Use regex for parsing

pyproject.toml

@@ -1,20 +1,9 @@
-[project]
-name = "ya-changelog"
-description = "Yet another changelog CLI tool."
-authors = ["Andrew Cassidy"]
-license = "AGPL"
-readme = "README.md"
-python = "^3.8"
-homepage = "https://github.com/drewcassidy/yet-another-changelog"
-repository = "https://github.com/drewcassidy/yet-another-changelog"
-
-keywords = ["changelog", "commandline"]
-
-dependencies = ["Click", "GitPython"]
-
 [build-system]
 requires = [
     "setuptools >= 35.0.2",
-    "setuptools_scm >= 2.0.0"
+    "setuptools_scm[toml] >= 3.4",
+    "wheel"
 ]
-build-backend = "setuptools.build_meta"
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools_scm]

setup.cfg

@@ -0,0 +1,29 @@
+[metadata]
+# until setuptools supports PEP621, this will have to do
+name = yaclog
+description = Yet another changelog CLI tool.
+author = Andrew Cassidy
+license = AGPLv3
+license_file = LICENSE.md
+long_description = file: README.md
+long_description_content_type = text/markdown
+url = https://github.com/drewcassidy/yet-another-changelog
+
+keywords = changelog, commandline, markdown
+classifiers =
+    Development Status :: 3 - Alpha
+    Intended Audience :: Developers
+    License :: OSI Approved :: GNU Affero General Public License v3
+    Operating System :: OS Independent
+    Programming Language :: Python :: 3 :: Only
+    Programming Language :: Python :: 3
+    Programming Language :: Python :: 3.8
+    Programming Language :: Python :: 3.9
+    Topic :: Text Processing :: Markup :: Markdown
+    Topic :: Software Development :: Version Control :: Git
+    Topic :: Utilities
+
+[options]
+install_requires = Click; GitPython
+python_requires >= 3.8
+packages = find:

yaclog/__init__.py

@@ -0,0 +1,11 @@
+import os
+from yaclog.changelog import Changelog
+
+
+def read(path: os.PathLike):
+    """
+    Create a new Changelog object from the given path
+    :param path: a path to a markdown changelog file
+    :return: a parsed Changelog object
+    """
+    return Changelog(path)

yaclog/changelog.py

@@ -62,44 +62,43 @@ class Changelog:
                 if line.isspace():
                     # skip empty lines
                     pass
-                elif line.startswith('## '):
+                elif match := re.fullmatch(
-                    # line is a version header
+                        r'^##\s+(?P<name>\S*)(?:\s+-\s+(?P<date>\S+))?\s*?(?P<extra>.*?)\s*#*$', line):
+                    # this is a version header in the form '## Name (- date) (tags*) (#*)'
                     version = VersionEntry()
                     section = ''
-                    split = line.removeprefix('##').strip().split()
 
-                    version.name, version.link, version.link_id = _strip_link(split[0])
+                    version.name, version.link, version.link_id = _strip_link(match['name'])
-
-                    if len(split) > 1:
-                        if split[1] == '-':
-                            split.remove('-')
 
+                    if match['date']:
                         try:
-                            version.date = datetime.date.fromisoformat(split[1].strip(string.punctuation))
+                            version.date = datetime.date.fromisoformat(match['date'].strip(string.punctuation))
                         except ValueError:
                             version.date = None
 
-                        version.tags = [s.strip(brackets) for s in split[2:]]
+                    if match['extra']:
+                        version.tags = [s.strip('[]') for s in re.findall(r'\[.*?]', match['extra'])]
+
                     self.versions.append(version)
 
-                elif line.startswith('### '):
+                elif match := re.fullmatch(r'###\s+(\S*?)(\s+#*)?', line):
-                    # line is a version section header
+                    # this is a version section header in the form '### Name' or '### Name ###'
-                    section = line.removeprefix('###').strip().title()
+                    section = match[1].title()
                     if section not in version.sections.keys():
                         version.sections[section] = []
 
+                elif match := re.fullmatch(r'\[(\S*)]:\s*(\S*)\n', line):
+                    # this is a link definition in the form '[id]: link', so add it to the link table
+                    self.links[match[1].lower()] = match[2]
+
+                elif line[0] in bullets or last_line.isspace():
+                    # bullet point or new paragraph
+                    # bullet points are preserved since some people like to use '+', '-' or '*' for different things
+                    version.sections[section].append(line.strip())
+
                 else:
-                    # line is an entry
+                    # not a bullet point, and no whitespace on last line, so append to the last entry
-                    if match := re.fullmatch(r'\[(.*)]:\s*(.*)\n', line):
+                    version.sections[section][-1] += '\n' + line.strip()
-                        # this is a link, so a it to the link table
-                        self.links[match[1].lower()] = match[2]
-                    elif line[0] in bullets or last_line.isspace():
-                        # bullet point or new paragraph
-                        # bullet points are preserved since some people like to use '+', '-' or '*' for different things
-                        version.sections[section].append(line.strip())
-                    else:
-                        # not a bullet point, and no whitespace on last line, so append to the last entry
-                        version.sections[section][-1] += '\n' + line.strip()
 
                 last_line = line
                 line = fp.readline()