test: add tests for session pair harvester (#91)

scripts/diff_analyzer.py

@@ -1,216 +0,0 @@
-#!/usr/bin/env python3
-"""
-Diff Analyzer — Parse unified diffs and categorize every change.
-
-Pipeline 6.1 for Compounding Intelligence.
-"""
-
-import re
-from dataclasses import dataclass, field, asdict
-from enum import Enum
-from typing import List, Dict, Any, Optional
-
-
-class ChangeCategory(Enum):
-    ADDED = "added"
-    DELETED = "deleted"
-    MODIFIED = "modified"
-    MOVED = "moved"
-    CONTEXT = "context"
-
-
-@dataclass
-class Hunk:
-    """A single diff hunk with header, line ranges, and category."""
-    header: str
-    old_start: int
-    old_count: int
-    new_start: int
-    new_count: int
-    lines: List[str] = field(default_factory=list)
-    category: ChangeCategory = ChangeCategory.CONTEXT
-
-    def to_dict(self) -> Dict[str, Any]:
-        d = asdict(self)
-        d["category"] = self.category.value
-        return d
-
-
-@dataclass
-class FileChange:
-    """A single file's changes."""
-    path: str
-    old_path: Optional[str] = None  # For renames
-    hunks: List[Hunk] = field(default_factory=list)
-    added_lines: int = 0
-    deleted_lines: int = 0
-    is_new: bool = False
-    is_deleted: bool = False
-    is_renamed: bool = False
-    is_binary: bool = False
-
-    def to_dict(self) -> Dict[str, Any]:
-        return {
-            "path": self.path,
-            "old_path": self.old_path,
-            "hunks": [h.to_dict() for h in self.hunks],
-            "added_lines": self.added_lines,
-            "deleted_lines": self.deleted_lines,
-            "is_new": self.is_new,
-            "is_deleted": self.is_deleted,
-            "is_renamed": self.is_renamed,
-            "is_binary": self.is_binary,
-        }
-
-
-@dataclass
-class ChangeSummary:
-    """Aggregate stats + per-file breakdown."""
-    files: List[FileChange] = field(default_factory=list)
-    total_added: int = 0
-    total_deleted: int = 0
-    total_files_changed: int = 0
-    total_hunks: int = 0
-    new_files: int = 0
-    deleted_files: int = 0
-    renamed_files: int = 0
-    binary_files: int = 0
-
-    def to_dict(self) -> Dict[str, Any]:
-        return {
-            "total_files_changed": self.total_files_changed,
-            "total_added": self.total_added,
-            "total_deleted": self.total_deleted,
-            "total_hunks": self.total_hunks,
-            "new_files": self.new_files,
-            "deleted_files": self.deleted_files,
-            "renamed_files": self.renamed_files,
-            "binary_files": self.binary_files,
-            "files": [f.to_dict() for f in self.files],
-        }
-
-
-class DiffAnalyzer:
-    """Parses unified diff format and produces structured ChangeSummary."""
-
-    HUNK_HEADER_RE = re.compile(r"^@@\s+-(\d+)(?:,(\d+))?\s+\+(\d+)(?:,(\d+))?\s+@@(.*)$")
-    DIFF_FILE_RE = re.compile(r"^diff --git a/(.*) b/(.*)")
-    RENAME_RE = re.compile(r"^rename from (.+)$")
-    RENAME_TO_RE = re.compile(r"^rename to (.+)$")
-    NEW_FILE_RE = re.compile(r"^new file mode")
-    DELETED_FILE_RE = re.compile(r"^deleted file mode")
-    BINARY_RE = re.compile(r"^Binary files .* differ")
-
-    def analyze(self, diff_text: str) -> ChangeSummary:
-        """Parse a unified diff and return a ChangeSummary."""
-        summary = ChangeSummary()
-        if not diff_text or not diff_text.strip():
-            return summary
-
-        # Split diff into per-file sections
-        file_diffs = self._split_files(diff_text)
-
-        for file_diff in file_diffs:
-            fc = self._parse_file_diff(file_diff)
-            summary.files.append(fc)
-            summary.total_added += fc.added_lines
-            summary.total_deleted += fc.deleted_lines
-            summary.total_hunks += len(fc.hunks)
-            if fc.is_new:
-                summary.new_files += 1
-            if fc.is_deleted:
-                summary.deleted_files += 1
-            if fc.is_renamed:
-                summary.renamed_files += 1
-            if fc.is_binary:
-                summary.binary_files += 1
-
-        summary.total_files_changed = len(summary.files)
-        return summary
-
-    def _split_files(self, diff_text: str) -> List[str]:
-        """Split a multi-file diff into individual file diffs."""
-        lines = diff_text.split("\n")
-        chunks = []
-        current = []
-        for line in lines:
-            if line.startswith("diff --git ") and current:
-                chunks.append("\n".join(current))
-                current = [line]
-            else:
-                current.append(line)
-        if current:
-            chunks.append("\n".join(current))
-        return chunks
-
-    def _parse_file_diff(self, diff_text: str) -> FileChange:
-        """Parse a single file's diff section."""
-        lines = diff_text.split("\n")
-        fc = FileChange(path="")
-
-        # Extract file paths
-        for line in lines:
-            m = self.DIFF_FILE_RE.match(line)
-            if m:
-                fc.path = m.group(2)
-                break
-
-        # Check for special states
-        for line in lines:
-            if self.NEW_FILE_RE.match(line):
-                fc.is_new = True
-            elif self.DELETED_FILE_RE.match(line):
-                fc.is_deleted = True
-            elif self.RENAME_RE.match(line):
-                fc.old_path = m.group(1) if (m := self.RENAME_RE.match(line)) else None
-                fc.is_renamed = True
-            elif self.BINARY_RE.match(line):
-                fc.is_binary = True
-                return fc  # No hunks for binary
-
-        # Rename TO
-        for line in lines:
-            m = self.RENAME_TO_RE.match(line)
-            if m and fc.is_renamed:
-                fc.path = m.group(1)
-
-        # Parse hunks
-        current_hunk = None
-        for line in lines:
-            m = self.HUNK_HEADER_RE.match(line)
-            if m:
-                if current_hunk:
-                    self._classify_hunk(current_hunk, fc)
-                    fc.hunks.append(current_hunk)
-                current_hunk = Hunk(
-                    header=m.group(5).strip(),
-                    old_start=int(m.group(1)),
-                    old_count=int(m.group(2) or 1),
-                    new_start=int(m.group(3)),
-                    new_count=int(m.group(4) or 1),
-                )
-            elif current_hunk and (line.startswith("+") or line.startswith("-") or line.startswith(" ")):
-                current_hunk.lines.append(line)
-
-        if current_hunk:
-            self._classify_hunk(current_hunk, fc)
-            fc.hunks.append(current_hunk)
-
-        return fc
-
-    def _classify_hunk(self, hunk: Hunk, fc: FileChange):
-        """Classify a hunk and count lines."""
-        added = sum(1 for l in hunk.lines if l.startswith("+"))
-        deleted = sum(1 for l in hunk.lines if l.startswith("-"))
-
-        fc.added_lines += added
-        fc.deleted_lines += deleted
-
-        if added > 0 and deleted == 0:
-            hunk.category = ChangeCategory.ADDED
-        elif deleted > 0 and added == 0:
-            hunk.category = ChangeCategory.DELETED
-        elif added > 0 and deleted > 0:
-            hunk.category = ChangeCategory.MODIFIED
-        else:
-            hunk.category = ChangeCategory.CONTEXT

scripts/session_pair_harvester.py

@@ -0,0 +1,234 @@
+#!/usr/bin/env python3
+"""
+Session Transcript → Training Pair Harvester
+
+Scans Hermes session JSONL files for Q&A patterns and extracts
+terse→rich training pairs. Outputs JSONL matching the timmy-config
+training pairs spec.
+
+Usage:
+  python3 scripts/session_pair_harvester.py ~/.hermes/sessions/
+  python3 scripts/session_pair_harvester.py session.jsonl --output pairs.jsonl
+  python3 scripts/session_pair_harvester.py --dir ~/.hermes/sessions/ --min-ratio 2.0
+
+Output format:
+  {"terse": "user short prompt", "rich": "ai detailed response", "source": "session_id", "model": "..."}
+"""
+
+import argparse
+import hashlib
+import json
+import sys
+from pathlib import Path
+from typing import Optional
+
+
+def compute_hash(text: str) -> str:
+    """Content hash for deduplication."""
+    return hashlib.sha256(text.encode()).hexdigest()[:16]
+
+
+def extract_pairs_from_session(session_data: dict, min_ratio: float = 1.5,
+                                min_response_words: int = 20) -> list:
+    """Extract terse→rich pairs from a single session object."""
+    pairs = []
+    conversations = session_data.get("conversations", [])
+    session_id = session_data.get("id", "unknown")
+    model = session_data.get("model", "unknown")
+
+    seen_hashes = set()
+
+    for i, msg in enumerate(conversations):
+        # Look for assistant/gpt responses
+        if msg.get("from") not in ("gpt", "assistant"):
+            continue
+
+        response_text = msg.get("value", "")
+        if not response_text or len(response_text.split()) < min_response_words:
+            continue
+
+        # Find the preceding human message
+        prompt_text = ""
+        for j in range(i - 1, -1, -1):
+            if conversations[j].get("from") == "human":
+                prompt_text = conversations[j].get("value", "")
+                break
+
+        if not prompt_text:
+            continue
+
+        # Filter: skip tool results, system messages embedded as human
+        if prompt_text.startswith("{") and "output" in prompt_text[:100]:
+            continue  # likely a tool result
+        if prompt_text.startswith("# SOUL.md") or prompt_text.startswith("You are"):
+            continue  # system prompt leak
+
+        # Quality filters
+        prompt_words = len(prompt_text.split())
+        response_words = len(response_text.split())
+
+        # Must have meaningful length ratio
+        if prompt_words == 0 or response_words == 0:
+            continue
+        ratio = response_words / prompt_words
+        if ratio < min_ratio:
+            continue
+
+        # Skip responses that are mostly code
+        code_blocks = response_text.count("```")
+        if code_blocks >= 4 and len(response_text.replace("```", "").strip()) < 50:
+            continue
+
+        # Skip responses with tool call artifacts
+        if "tool_call" in response_text[:100] or "function_call" in response_text[:100]:
+            continue
+
+        # Deduplicate by content hash
+        content_hash = compute_hash(prompt_text + response_text[:200])
+        if content_hash in seen_hashes:
+            continue
+        seen_hashes.add(content_hash)
+
+        # Clean up response: remove markdown headers if too many
+        clean_response = response_text
+
+        pairs.append({
+            "terse": prompt_text.strip(),
+            "rich": clean_response.strip(),
+            "source": session_id,
+            "model": model,
+            "prompt_words": prompt_words,
+            "response_words": response_words,
+            "ratio": round(ratio, 2),
+        })
+
+    return pairs
+
+
+def extract_from_jsonl_file(filepath: str, **kwargs) -> list:
+    """Extract pairs from a session JSONL file."""
+    pairs = []
+    path = Path(filepath)
+
+    if not path.exists():
+        print(f"Warning: {filepath} not found", file=sys.stderr)
+        return pairs
+
+    content = path.read_text()
+    lines = content.strip().split("\n")
+
+    for line in lines:
+        line = line.strip()
+        if not line:
+            continue
+        try:
+            session = json.loads(line)
+        except json.JSONDecodeError:
+            continue
+
+        session_pairs = extract_pairs_from_session(session, **kwargs)
+        pairs.extend(session_pairs)
+
+    return pairs
+
+
+def deduplicate_pairs(pairs: list) -> list:
+    """Remove duplicate pairs across files."""
+    seen = set()
+    unique = []
+    for pair in pairs:
+        key = compute_hash(pair["terse"] + pair["rich"][:200])
+        if key not in seen:
+            seen.add(key)
+            unique.append(pair)
+    return unique
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Harvest training pairs from session transcripts")
+    parser.add_argument("input", nargs="?", help="Session JSONL file or directory")
+    parser.add_argument("--dir", "-d", help="Directory to scan for session files")
+    parser.add_argument("--output", "-o", default="harvested_pairs.jsonl", help="Output file")
+    parser.add_argument("--min-ratio", type=float, default=1.5, help="Min response/prompt word ratio")
+    parser.add_argument("--min-words", type=int, default=20, help="Min response word count")
+    parser.add_argument("--dry-run", action="store_true", help="Print stats without writing")
+    args = parser.parse_args()
+
+    all_pairs = []
+    files_scanned = 0
+
+    scan_dir = args.dir or args.input
+    if not scan_dir:
+        parser.print_help()
+        sys.exit(1)
+
+    scan_path = Path(scan_dir)
+    if scan_path.is_dir():
+        jsonl_files = sorted(scan_path.rglob("*.jsonl"))
+        print(f"Scanning {len(jsonl_files)} files in {scan_dir}...", file=sys.stderr)
+        for fpath in jsonl_files:
+            pairs = extract_from_jsonl_file(
+                str(fpath),
+                min_ratio=args.min_ratio,
+                min_response_words=args.min_words
+            )
+            all_pairs.extend(pairs)
+            files_scanned += 1
+    else:
+        pairs = extract_from_jsonl_file(
+            str(scan_path),
+            min_ratio=args.min_ratio,
+            min_response_words=args.min_words
+        )
+        all_pairs.extend(pairs)
+        files_scanned = 1
+
+    # Deduplicate
+    unique_pairs = deduplicate_pairs(all_pairs)
+
+    # Stats
+    if unique_pairs:
+        avg_prompt = sum(p["prompt_words"] for p in unique_pairs) / len(unique_pairs)
+        avg_response = sum(p["response_words"] for p in unique_pairs) / len(unique_pairs)
+        avg_ratio = sum(p["ratio"] for p in unique_pairs) / len(unique_pairs)
+    else:
+        avg_prompt = avg_response = avg_ratio = 0
+
+    stats = {
+        "files_scanned": files_scanned,
+        "raw_pairs": len(all_pairs),
+        "unique_pairs": len(unique_pairs),
+        "duplicates_removed": len(all_pairs) - len(unique_pairs),
+        "avg_prompt_words": round(avg_prompt, 1),
+        "avg_response_words": round(avg_response, 1),
+        "avg_ratio": round(avg_ratio, 2),
+    }
+
+    print(json.dumps(stats, indent=2), file=sys.stderr)
+
+    if args.dry_run:
+        # Print sample pairs
+        for pair in unique_pairs[:3]:
+            print(f"\n--- Source: {pair['source']} (ratio: {pair['ratio']}) ---", file=sys.stderr)
+            print(f"TERSE: {pair['terse'][:100]}...", file=sys.stderr)
+            print(f"RICH: {pair['rich'][:150]}...", file=sys.stderr)
+        return
+
+    # Write output
+    output_path = Path(args.output)
+    with open(output_path, "w") as f:
+        for pair in unique_pairs:
+            # Strip internal fields for output
+            output = {
+                "terse": pair["terse"],
+                "rich": pair["rich"],
+                "source": pair["source"],
+                "model": pair["model"],
+            }
+            f.write(json.dumps(output) + "\n")
+
+    print(f"\nWrote {len(unique_pairs)} pairs to {output_path}", file=sys.stderr)
+
+
+if __name__ == "__main__":
+    main()

scripts/test_diff_analyzer.py

@@ -1,189 +0,0 @@
-#!/usr/bin/env python3
-"""Tests for scripts/diff_analyzer.py — 10 tests."""
-
-import sys
-import os
-sys.path.insert(0, os.path.dirname(__file__) or ".")
-
-import importlib.util
-spec = importlib.util.spec_from_file_location("da", os.path.join(os.path.dirname(__file__) or ".", "diff_analyzer.py"))
-mod = importlib.util.module_from_spec(spec)
-spec.loader.exec_module(mod)
-DiffAnalyzer = mod.DiffAnalyzer
-ChangeCategory = mod.ChangeCategory
-
-
-SAMPLE_ADD = """diff --git a/new.py b/new.py
-new file mode 100644
---- /dev/null
-+++ b/new.py
-@@ -0,0 +1,3 @@
-+def hello():
-+    print("world")
-+    return True
-"""
-
-SAMPLE_DELETE = """diff --git a/old.py b/old.py
-deleted file mode 100644
---- a/old.py
-+++ /dev/null
-@@ -1,2 +0,0 @@
--def goodbye():
--    pass
-"""
-
-SAMPLE_MODIFY = """diff --git a/app.py b/app.py
---- a/app.py
-+++ b/app.py
-@@ -1,3 +1,4 @@
- def main():
--    print("old")
-+    print("new")
-+    print("extra")
-     return 0
-"""
-
-SAMPLE_RENAME = """diff --git a/old_name.py b/new_name.py
-rename from old_name.py
-rename to new_name.py
---- a/old_name.py
-+++ b/new_name.py
-@@ -1,1 +1,1 @@
--old content
-+new content
-"""
-
-SAMPLE_MULTI = """diff --git a/a.py b/a.py
---- a/a.py
-+++ b/a.py
-@@ -1,1 +1,2 @@
- existing
-+added line
-diff --git b/b.py b/b.py
-new file mode 100644
---- /dev/null
-+++ b/b.py
-@@ -0,0 +1,1 @@
-+new file
-"""
-
-SAMPLE_BINARY = """diff --git a/img.png b/img.png
-Binary files a/img.png and b/img.png differ
-"""
-
-
-def test_empty():
-    a = DiffAnalyzer()
-    s = a.analyze("")
-    assert s.total_files_changed == 0
-    print("PASS: test_empty")
-
-def test_addition():
-    a = DiffAnalyzer()
-    s = a.analyze(SAMPLE_ADD)
-    assert s.total_files_changed == 1
-    assert s.total_added == 3
-    assert s.total_deleted == 0
-    assert s.new_files == 1
-    assert s.files[0].hunks[0].category == ChangeCategory.ADDED
-    print("PASS: test_addition")
-
-def test_deletion():
-    a = DiffAnalyzer()
-    s = a.analyze(SAMPLE_DELETE)
-    assert s.total_deleted == 2
-    assert s.deleted_files == 1
-    assert s.files[0].hunks[0].category == ChangeCategory.DELETED
-    print("PASS: test_deletion")
-
-def test_modification():
-    a = DiffAnalyzer()
-    s = a.analyze(SAMPLE_MODIFY)
-    assert s.total_added == 2
-    assert s.total_deleted == 1
-    assert s.files[0].hunks[0].category == ChangeCategory.MODIFIED
-    print("PASS: test_modification")
-
-def test_rename():
-    a = DiffAnalyzer()
-    s = a.analyze(SAMPLE_RENAME)
-    assert s.renamed_files == 1
-    assert s.files[0].old_path == "old_name.py"
-    assert s.files[0].path == "new_name.py"
-    assert s.files[0].is_renamed == True
-    print("PASS: test_rename")
-
-def test_multiple_files():
-    a = DiffAnalyzer()
-    s = a.analyze(SAMPLE_MULTI)
-    assert s.total_files_changed == 2
-    assert s.new_files == 1
-    print("PASS: test_multiple_files")
-
-def test_binary():
-    a = DiffAnalyzer()
-    s = a.analyze(SAMPLE_BINARY)
-    assert s.binary_files == 1
-    assert s.files[0].is_binary == True
-    assert len(s.files[0].hunks) == 0
-    print("PASS: test_binary")
-
-def test_to_dict():
-    a = DiffAnalyzer()
-    s = a.analyze(SAMPLE_MODIFY)
-    d = s.to_dict()
-    assert "total_files_changed" in d
-    assert "files" in d
-    assert isinstance(d["files"], list)
-    print("PASS: test_to_dict")
-
-def test_context_only():
-    diff = """diff --git a/f.py b/f.py
---- a/f.py
-+++ b/f.py
-@@ -1,3 +1,3 @@
- line1
--old
-+new
- line3
-"""
-    a = DiffAnalyzer()
-    s = a.analyze(diff)
-    # Has both added and deleted = MODIFIED
-    assert s.files[0].hunks[0].category == ChangeCategory.MODIFIED
-    print("PASS: test_context_only")
-
-def test_multi_hunk():
-    diff = """diff --git a/f.py b/f.py
---- a/f.py
-+++ b/f.py
-@@ -1,1 +1,2 @@
- existing
-+first addition
-@@ -10,1 +11,2 @@
- more
-+second addition
-"""
-    a = DiffAnalyzer()
-    s = a.analyze(diff)
-    assert s.total_hunks == 2
-    assert s.total_added == 2
-    print("PASS: test_multi_hunk")
-
-
-def run_all():
-    test_empty()
-    test_addition()
-    test_deletion()
-    test_modification()
-    test_rename()
-    test_multiple_files()
-    test_binary()
-    test_to_dict()
-    test_context_only()
-    test_multi_hunk()
-    print("\nAll 10 tests passed!")
-
-
-if __name__ == "__main__":
-    run_all()

scripts/test_session_pair_harvester.py

@@ -0,0 +1,90 @@
+#!/usr/bin/env python3
+"""Tests for session_pair_harvester."""
+
+import json
+import sys
+import os
+import tempfile
+
+sys.path.insert(0, os.path.dirname(__file__))
+from session_pair_harvester import extract_pairs_from_session, deduplicate_pairs, compute_hash
+
+
+def test_basic_extraction():
+    session = {
+        "id": "test_001",
+        "model": "test-model",
+        "conversations": [
+            {"from": "system", "value": "You are helpful."},
+            {"from": "human", "value": "What is Python?"},
+            {"from": "gpt", "value": "Python is a high-level programming language known for its readability and versatility. It supports multiple paradigms including procedural, object-oriented, and functional programming. Python is widely used in web development, data science, machine learning, and automation."},
+        ]
+    }
+    pairs = extract_pairs_from_session(session, min_ratio=1.5, min_response_words=10)
+    assert len(pairs) == 1
+    assert pairs[0]["terse"] == "What is Python?"
+    assert "programming language" in pairs[0]["rich"]
+    assert pairs[0]["source"] == "test_001"
+    print("PASS: test_basic_extraction")
+
+
+def test_filters_short_responses():
+    session = {
+        "id": "test_002",
+        "model": "test",
+        "conversations": [
+            {"from": "human", "value": "Hi"},
+            {"from": "gpt", "value": "Hello!"},
+        ]
+    }
+    pairs = extract_pairs_from_session(session, min_ratio=1.5, min_response_words=20)
+    assert len(pairs) == 0
+    print("PASS: test_filters_short_responses")
+
+
+def test_skips_tool_results():
+    session = {
+        "id": "test_003",
+        "model": "test",
+        "conversations": [
+            {"from": "human", "value": '{"output": "file content", "exit_code": 0}'},
+            {"from": "gpt", "value": "The file was read successfully. Now let me analyze the content and provide a detailed summary of what was found in the file system."},
+        ]
+    }
+    pairs = extract_pairs_from_session(session, min_ratio=1.5, min_response_words=10)
+    assert len(pairs) == 0
+    print("PASS: test_skips_tool_results")
+
+
+def test_deduplication():
+    pairs = [
+        {"terse": "What is X?", "rich": "X is Y.", "source": "s1", "model": "m"},
+        {"terse": "What is X?", "rich": "X is Y.", "source": "s2", "model": "m"},
+        {"terse": "What is Z?", "rich": "Z is W.", "source": "s1", "model": "m"},
+    ]
+    unique = deduplicate_pairs(pairs)
+    assert len(unique) == 2
+    print("PASS: test_deduplication")
+
+
+def test_ratio_filter():
+    session = {
+        "id": "test_005",
+        "model": "test",
+        "conversations": [
+            {"from": "human", "value": "Explain quantum computing in detail with examples and applications"},
+            {"from": "gpt", "value": "OK."},
+        ]
+    }
+    pairs = extract_pairs_from_session(session, min_ratio=1.5, min_response_words=10)
+    assert len(pairs) == 0  # response too short relative to prompt
+    print("PASS: test_ratio_filter")
+
+
+if __name__ == "__main__":
+    test_basic_extraction()
+    test_filters_short_responses()
+    test_skips_tool_results()
+    test_deduplication()
+    test_ratio_filter()
+    print("\nAll tests passed.")