdownLinkTreeprocessor(md), 'mdlink', 0,
-        )
-
 
 if __name__ == '__main__':
     main()

tests/test_markdown.py

@@ -0,0 +1,48 @@
+from datetime import datetime
+
+import pytest
+import markdown
+
+from sg.markdown import convert_markdown, markdown_factory
+
+
+@pytest.mark.parametrize("input_, expected", [
+    # inline
+    ('[test](test.md)', 'test.html'),
+    ('[test](test.md "test")', 'test.html'),
+    ('[test](a/test.md)', 'a/test.html'),
+    ('[test](a/test.md "test")', 'a/test.html'),
+    ('[test](/test.md)', '/test.html'),
+    ('[test](/test.md "test")', '/test.html'),
+    ('[test](/a/test.md)', '/a/test.html'),
+    ('[test](/a/test.md "test")', '/a/test.html'),
+    # reference
+    ('[test][]\n[test]: test.md ''', 'test.html'),
+    ('[test][]\n[test]: test.md "test"', 'test.html'),
+    ('[test][]\n[test]: a/test.md', 'a/test.html'),
+    ('[test][]\n[test]: a/test.md "test"', 'a/test.html'),
+    ('[test][]\n[test]: /test.md', '/test.html'),
+    ('[test][]\n[test]: /test.md "test"', '/test.html'),
+    ('[test][]\n[test]: /a/test.md', '/a/test.html'),
+    ('[test][]\n[test]: /a/test.md "test"', '/a/test.html'),
+])
+def test_convert_markdown_links(input_, expected):
+    md = markdown_factory()
+    html, _ = convert_markdown(md, input_)
+    assert expected in html
+
+
+@pytest.mark.parametrize("input_, expected", [
+    ('foo: bar', {'foo': 'bar'}),
+    ('tags: this, is, a, test\n', {'tags': ['this', 'is', 'a', 'test']}),
+    ('date: 2020-01-01 12:10', {'date': datetime(2020, 1, 1, 12, 10)}),
+])
+def test_convert_metadata(input_, expected):
+    md = markdown_factory()
+    _, meta = convert_markdown(md, input_)
+    assert expected == meta
+
+
+def test_markdown_factory():
+    md = markdown_factory()
+    assert isinstance(md, markdown.Markdown)

tests/test_sg.py

@@ -1,48 +0,0 @@
-from datetime import datetime
-
-import markdown
-import pytest
-
-from sg import sg
-
-
-@pytest.mark.parametrize("input_, expected", [
-    # inline
-    ('[test](test.md)', 'test.html'),
-    ('[test](test.md "test")', 'test.html'),
-    ('[test](a/test.md)', 'a/test.html'),
-    ('[test](a/test.md "test")', 'a/test.html'),
-    ('[test](/test.md)', '/test.html'),
-    ('[test](/test.md "test")', '/test.html'),
-    ('[test](/a/test.md)', '/a/test.html'),
-    ('[test](/a/test.md "test")', '/a/test.html'),
-    # reference
-    ('[test][]\n[test]: test.md ''', 'test.html'),
-    ('[test][]\n[test]: test.md "test"', 'test.html'),
-    ('[test][]\n[test]: a/test.md', 'a/test.html'),
-    ('[test][]\n[test]: a/test.md "test"', 'a/test.html'),
-    ('[test][]\n[test]: /test.md', '/test.html'),
-    ('[test][]\n[test]: /test.md "test"', '/test.html'),
-    ('[test][]\n[test]: /a/test.md', '/a/test.html'),
-    ('[test][]\n[test]: /a/test.md "test"', '/a/test.html'),
-])
-def test_convert_markdown_links(input_, expected):
-    md = sg.markdown_factory()
-    html, _ = sg.convert_markdown(md, input_)
-    assert expected in html
-
-
-@pytest.mark.parametrize("input_, expected", [
-    ('foo: bar', {'foo': 'bar'}),
-    ('tags: this, is, a, test\n', {'tags': ['this', 'is', 'a', 'test']}),
-    ('date: 2020-01-01 12:10', {'date': datetime(2020, 1, 1, 12, 10)}),
-])
-def test_convert_metadata(input_, expected):
-    md = sg.markdown_factory()
-    _, meta = sg.convert_markdown(md, input_)
-    assert expected == meta
-
-
-def test_markdown_factory():
-    md = sg.markdown_factory()
-    assert isinstance(md, markdown.Markdown)