claude-code-log 1.1.1__tar.gz → 1.2.0__tar.gz

{claude_code_log-1.1.1 → claude_code_log-1.2.0}/CHANGELOG.md

@@ -6,6 +6,25 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 
+## [1.2.0] - 2026-04-19
+
+### Changed
+
+- **Preserve agentId anchors in parallel-Task stitch + tool-param UI fix (#115)**
+- **Per-level output files for --detail and --compact (#114)**
+- **Handle custom-title, agent-name, and agent-color transcript entry types (#113)**
+- **Ignore 'last-prompt' message type (#112)**
+- **Detail levels and compact rendering of conversations (#96)**
+- **Skip PassthroughTranscriptEntry in _render_messages**
+- **Integrate agent transcripts into the DAG (Phase C) (#99)**
+- **Implement DAG-based message ordering (Phases A+B) (#97)**
+- **Fix slow test hitting real ~/.claude/projects (5GB) (#109)**
+- **feat: add --session-id flag for exporting a single  (#103)**
+- **Fix search broken when HTML saved with different filename (#106)**
+- **Add Grep tool renderer with pattern in title (#107)**
+- **Fix TUI square bracket escaping issue (#105)**
+
+
 ## [1.1.1] - 2026-03-10
 
 ### Changed

{claude_code_log-1.1.1 → claude_code_log-1.2.0}/CLAUDE.md

@@ -47,6 +47,8 @@ claude-code-log --from-date "last week"
 
 See @CONTRIBUTING.md for detailed development setup, testing, architecture, and release process.
 
+**Before pushing, always remind the user to run `just ci`.**
+
 ### Claude-Specific Testing Tips
 
 **Always use `-n auto` for parallel test execution:**

{claude_code_log-1.1.1 → claude_code_log-1.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: claude-code-log
-Version: 1.1.1
+Version: 1.2.0
 Summary: Convert Claude Code transcript JSONL files to HTML
 Project-URL: Homepage, https://github.com/daaain/claude-code-log
 Project-URL: Issues, https://github.com/daaain/claude-code-log/issues
@@ -66,6 +66,7 @@ uvx claude-code-log@latest --open-browser
 - **Rich Message Types**: Support for user/assistant messages, tool use/results, thinking content, images
 - **System Command Visibility**: Show system commands (like `init`) in expandable details with structured parsing
 - **Markdown Rendering**: Server-side markdown rendering with syntax highlighting using mistune
+- **Detail Levels & Compact Mode**: `--detail full|high|low|minimal` filters by verbosity and `--compact` merges repeated section headings — pairs well with `--format md` to feed past conversations back to an LLM for analysis or experience building
 - **Floating Navigation**: Always-available back-to-top button and filter controls
 - **CLI Interface**: Simple command-line tool using Click
 
@@ -81,6 +82,7 @@ This tool helps you answer questions like:
 - **"How can I share my Claude Code conversation with others?"**
 - **"What's the timeline of my project development?"**
 - **"How can I analyse patterns in my Claude Code usage?"**
+- **"How can I feed a past session back to an LLM for analysis or experience building?"**
 
 ## Usage
 
@@ -164,6 +166,27 @@ claude-code-log /path/to/directory --from-date "yesterday" --to-date "today"
 claude-code-log /path/to/directory --from-date "3 days ago" --to-date "yesterday"
 ```
 
+### Feeding Past Conversations to an LLM
+
+The combination `--detail low --format md --compact` produces condensed Markdown suitable as context for an LLM to review or distill patterns from past work:
+
+```bash
+# Session → compact Markdown for LLM review
+claude-code-log transcript.jsonl --detail low --format md --compact -o session.md
+
+# Whole project history
+claude-code-log /path/to/project --detail low --format md --compact
+```
+
+`--detail` levels (smallest → largest output):
+
+- `minimal` — user + assistant text only
+- `low` — interaction-focused; keeps WebSearch, WebFetch, and Task (agent delegations) as key signals
+- `high` — detailed but cleaned; drops system/hook noise
+- `full` — everything (default)
+
+`--compact` merges consecutive same-type sections in Markdown so runs of assistant responses share one heading instead of repeating `### 🤖 Assistant:` for each.
+
 ## Project Hierarchy Output
 
 When processing all projects, the tool generates:
@@ -228,6 +251,7 @@ Markdown export provides a lightweight, portable alternative to HTML:
 - **Code Preservation**: Syntax highlighting hints via fenced code blocks
 - **Embedded Viewer**: TUI includes built-in Markdown viewer with table of contents
 - **Image Support**: Configurable image handling (placeholder, embedded base64, or referenced files)
+- **`--compact` Mode**: Merge consecutive same-type section headings — most useful with `--detail low` or `minimal` where tool stripping creates runs of Assistant or User sections
 
 ## Installation
 

{claude_code_log-1.1.1 → claude_code_log-1.2.0}/README.md

@@ -42,6 +42,7 @@ uvx claude-code-log@latest --open-browser
 - **Rich Message Types**: Support for user/assistant messages, tool use/results, thinking content, images
 - **System Command Visibility**: Show system commands (like `init`) in expandable details with structured parsing
 - **Markdown Rendering**: Server-side markdown rendering with syntax highlighting using mistune
+- **Detail Levels & Compact Mode**: `--detail full|high|low|minimal` filters by verbosity and `--compact` merges repeated section headings — pairs well with `--format md` to feed past conversations back to an LLM for analysis or experience building
 - **Floating Navigation**: Always-available back-to-top button and filter controls
 - **CLI Interface**: Simple command-line tool using Click
 
@@ -57,6 +58,7 @@ This tool helps you answer questions like:
 - **"How can I share my Claude Code conversation with others?"**
 - **"What's the timeline of my project development?"**
 - **"How can I analyse patterns in my Claude Code usage?"**
+- **"How can I feed a past session back to an LLM for analysis or experience building?"**
 
 ## Usage
 
@@ -140,6 +142,27 @@ claude-code-log /path/to/directory --from-date "yesterday" --to-date "today"
 claude-code-log /path/to/directory --from-date "3 days ago" --to-date "yesterday"
 ```
 
+### Feeding Past Conversations to an LLM
+
+The combination `--detail low --format md --compact` produces condensed Markdown suitable as context for an LLM to review or distill patterns from past work:
+
+```bash
+# Session → compact Markdown for LLM review
+claude-code-log transcript.jsonl --detail low --format md --compact -o session.md
+
+# Whole project history
+claude-code-log /path/to/project --detail low --format md --compact
+```
+
+`--detail` levels (smallest → largest output):
+
+- `minimal` — user + assistant text only
+- `low` — interaction-focused; keeps WebSearch, WebFetch, and Task (agent delegations) as key signals
+- `high` — detailed but cleaned; drops system/hook noise
+- `full` — everything (default)
+
+`--compact` merges consecutive same-type sections in Markdown so runs of assistant responses share one heading instead of repeating `### 🤖 Assistant:` for each.
+
 ## Project Hierarchy Output
 
 When processing all projects, the tool generates:
@@ -204,6 +227,7 @@ Markdown export provides a lightweight, portable alternative to HTML:
 - **Code Preservation**: Syntax highlighting hints via fenced code blocks
 - **Embedded Viewer**: TUI includes built-in Markdown viewer with table of contents
 - **Image Support**: Configurable image handling (placeholder, embedded base64, or referenced files)
+- **`--compact` Mode**: Merge consecutive same-type section headings — most useful with `--detail low` or `minimal` where tool stripping creates runs of Assistant or User sections
 
 ## Installation
 

{claude_code_log-1.1.1 → claude_code_log-1.2.0}/claude_code_log/cache.py

@@ -1140,8 +1140,17 @@ class CacheManager:
 
         return row["page_size_config"] if row else None
 
-    def get_page_data(self, page_number: int) -> Optional[PageCacheData]:
-        """Get cache data for a specific page."""
+    def get_page_data(
+        self, page_number: int, variant_suffix: str = ""
+    ) -> Optional[PageCacheData]:
+        """Get cache data for a specific page.
+
+        Args:
+            page_number: Page number (1-indexed).
+            variant_suffix: Detail/compact variant infix (e.g. ``""``,
+                ``".low"``, ``".low.compact"``). Each variant has its
+                own pagination cache row.
+        """
         if self._project_id is None:
             return None
 
@@ -1149,8 +1158,9 @@ class CacheManager:
             # Get page info
             page_row = conn.execute(
                 """SELECT * FROM html_pages
-                   WHERE project_id = ? AND page_number = ?""",
-                (self._project_id, page_number),
+                   WHERE project_id = ? AND variant_suffix = ?
+                         AND page_number = ?""",
+                (self._project_id, variant_suffix, page_number),
             ).fetchone()
 
             if not page_row:
@@ -1247,8 +1257,16 @@ class CacheManager:
         total_output_tokens: int,
         total_cache_creation_tokens: int,
         total_cache_read_tokens: int,
+        variant_suffix: str = "",
     ) -> None:
-        """Update or insert page cache entry."""
+        """Update or insert page cache entry.
+
+        Args:
+            variant_suffix: Detail/compact variant infix (e.g. ``""``,
+                ``".low"``). Each variant owns its own cache row; the
+                UNIQUE constraint is on (project_id, variant_suffix,
+                page_number).
+        """
         if self._project_id is None or not session_ids:
             return
 
@@ -1256,13 +1274,14 @@ class CacheManager:
             # Insert or update page
             conn.execute(
                 """INSERT INTO html_pages
-                   (project_id, page_number, html_path, page_size_config, message_count,
+                   (project_id, variant_suffix, page_number, html_path,
+                    page_size_config, message_count,
                     first_session_id, last_session_id, first_timestamp, last_timestamp,
                     total_input_tokens, total_output_tokens,
                     total_cache_creation_tokens, total_cache_read_tokens,
                     generated_at, library_version)
-                   VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
-                   ON CONFLICT(project_id, page_number)
+                   VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
+                   ON CONFLICT(project_id, variant_suffix, page_number)
                    DO UPDATE SET
                        html_path = excluded.html_path,
                        page_size_config = excluded.page_size_config,
@@ -1279,6 +1298,7 @@ class CacheManager:
                        library_version = excluded.library_version""",
                 (
                     self._project_id,
+                    variant_suffix,
                     page_number,
                     html_path,
                     page_size_config,
@@ -1299,8 +1319,9 @@ class CacheManager:
             # Get the page ID
             row = conn.execute(
                 """SELECT id FROM html_pages
-                   WHERE project_id = ? AND page_number = ?""",
-                (self._project_id, page_number),
+                   WHERE project_id = ? AND variant_suffix = ?
+                         AND page_number = ?""",
+                (self._project_id, variant_suffix, page_number),
             ).fetchone()
             page_id = row["id"]
 
@@ -1318,13 +1339,18 @@ class CacheManager:
             conn.commit()
 
     def is_page_stale(
-        self, page_number: int, page_size_config: int
+        self,
+        page_number: int,
+        page_size_config: int,
+        variant_suffix: str = "",
     ) -> tuple[bool, str]:
         """Check if a page needs regeneration.
 
         Args:
-            page_number: The page number to check
-            page_size_config: The current page size configuration
+            page_number: The page number to check.
+            page_size_config: The current page size configuration.
+            variant_suffix: Detail/compact variant infix; each variant's
+                cache is checked independently.
 
         Returns:
             Tuple of (is_stale: bool, reason: str)
@@ -1334,7 +1360,7 @@ class CacheManager:
         if self._project_id is None:
             return True, "no_cache"
 
-        page_data = self.get_page_data(page_number)
+        page_data = self.get_page_data(page_number, variant_suffix)
         if page_data is None:
             return True, "not_cached"
 
@@ -1409,15 +1435,21 @@ class CacheManager:
 
         return html_paths
 
-    def get_page_count(self) -> int:
-        """Get the number of cached pages for this project."""
+    def get_page_count(self, variant_suffix: str = "") -> int:
+        """Get the number of cached pages for this project and variant.
+
+        Orphan-cleanup loops in the converter must pass the same
+        ``variant_suffix`` used for rendering, otherwise they'd count
+        pages across variants and delete live pages of another variant.
+        """
         if self._project_id is None:
             return 0
 
         with self._get_connection() as conn:
             row = conn.execute(
-                """SELECT COUNT(*) as cnt FROM html_pages WHERE project_id = ?""",
-                (self._project_id,),
+                """SELECT COUNT(*) as cnt FROM html_pages
+                   WHERE project_id = ? AND variant_suffix = ?""",
+                (self._project_id, variant_suffix),
             ).fetchone()
 
         return row["cnt"] if row else 0
@@ -1556,6 +1588,47 @@ def get_all_cached_projects(
     return result
 
 
+def find_session_in_cache(
+    session_id: str,
+    projects_dir: Path,
+    db_path: Optional[Path] = None,
+) -> list[tuple[str, str]]:
+    """Find a session by ID or prefix across all projects in cache.
+
+    Standalone function that queries the cache database directly.
+
+    Args:
+        session_id: Full session ID or prefix to match.
+        projects_dir: Path to the projects directory (for default DB location).
+        db_path: Optional explicit path to the cache database.
+
+    Returns:
+        List of (project_path, full_session_id) tuples for all matches.
+    """
+    actual_db_path = db_path or get_cache_db_path(projects_dir)
+    if not actual_db_path.exists():
+        return []
+
+    try:
+        conn = sqlite3.connect(actual_db_path, timeout=30.0)
+        conn.row_factory = sqlite3.Row
+        try:
+            rows = conn.execute(
+                """SELECT p.project_path, s.session_id
+                   FROM sessions s
+                   JOIN projects p ON s.project_id = p.id
+                   WHERE s.session_id = ? OR s.session_id LIKE ?
+                   ORDER BY s.first_timestamp DESC""",
+                (session_id, f"{session_id}%"),
+            ).fetchall()
+            return [(row["project_path"], row["session_id"]) for row in rows]
+        finally:
+            conn.close()
+    except (sqlite3.Error, OSError) as e:
+        logger.debug("Failed to find session in cache: %s", e)
+        return []
+
+
 __all__ = [
     "CacheManager",
     "CachedFileInfo",
@@ -1563,6 +1636,7 @@ __all__ = [
     "PageCacheData",
     "ProjectCache",
     "SessionCacheData",
+    "find_session_in_cache",
     "get_all_cached_projects",
     "get_cache_db_path",
     "get_library_version",

{claude_code_log-1.1.1 → claude_code_log-1.2.0}/claude_code_log/cli.py

@@ -14,11 +14,13 @@ from .converter import (
     convert_jsonl_to,
     convert_jsonl_to_html,
     ensure_fresh_cache,
+    generate_single_session_file,
     get_file_extension,
     process_projects_hierarchy,
 )
 from .cache import (
     CacheManager,
+    find_session_in_cache,
     get_all_cached_projects,
     get_cache_db_path,
     get_library_version,
@@ -506,6 +508,31 @@ def _clear_output_files(input_path: Path, all_projects: bool, file_ext: str) ->
     default=2000,
     help="Maximum messages per page for combined transcript (default: 2000). Sessions are never split across pages.",
 )
+@click.option(
+    "--session-id",
+    default=None,
+    help="Export a single session by ID (full ID or prefix). Project path is optional — looks up the session globally via cache.",
+)
+@click.option(
+    "--detail",
+    type=click.Choice(["full", "high", "low", "minimal"], case_sensitive=False),
+    default="full",
+    help=(
+        "Detail level for output. "
+        "full: everything; "
+        "high: detailed but cleaned (no system/hook noise); "
+        "low: interaction-focused + key signals; "
+        "minimal: user + assistant messages only."
+    ),
+)
+@click.option(
+    "--compact",
+    is_flag=True,
+    help=(
+        "Merge consecutive same-category headings in Markdown output. "
+        "Markdown-only — a no-op for HTML."
+    ),
+)
 @click.option(
     "--debug",
     is_flag=True,
@@ -528,6 +555,9 @@ def main(
     output_format: str,
     image_export_mode: Optional[str],
     page_size: int,
+    session_id: Optional[str],
+    detail: str,
+    compact: bool,
     debug: bool,
 ) -> None:
     """Convert Claude transcript JSONL files to HTML or Markdown.
@@ -537,6 +567,10 @@ def main(
     # Configure logging to show warnings and above
     logging.basicConfig(level=logging.WARNING, format="%(levelname)s: %(message)s")
 
+    from .models import DetailLevel
+
+    detail_level = DetailLevel(detail.lower())
+
     try:
         # Handle TUI mode
         if tui:
@@ -648,6 +682,60 @@ def main(
                 _launch_tui_with_cache_check(input_path)
                 return
 
+        # Handle --session-id: export a single session by ID
+        if session_id is not None:
+            if input_path is None:
+                # Global lookup via cache
+                effective_projects_dir = projects_dir or get_default_projects_dir()
+                matches = find_session_in_cache(session_id, effective_projects_dir)
+                if not matches:
+                    click.echo(
+                        f"Error: Session '{session_id}' not found in cache. "
+                        "Try providing a project directory path, or run "
+                        "claude-code-log first to populate the cache.",
+                        err=True,
+                    )
+                    sys.exit(1)
+                if len(matches) > 1:
+                    # Check if all matches resolve to the same session ID
+                    unique_ids = {m[1] for m in matches}
+                    if len(unique_ids) > 1:
+                        click.echo(
+                            f"Error: Ambiguous session ID prefix '{session_id}' "
+                            "matches multiple sessions:",
+                            err=True,
+                        )
+                        for proj_path, sid in matches:
+                            click.echo(f"  {sid[:8]} in {proj_path}", err=True)
+                        sys.exit(1)
+                input_path = Path(matches[0][0])
+                session_id = matches[0][1]
+            else:
+                # Convert project path if needed
+                if not input_path.exists() or (
+                    input_path.is_dir() and not list(input_path.glob("*.jsonl"))
+                ):
+                    claude_path = convert_project_path_to_claude_dir(
+                        input_path, projects_dir
+                    )
+                    if claude_path.exists():
+                        input_path = claude_path
+
+            output_path = generate_single_session_file(
+                output_format,
+                input_path,
+                session_id,
+                output,
+                not no_cache,
+                image_export_mode,
+                detail=detail_level,
+                compact=compact,
+            )
+            click.echo(f"Successfully exported session to {output_path}")
+            if open_browser:
+                click.launch(str(output_path))
+            return
+
         # Handle default case - process all projects hierarchy if no input path and --all-projects flag
         if input_path is None:
             input_path = projects_dir or get_default_projects_dir()
@@ -685,6 +773,8 @@ def main(
                 output_format,
                 image_export_mode,
                 page_size=page_size,
+                detail=detail_level,
+                compact=compact,
             )
 
             # Count processed projects
@@ -737,6 +827,11 @@ def main(
             not no_cache,
             image_export_mode=image_export_mode,
             page_size=page_size,
+            detail=detail_level,
+            compact=compact,
+            # User's `-o` path is a one-off export, not a cached artifact:
+            # don't occupy a cache slot keyed by an arbitrary destination.
+            update_cache=output is None,
         )
         if input_path.is_file():
             click.echo(f"Successfully converted {input_path} to {output_path}")