python-ai-crawler-scraper/obsidian_writer.py at main · Diatonic-AI/python-ai-crawler-scraper · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
#!/usr/bin/env python3
"""Obsidian-compatible Markdown file writer with YAML frontmatter."""

from datetime import datetime
from pathlib import Path
from typing import Dict, List
from urllib.parse import urlparse


class ObsidianWriter:
    """Writes pages as Obsidian-compatible Markdown files."""

    def __init__(self, vault_dir: Path):
        """Initialize writer with vault directory."""
        self.vault_dir = vault_dir
        self.vault_dir.mkdir(parents=True, exist_ok=True)

    def write_page(self, page_data: Dict, backlinks: List[Dict]) -> Path:
        """
        Write page to vault as Markdown file with frontmatter.
        """
        # Place under domain subfolder for organization
        url = page_data.get('url') or page_data.get('source_url', '')
        domain = urlparse(url).netloc if url else "unknown"
        domain_dir = self.vault_dir / domain
        domain_dir.mkdir(parents=True, exist_ok=True)

        filename = f"{page_data['slug']}.md"
        filepath = domain_dir / filename

        frontmatter = self._build_frontmatter(page_data, backlinks)
        markdown = page_data.get('markdown_content', '')

        # Append "See also" and "Backlinks" sections if available
        see_also = page_data.get('semantic_similar') or []
        if see_also:
            markdown += "\n\n## See also\n"
            for s in see_also:
                sid = s.get('id')
                if sid:
                    markdown += f"- [[{sid}]] ({s.get('score', 0):.2f})\n"
        if backlinks:
            markdown += "\n\n## Backlinks\n"
            for bl in backlinks[:20]:
                title = bl.get('title') or bl.get('source_url') or ''
                markdown += f"- [[{title}]]\n"

        with open(filepath, 'w', encoding='utf-8') as f:
            f.write("---\n")
            f.write(frontmatter)
            f.write("---\n\n")
            f.write(markdown)

        return filepath

    def _build_frontmatter(self, page_data: Dict, backlinks: List[Dict]) -> str:
        """Build YAML frontmatter."""
        lines = []

        lines.append(f"title: \"{page_data.get('title', 'Untitled')}\"")
        lines.append(f"source_url: \"{page_data.get('url', '')}\"")
        lines.append(f"slug: \"{page_data.get('slug', '')}\"")
        lines.append(f"crawled_at: \"{page_data.get('created_at', '')}\"")
        lines.append(f"updated_at: \"{page_data.get('updated_at', '')}\"")
        lines.append(f"crawl_depth: {page_data.get('crawl_depth', 0)}")
        lines.append(f"word_count: {page_data.get('word_count', 0)}")
        lines.append(f"checksum: \"{page_data.get('checksum', '')}\"")

        # Optional enrichments
        if page_data.get('summary'):
            lines.append(f"summary: \"{page_data.get('summary').replace('\\', '\\\\').replace('"', '\\"')}\"")
        if page_data.get('type'):
            lines.append(f"page_type: \"{page_data.get('type')}\"")
        if page_data.get('lang'):
            lines.append(f"lang: \"{page_data.get('lang')}\"")
        if page_data.get('embeddings_id'):
            lines.append(f"embeddings_id: \"{page_data.get('embeddings_id')}\"")

        # Tags
        metadata = page_data.get('metadata', {})
        tags = metadata.get('tags', ['web', 'crawled'])
        tags_yaml = ", ".join(tags)
        lines.append(f"tags: [{tags_yaml}]")

        # Validation results (non-blocking)
        ext_errors = page_data.get('external_link_errors') or []
        if ext_errors:
            lines.append("external_link_errors:")
            for e in ext_errors:
                lines.append(f"  - \"{e}\"")
        missing = page_data.get('missing_internal_links') or []
        if missing:
            lines.append("missing_internal_links:")
            for m in missing:
                lines.append(f"  - \"{m}\"")
        similar = page_data.get('semantic_similar') or []
        if similar:
            lines.append("semantic_similar:")
            for s in similar:
                lines.append(f"  - id: \"{s.get('id','')}\"\n    score: {s.get('score', 0)}")
        if page_data.get('last_validated'):
            lines.append(f"last_validated: \"{page_data.get('last_validated')}\"")

        # Backlinks (titles only for now)
        if backlinks:
            lines.append("backlinks:")
            for bl in backlinks[:20]:
                lines.append(f"  - \"{bl.get('title', bl.get('source_url', ''))}\"")

        return "\n".join(lines) + "\n"


if __name__ == "__main__":
    # Test writer
    test_dir = Path("test_vault")
    writer = ObsidianWriter(test_dir)

    test_data = {
        'title': 'Test Page',
        'slug': 'test-page',
        'url': 'https://example.com/test',
        'markdown_content': '# Test\n\nThis is test content.',
        'created_at': datetime.utcnow().isoformat(),
        'word_count': 5,
        'checksum': 'abc123',
        'crawl_depth': 1
    }

    filepath = writer.write_page(test_data, [])
    print(f"✅ Written to: {filepath}")

    # Cleanup
    filepath.unlink()
    test_dir.rmdir()