Create a changelog analyzer

Here is a functional Changelog Analyzer in Python that parses changelog files (in Markdown format commonly used in CHANGELOG.md) and provides a structured summary of versions, dates, and categorized changes (e.g., Added, Fixed, Changed).

python
CopyEdit
import re
from collections import defaultdict
from pprint import pprint

class ChangelogAnalyzer:
    def __init__(self, changelog_text):
        self.changelog_text = changelog_text
        self.entries = defaultdict(dict)

    def parse(self):
        # Pattern for version headers like ## [1.0.0] - 2023-12-31
        version_header_pattern = re.compile(r'^##s*[?([d.]+)]?s*-s*(d{4}-d{2}-d{2})?', re.MULTILINE)
        # Pattern for change type headers like ### Added
        change_type_pattern = re.compile(r'^###s*(Added|Changed|Deprecated|Removed|Fixed|Security)', re.MULTILINE)

        pos = 0
        while True:
            version_match = version_header_pattern.search(self.changelog_text, pos)
            if not version_match:
                break

            version = version_match.group(1)
            date = version_match.group(2) or "Unreleased"
            start = version_match.end()

            next_version_match = version_header_pattern.search(self.changelog_text, start)
            end = next_version_match.start() if next_version_match else len(self.changelog_text)

            version_block = self.changelog_text[start:end]
            changes = defaultdict(list)

            for change_match in change_type_pattern.finditer(version_block):
                change_type = change_match.group(1)
                change_start = change_match.end()
                next_change_match = change_type_pattern.search(version_block, change_start)
                change_end = next_change_match.start() if next_change_match else len(version_block)
                items_block = version_block[change_start:change_end]
                items = re.findall(r'- (.+)', items_block)
                changes[change_type].extend(items)

            self.entries[version] = {
                "date": date,
                "changes": dict(changes)
            }

            pos = end

    def get_summary(self):
        return dict(self.entries)


# Example usage:
if __name__ == "__main__":
    with open("CHANGELOG.md", "r") as file:
        changelog_content = file.read()

    analyzer = ChangelogAnalyzer(changelog_content)
    analyzer.parse()
    summary = analyzer.get_summary()
    pprint(summary)

Features:

• Extracts version numbers and dates.

• Categorizes changes under headings like Added, Fixed, Changed, etc.

• Handles Markdown-style changelog formatting (e.g., Keep a Changelog format).

• Can be extended to export JSON or integrate into CI pipelines.

Let me know if you want a web interface, JSON export, Git integration, or support for alternative changelog formats.