Dont use YAML for the changelog

No point having an extra dependency just to parse the changelog, which
doesnt have a ot of structured information, anyway.

setup/changelog.py

@@ -0,0 +1,169 @@
+#!/usr/bin/env python
+# vim:fileencoding=utf-8
+# License: GPL v3 Copyright: 2020, Kovid Goyal <kovid at kovidgoyal.net>
+
+from datetime import date
+
+
+def parse(raw):
+    entries = []
+    current_entry = None
+    current_section = 'new features'
+
+    def normal(linenum, line, stripped_line):
+        nonlocal current_entry, current_section
+        if not stripped_line:
+            return normal
+        if stripped_line.startswith('{' '{' '{'):
+            parts = line.split()[1:]
+            if len(parts) != 2:
+                raise ValueError(f'The entry start line is malformed: {line}')
+            if current_entry is not None:
+                raise ValueError(f'Start of entry while previous entry is still active at line: {linenum}')
+            version, draw = parts
+            d = date(*map(int, draw.split('-')))
+            current_entry = {'version': version, 'date': d, 'new features': [], 'bug fixes': [], 'improved recipes': [], 'new recipes': []}
+            current_section = 'new features'
+            return in_entry
+        raise ValueError(f'Invalid content at line {linenum}: {line}')
+
+    def in_entry(linenum, line, stripped_line):
+        nonlocal current_section, current_entry
+        if stripped_line == '}' '}' '}':
+            if current_entry is None:
+                raise ValueError(f'Entry terminator without active entry at line: {linenum}')
+            entries.append(current_entry)
+            current_entry = None
+            return normal
+        if line.startswith(':: '):
+            current_section = line[3:].strip()
+            if current_section not in ('new features', 'bug fixes', 'new recipes', 'improved recipes'):
+                raise ValueError(f'Unknown section: {current_section}')
+            return in_entry
+        if line.startswith('-'):
+            return start_item(linenum, line, stripped_line)
+        if not stripped_line:
+            return in_entry
+        raise ValueError(f'Invalid content at line {linenum}: {line}')
+
+    def start_item(linenum, line, stripped_line):
+        line = line[1:].lstrip()
+        items = current_entry[current_section]
+        if current_section == 'improved recipes':
+            items.append(line.rstrip())
+            return in_entry
+        if current_section == 'new recipes':
+            idx = line.rfind('by ')
+            if idx == -1:
+                items.append({'title': line.strip()})
+            else:
+                items.append({'title': line[:idx].strip(), 'author': line[idx + 3:].strip()})
+            return in_entry
+        item = {}
+        if line.startswith('['):
+            idx = line.find(']')
+            if idx == -1:
+                raise ValueError(f'No closing ] found in line: {linenum}')
+            for x in line[1:idx].split():
+                if x == 'major':
+                    item['type'] = x
+                    continue
+                num = int(x)
+                item.setdefault('tickets', []).append(num)
+            item['title'] = line[idx+1:].strip()
+        else:
+            item['title'] = line.strip()
+        items.append(item)
+        return in_item
+
+    def finalize_item(item):
+        if 'description' in item and not item['description']:
+            del item['description']
+        return item
+
+    def in_item(linenum, line, stripped_line):
+        item = current_entry[current_section][-1]
+        if line.startswith('::'):
+            finalize_item(item)
+            return in_entry(linenum, line, stripped_line)
+        if line.startswith('-'):
+            finalize_item(item)
+            return start_item(linenum, line, stripped_line)
+        if line.startswith('}' '}' '}'):
+            return in_entry(linenum, line, stripped_line)
+        if not stripped_line:
+            if 'description' not in item:
+                item['description'] = ''
+            return in_item
+        if 'description' in item:
+            item['description'] += stripped_line
+        else:
+            item['title'] += stripped_line
+        return in_item
+
+    state = normal
+    for i, line in enumerate(raw.splitlines()):
+        if line.startswith('#'):
+            continue
+        stripped_line = line.strip()
+        state = state(i + 1, line, stripped_line)
+    return entries
+
+
+def migrate():
+    from yaml import safe_load
+
+    def output_item(item, lines):
+        meta = []
+        if item.get('type') == 'major':
+            meta.append(item['type'])
+        for x in item.get('tickets', ()):
+            meta.append(str(x))
+        title = item['title']
+        if meta:
+            meta = ' '.join(meta)
+            title = f'[{meta}] {title}'
+        lines.append(f'- {title}')
+        d = item.get('description')
+        if d:
+            lines.append(''), lines.append(d)
+        lines.append('')
+
+    for name in ('Changelog.yaml', 'Changelog.old.yaml'):
+        entries = safe_load(open(name).read())
+        lines = []
+        for entry in entries:
+            lines.append('')
+            lines.append('{' '{' '{' f' {entry["version"]} {entry["date"]}')
+            for w in ('new features', 'bug fixes'):
+                nf = entry.get(w)
+                if nf:
+                    lines.append(f':: {w}'), lines.append('')
+                    for x in nf:
+                        output_item(x, lines)
+                    lines.append('')
+            nr = entry.get('new recipes')
+            if nr:
+                lines.append(':: new recipes'), lines.append('')
+                for r in nr:
+                    aut = r.get('author') or r.get('authors')
+                    title = r['title']
+                    if title:
+                        if aut:
+                            lines.append(f'- {title} by {aut}')
+                        else:
+                            lines.append(f'- {title}')
+                lines.append('')
+            ir = entry.get('improved recipes')
+            if ir:
+                lines.append(':: improved recipes'), lines.append('')
+                for r in ir:
+                    lines.append(f'- {r}')
+                lines.append('')
+            with open(name.replace('yaml', 'txt'), 'w') as f:
+                f.write('\n'.join(lines))
+            lines.append(''), lines.append('}' '}' '}'), lines.append('')
+
+
+if __name__ == '__main__':
+    migrate()