notepad-cleanup 0.2.1__tar.gz → 0.2.3__tar.gz

notepad_cleanup-0.2.3/PKG-INFO

@@ -0,0 +1,208 @@
+Metadata-Version: 2.4
+Name: notepad-cleanup
+Version: 0.2.3
+Summary: Extract and organize text from Windows 11 Notepad tabs using AI
+Home-page: https://github.com/DazzleTools/notepad-cleanup
+Author: djdarcy
+Author-email: djdarcy <6962246+djdarcy@users.noreply.github.com>
+License: GPL-3.0-or-later
+Project-URL: Homepage, https://github.com/DazzleTools/notepad-cleanup
+Project-URL: Repository, https://github.com/DazzleTools/notepad-cleanup
+Project-URL: Issues, https://github.com/DazzleTools/notepad-cleanup/issues
+Project-URL: Discussions, https://github.com/DazzleTools/notepad-cleanup/discussions
+Project-URL: Changelog, https://github.com/DazzleTools/notepad-cleanup/blob/main/CHANGELOG.md
+Keywords: windows,notepad,extraction,organization,ai,dedup,tabs,clipboard
+Classifier: Development Status :: 3 - Alpha
+Classifier: Environment :: Console
+Classifier: Intended Audience :: End Users/Desktop
+Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
+Classifier: Operating System :: Microsoft :: Windows
+Classifier: Operating System :: Microsoft :: Windows :: Windows 11
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Utilities
+Classifier: Topic :: Text Processing
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: pywinauto>=0.6.9
+Requires-Dist: pywin32>=306
+Requires-Dist: psutil>=5.9.0
+Requires-Dist: click>=8.1.0
+Requires-Dist: rich>=13.0.0
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0.0; extra == "dev"
+Requires-Dist: pytest-cov>=4.0.0; extra == "dev"
+Dynamic: author
+Dynamic: home-page
+Dynamic: license-file
+Dynamic: requires-python
+
+# notepad-cleanup
+
+[![PyPI](https://img.shields.io/pypi/v/notepad-cleanup?color=green)](https://pypi.org/project/notepad-cleanup/)
+[![Release Date](https://img.shields.io/github/release-date/DazzleTools/notepad-cleanup?color=green)](https://github.com/DazzleTools/notepad-cleanup/releases)
+[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
+[![License: GPL v3](https://img.shields.io/badge/license-GPL%20v3-green.svg)](https://www.gnu.org/licenses/gpl-3.0.html)
+[![Installs](https://img.shields.io/endpoint?url=https://gist.githubusercontent.com/djdarcy/a165031b15ecf9b6fdac066d0222a591/raw/installs.json)](https://dazzletools.github.io/notepad-cleanup/stats/#installs)
+[![GitHub Discussions](https://img.shields.io/github/discussions/DazzleTools/notepad-cleanup)](https://github.com/DazzleTools/notepad-cleanup/discussions)
+[![Platform](https://img.shields.io/badge/platform-Windows%2011-lightgrey.svg)](https://github.com/DazzleTools/notepad-cleanup)
+
+Extract and organize text from all open Windows 11 Notepad tabs using AI-powered categorization.
+
+## What It Does
+
+Windows 11 Notepad supports multiple tabs, making it easy to accumulate dozens of text snippets, code fragments, notes, and temporary data across multiple windows. **notepad-cleanup** extracts all that text in one command, deduplicates it against previous sessions, and organizes it into categorized folders using AI.
+
+## Installation
+
+```bash
+pip install notepad-cleanup
+```
+
+For virtual environments, source installs, and Claude Code CLI setup, see [docs/install.md](docs/install.md).
+
+```bash
+notepad-cleanup extract                    # Extract all Notepad tabs
+notepad-cleanup compare --last --link auto # Find and link duplicates
+notepad-cleanup organize --last            # AI categorization (duplicates become symlinks)
+notepad-cleanup links separate --last      # Optional: split out linked files to see only new content
+```
+
+## Features
+
+- **Two-phase extraction** -- Silent `WM_GETTEXT` for loaded tabs, UI Automation for unloaded tabs
+- **Cross-session deduplication** -- Compare against historical sessions with exact and [fuzzy matching](docs/fuzzy-matching.md)
+- **Filesystem linking** -- Replace duplicates with hardlinks, symlinks, or DazzleLink descriptors
+- **Link-aware organization** -- AI categorizes all files, but duplicates become symlinks in `organized/` instead of copies. Preserves the connection network back to canonical sources
+- **Separate/join links** -- Split `organized/` into new files vs linked duplicates, or rejoin them
+- **Configuration system** -- Unified folder registry with [`...` notation](docs/config.md), MRU history, persistent settings
+- **Diff integration** -- Auto-generated scripts for Beyond Compare, WinMerge, VS Code, etc.
+
+## Usage
+
+### First-time setup
+
+```bash
+notepad-cleanup config add "C:\Users\YourName\Desktop\Notepad Organize"
+notepad-cleanup config set search "...1"
+notepad-cleanup config set diff_tool bcomp
+```
+
+### Daily workflow
+
+```bash
+notepad-cleanup extract                    # 1. Extract all tabs
+notepad-cleanup compare --last             # 2. Find duplicates
+notepad-cleanup diff --last                # 3. Spot-check in diff tool
+notepad-cleanup compare --last --link auto # 4. Link duplicates
+notepad-cleanup organize --last            # 5. AI categorization (dupes become symlinks)
+notepad-cleanup links separate --last      # 6. Optional: see only new files
+notepad-cleanup links join --last          # 7. Optional: rejoin everything
+```
+
+After setup, `--last` auto-uses the most recent extraction. No path copy-pasting.
+
+### Commands
+
+| Command | Purpose |
+|---------|---------|
+| `extract` | Extract text from all open Notepad windows/tabs |
+| `compare` | Find duplicates across historical sessions |
+| `organize` | AI-powered categorization (symlinks for duplicates, copies for new) |
+| `links` | Separate linked files from organized/, or rejoin them |
+| `diff` | Launch diff script to spot-check matched pairs |
+| `config` | Manage folders, search dirs, diff tool, settings |
+| `run` | Extract + organize in one step |
+
+For full parameter documentation, see [docs/parameters.md](docs/parameters.md).
+
+## Documentation
+
+| Doc | Contents |
+|-----|----------|
+| [Parameters](docs/parameters.md) | Full command reference with all options |
+| [Configuration](docs/config.md) | Folder registry, `...` notation, MRU, search dirs |
+| [Fuzzy Matching](docs/fuzzy-matching.md) | Threshold formula, derivation, customization |
+
+## Output Structure
+
+```
+notepad-cleanup/nc-2026-03-16__08-15-30/
+├── manifest.json                  # Extraction metadata
+├── window01/
+│   ├── tab01.txt                  # Raw extracted files
+│   ├── tab02.txt
+│   └── tab03.txt
+├── window02/
+│   └── tab01.txt
+├── organized/                     # AI-organized output (after organize step)
+│   ├── code-snippets/
+│   │   ├── process-data.py
+│   │   └── batch-rename.bat
+│   ├── personal-notes/
+│   │   └── grocery-list.txt
+│   └── _summary.md               # Organization summary
+├── _compare_results.json          # Dedup comparison cache
+├── _compare_diffs.cmd             # Diff script for spot-checking
+├── _dedup_links.json              # Link manifest (if --link used)
+├── _organize_prompt.md            # AI prompt used
+└── _organize_log.txt              # Claude CLI output
+```
+
+## How It Works
+
+### Phase 1: Silent Extraction
+
+Uses `WM_GETTEXT` message to read text from `RichEditD2DPT` child windows. This is completely silent and invisible -- no focus changes, no window activation, no disruption to your workflow.
+
+**Limitation:** Only works for tabs that have been loaded (visited) at least once in the current Notepad session. Unloaded tabs have no `RichEditD2DPT` control yet, so they cannot be read silently.
+
+### Phase 2: Tab Switching (Announced)
+
+For unloaded tabs, uses UI Automation (`TabItem.Select()`) to activate each tab, which forces Windows to load the `RichEditD2DPT` control. Once loaded, the same `WM_GETTEXT` method reads the content.
+
+**Warning:** This steals focus and activates Notepad windows. The tool warns you before Phase 2 starts and waits for confirmation. Do not type or click during Phase 2.
+
+### Organization with AI
+
+After extraction, Claude Code CLI:
+1. Reads `manifest.json` to understand the collection
+2. Reads each extracted file to determine content type
+3. Returns a JSON plan with categories and renamed filenames
+4. The tool executes the plan locally (copy files to organized folders)
+
+## Requirements
+
+- **Windows 11** (uses Windows 11 Notepad tab features)
+- **Python 3.10+**
+- **Claude Code CLI** (optional, for organize step)
+
+For detailed installation instructions, see [docs/install.md](docs/install.md).
+
+## Development
+
+```bash
+git clone https://github.com/DazzleTools/notepad-cleanup.git
+cd notepad-cleanup
+python -m venv venv
+venv\Scripts\activate
+pip install -e .
+```
+
+## Contributing
+
+Contributions are welcome! See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
+
+Like the project?
+
+[!["Buy Me A Coffee"](https://www.buymeacoffee.com/assets/img/custom_images/orange_img.png)](https://www.buymeacoffee.com/djdarcy)
+
+## License
+
+notepad-cleanup, Copyright (C) 2026 Dustin Darcy. 
+
+This project is licensed under the GNU General Public License v3.0 — see [LICENSE](LICENSE) for full details.

notepad_cleanup-0.2.3/README.md

@@ -0,0 +1,165 @@
+# notepad-cleanup
+
+[![PyPI](https://img.shields.io/pypi/v/notepad-cleanup?color=green)](https://pypi.org/project/notepad-cleanup/)
+[![Release Date](https://img.shields.io/github/release-date/DazzleTools/notepad-cleanup?color=green)](https://github.com/DazzleTools/notepad-cleanup/releases)
+[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
+[![License: GPL v3](https://img.shields.io/badge/license-GPL%20v3-green.svg)](https://www.gnu.org/licenses/gpl-3.0.html)
+[![Installs](https://img.shields.io/endpoint?url=https://gist.githubusercontent.com/djdarcy/a165031b15ecf9b6fdac066d0222a591/raw/installs.json)](https://dazzletools.github.io/notepad-cleanup/stats/#installs)
+[![GitHub Discussions](https://img.shields.io/github/discussions/DazzleTools/notepad-cleanup)](https://github.com/DazzleTools/notepad-cleanup/discussions)
+[![Platform](https://img.shields.io/badge/platform-Windows%2011-lightgrey.svg)](https://github.com/DazzleTools/notepad-cleanup)
+
+Extract and organize text from all open Windows 11 Notepad tabs using AI-powered categorization.
+
+## What It Does
+
+Windows 11 Notepad supports multiple tabs, making it easy to accumulate dozens of text snippets, code fragments, notes, and temporary data across multiple windows. **notepad-cleanup** extracts all that text in one command, deduplicates it against previous sessions, and organizes it into categorized folders using AI.
+
+## Installation
+
+```bash
+pip install notepad-cleanup
+```
+
+For virtual environments, source installs, and Claude Code CLI setup, see [docs/install.md](docs/install.md).
+
+```bash
+notepad-cleanup extract                    # Extract all Notepad tabs
+notepad-cleanup compare --last --link auto # Find and link duplicates
+notepad-cleanup organize --last            # AI categorization (duplicates become symlinks)
+notepad-cleanup links separate --last      # Optional: split out linked files to see only new content
+```
+
+## Features
+
+- **Two-phase extraction** -- Silent `WM_GETTEXT` for loaded tabs, UI Automation for unloaded tabs
+- **Cross-session deduplication** -- Compare against historical sessions with exact and [fuzzy matching](docs/fuzzy-matching.md)
+- **Filesystem linking** -- Replace duplicates with hardlinks, symlinks, or DazzleLink descriptors
+- **Link-aware organization** -- AI categorizes all files, but duplicates become symlinks in `organized/` instead of copies. Preserves the connection network back to canonical sources
+- **Separate/join links** -- Split `organized/` into new files vs linked duplicates, or rejoin them
+- **Configuration system** -- Unified folder registry with [`...` notation](docs/config.md), MRU history, persistent settings
+- **Diff integration** -- Auto-generated scripts for Beyond Compare, WinMerge, VS Code, etc.
+
+## Usage
+
+### First-time setup
+
+```bash
+notepad-cleanup config add "C:\Users\YourName\Desktop\Notepad Organize"
+notepad-cleanup config set search "...1"
+notepad-cleanup config set diff_tool bcomp
+```
+
+### Daily workflow
+
+```bash
+notepad-cleanup extract                    # 1. Extract all tabs
+notepad-cleanup compare --last             # 2. Find duplicates
+notepad-cleanup diff --last                # 3. Spot-check in diff tool
+notepad-cleanup compare --last --link auto # 4. Link duplicates
+notepad-cleanup organize --last            # 5. AI categorization (dupes become symlinks)
+notepad-cleanup links separate --last      # 6. Optional: see only new files
+notepad-cleanup links join --last          # 7. Optional: rejoin everything
+```
+
+After setup, `--last` auto-uses the most recent extraction. No path copy-pasting.
+
+### Commands
+
+| Command | Purpose |
+|---------|---------|
+| `extract` | Extract text from all open Notepad windows/tabs |
+| `compare` | Find duplicates across historical sessions |
+| `organize` | AI-powered categorization (symlinks for duplicates, copies for new) |
+| `links` | Separate linked files from organized/, or rejoin them |
+| `diff` | Launch diff script to spot-check matched pairs |
+| `config` | Manage folders, search dirs, diff tool, settings |
+| `run` | Extract + organize in one step |
+
+For full parameter documentation, see [docs/parameters.md](docs/parameters.md).
+
+## Documentation
+
+| Doc | Contents |
+|-----|----------|
+| [Parameters](docs/parameters.md) | Full command reference with all options |
+| [Configuration](docs/config.md) | Folder registry, `...` notation, MRU, search dirs |
+| [Fuzzy Matching](docs/fuzzy-matching.md) | Threshold formula, derivation, customization |
+
+## Output Structure
+
+```
+notepad-cleanup/nc-2026-03-16__08-15-30/
+├── manifest.json                  # Extraction metadata
+├── window01/
+│   ├── tab01.txt                  # Raw extracted files
+│   ├── tab02.txt
+│   └── tab03.txt
+├── window02/
+│   └── tab01.txt
+├── organized/                     # AI-organized output (after organize step)
+│   ├── code-snippets/
+│   │   ├── process-data.py
+│   │   └── batch-rename.bat
+│   ├── personal-notes/
+│   │   └── grocery-list.txt
+│   └── _summary.md               # Organization summary
+├── _compare_results.json          # Dedup comparison cache
+├── _compare_diffs.cmd             # Diff script for spot-checking
+├── _dedup_links.json              # Link manifest (if --link used)
+├── _organize_prompt.md            # AI prompt used
+└── _organize_log.txt              # Claude CLI output
+```
+
+## How It Works
+
+### Phase 1: Silent Extraction
+
+Uses `WM_GETTEXT` message to read text from `RichEditD2DPT` child windows. This is completely silent and invisible -- no focus changes, no window activation, no disruption to your workflow.
+
+**Limitation:** Only works for tabs that have been loaded (visited) at least once in the current Notepad session. Unloaded tabs have no `RichEditD2DPT` control yet, so they cannot be read silently.
+
+### Phase 2: Tab Switching (Announced)
+
+For unloaded tabs, uses UI Automation (`TabItem.Select()`) to activate each tab, which forces Windows to load the `RichEditD2DPT` control. Once loaded, the same `WM_GETTEXT` method reads the content.
+
+**Warning:** This steals focus and activates Notepad windows. The tool warns you before Phase 2 starts and waits for confirmation. Do not type or click during Phase 2.
+
+### Organization with AI
+
+After extraction, Claude Code CLI:
+1. Reads `manifest.json` to understand the collection
+2. Reads each extracted file to determine content type
+3. Returns a JSON plan with categories and renamed filenames
+4. The tool executes the plan locally (copy files to organized folders)
+
+## Requirements
+
+- **Windows 11** (uses Windows 11 Notepad tab features)
+- **Python 3.10+**
+- **Claude Code CLI** (optional, for organize step)
+
+For detailed installation instructions, see [docs/install.md](docs/install.md).
+
+## Development
+
+```bash
+git clone https://github.com/DazzleTools/notepad-cleanup.git
+cd notepad-cleanup
+python -m venv venv
+venv\Scripts\activate
+pip install -e .
+```
+
+## Contributing
+
+Contributions are welcome! See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
+
+Like the project?
+
+[!["Buy Me A Coffee"](https://www.buymeacoffee.com/assets/img/custom_images/orange_img.png)](https://www.buymeacoffee.com/djdarcy)
+
+## License
+
+notepad-cleanup, Copyright (C) 2026 Dustin Darcy. 
+
+This project is licensed under the GNU General Public License v3.0 — see [LICENSE](LICENSE) for full details.

{notepad_cleanup-0.2.1 → notepad_cleanup-0.2.3}/notepad_cleanup/_version.py

@@ -21,13 +21,13 @@ Version levels:
 # Version components - edit these for version bumps
 MAJOR = 0
 MINOR = 2
-PATCH = 1
+PATCH = 3
 PHASE = None  # Per-MINOR feature set: None, "alpha", "beta", "rc1", etc.
 PRE_RELEASE_NUM = 0  # PEP 440 pre-release number (e.g., a1, b2)
 PROJECT_PHASE = "prealpha"  # Project-wide: "prealpha", "alpha", "beta", "stable"
 
 # Auto-updated by git hooks - do not edit manually
-__version__ = "0.2.1_main_14-20260316-16a4e3a7"
+__version__ = "0.2.3_main_16-20260317-af4f15c0"
 __app_name__ = "notepad-cleanup"
 
 

{notepad_cleanup-0.2.1 → notepad_cleanup-0.2.3}/notepad_cleanup/cli.py

@@ -13,13 +13,15 @@ from .discovery import find_notepad_windows, get_richedit_children, get_tab_coun
 from .extractor import extract_phase1, extract_phase2, merge_results
 from .saver import save_extraction
 from .organizer import (generate_prompt, invoke_claude_cli, save_prompt_to_file,
-                         find_claude_cli, parse_plan, execute_plan)
+                         find_claude_cli, parse_plan, execute_plan,
+                         separate_links, join_links)
 from .dedup import (find_session_dirs, build_hash_index, find_duplicates,
                     find_content_files,
                     generate_unified_diff, near_match_threshold,
                     resolve_diff_tool, launch_diff_tool, create_links,
                     write_link_manifest, generate_diff_script,
                     save_compare_results, load_compare_results,
+                    get_linked_paths,
                     LINK_STRATEGIES, CACHE_FILENAME)
 from .config import (
     load_config, save_config, config_get, config_set, config_unset,
@@ -546,8 +548,14 @@ def organize(folder, last, backend, dry_run, verbose):
 
     console.print("\n[bold]Notepad Cleanup -- Organize[/bold]\n")
 
-    # Generate prompt
-    prompt = generate_prompt(manifest_path)
+    # Load dedup context if available (for link-aware organize)
+    linked_paths = get_linked_paths(folder)
+    if linked_paths:
+        console.print(f"  [dim]Found {len(linked_paths)} dedup-linked files "
+                      f"(will symlink, not copy)[/dim]")
+
+    # Generate prompt (linked files excluded from AI task)
+    prompt = generate_prompt(manifest_path, linked_paths=linked_paths)
 
     if backend == "prompt-only" or dry_run:
         prompt_file = save_prompt_to_file(prompt, folder)
@@ -606,12 +614,14 @@ def organize(folder, last, backend, dry_run, verbose):
 
     console.print(f"  Claude proposed a plan for [bold]{len(plan)}[/bold] files\n")
 
-    # Step 3: Execute the plan locally
+    # Step 3: Execute the plan locally (link-aware)
     with console.status("Organizing files..."):
-        summary, stats = execute_plan(plan, folder)
+        summary, stats = execute_plan(plan, folder, linked_paths=linked_paths)
 
     console.print("[bold green]Organization complete![/bold green]\n")
-    console.print(f"  Files organized:  {stats['copied']}")
+    console.print(f"  Files copied:     {stats['copied']}")
+    if stats.get('linked', 0):
+        console.print(f"  Files linked:     {stats['linked']} (symlink/hardlink to canonical)")
     if stats['errors']:
         console.print(f"  [red]Errors:         {stats['errors']}[/red]")
 
@@ -1011,6 +1021,89 @@ def diff_cmd(folder, last):
         console.print(f"  You can run it manually: {script_path}\n")
 
 
+@main.command("links")
+@click.argument("action", type=click.Choice(["separate", "join"]))
+@click.argument("folder", type=click.Path(), required=False, default=None)
+@click.option("--last", is_flag=True, help="Use the most recent extraction")
+@click.option("--dir-name", default="organized-links",
+              help="Name for the links directory (default: organized-links)")
+@click.option("--dry-run", is_flag=True, help="Preview without moving files")
+def links_cmd(action, folder, last, dir_name, dry_run):
+    """Separate or rejoin linked files in organized/.
+
+    \b
+    separate: Move linked files from organized/ into a parallel tree,
+              preserving category structure. Shows only new files in organized/.
+    join:     Move linked files back from the parallel tree into organized/,
+              restoring the full collection.
+
+    \b
+    Examples:
+      notepad-cleanup links separate --last              Split links out
+      notepad-cleanup links separate --last --dry-run    Preview the split
+      notepad-cleanup links join --last                  Rejoin links
+      notepad-cleanup links join --last --dry-run        Preview the rejoin
+    """
+
+    folder = resolve_folder(folder, use_last=last)
+    if folder is None:
+        console.print("[red]No extraction folder specified.[/red]")
+        console.print("  Provide a FOLDER argument or use --last.")
+        return
+    folder = Path(folder)
+    if not folder.exists():
+        console.print(f"[red]Folder does not exist: {folder}[/red]")
+        return
+
+    if last:
+        console.print(f"\n  [dim]Using last extraction: {folder}[/dim]")
+
+    organized_dir = folder / "organized"
+    links_dir = folder / dir_name
+
+    if action == "separate":
+        if not organized_dir.exists():
+            console.print(f"\n  [yellow]No organized/ folder found in {folder}[/yellow]")
+            console.print("  Run 'notepad-cleanup organize' first.\n")
+            return
+
+        label = "Separate Links -- Dry Run" if dry_run else "Separate Links"
+        console.print(f"\n[bold]{label}[/bold]\n")
+        console.print(f"  From: {organized_dir}")
+        console.print(f"  To:   {links_dir}\n")
+
+        stats, details = separate_links(organized_dir, links_dir_name=dir_name, dry_run=dry_run)
+
+        console.print(f"  Links moved:    {stats['moved']}")
+        console.print(f"  Real files:     {stats['real_kept']}")
+        if stats['errors']:
+            console.print(f"  [red]Errors:       {stats['errors']}[/red]")
+
+    elif action == "join":
+        if not links_dir.exists():
+            console.print(f"\n  [yellow]No {dir_name}/ folder found in {folder}[/yellow]")
+            console.print("  Run 'notepad-cleanup links separate' first.\n")
+            return
+
+        label = "Join Links -- Dry Run" if dry_run else "Join Links"
+        console.print(f"\n[bold]{label}[/bold]\n")
+        console.print(f"  From: {links_dir}")
+        console.print(f"  To:   {organized_dir}\n")
+
+        stats, details = join_links(organized_dir, links_dir_name=dir_name, dry_run=dry_run)
+
+        console.print(f"  Links restored: {stats['moved']}")
+        if stats['errors']:
+            console.print(f"  [red]Errors:       {stats['errors']}[/red]")
+
+    if details:
+        console.print()
+        for d in details:
+            console.print(d)
+
+    console.print()
+
+
 @main.command()
 @click.option("--output-dir", "-o", type=click.Path(), default=None,
               help="Output directory")

{notepad_cleanup-0.2.1 → notepad_cleanup-0.2.3}/notepad_cleanup/dedup.py

@@ -1062,24 +1062,59 @@ def _create_single_link(match: DedupMatch, strategy: str, backup: bool) -> LinkR
 def _create_dazzlelink_file(link_path: Path, target_path: Path):
     """Create a .dazzlelink JSON descriptor file.
 
-    This is a lightweight cross-platform alternative when native symlinks
-    aren't available. The file is a JSON descriptor that tools can follow.
+    This is a cross-platform alternative when native symlinks aren't
+    available. The file is a JSON descriptor compatible with the DazzleLink
+    tool (https://github.com/DazzleTools/dazzlelink). Default mode is "open"
+    so double-clicking opens the target in its native application.
     """
     from datetime import datetime
+    import time
+
+    resolved_target = target_path.resolve()
+    now = datetime.now()
+    now_ts = time.time()
+
+    # Get target file info
+    target_exists = resolved_target.exists()
+    target_size = 0
+    target_ext = resolved_target.suffix
+    target_timestamps = {}
+    if target_exists:
+        try:
+            stat = resolved_target.stat()
+            target_size = stat.st_size
+            target_timestamps = {
+                "created": stat.st_ctime,
+                "modified": stat.st_mtime,
+                "accessed": stat.st_atime,
+                "created_iso": datetime.fromtimestamp(stat.st_ctime).isoformat(),
+                "modified_iso": datetime.fromtimestamp(stat.st_mtime).isoformat(),
+                "accessed_iso": datetime.fromtimestamp(stat.st_atime).isoformat(),
+            }
+        except OSError:
+            pass
 
     dazzlelink_data = {
         "schema_version": 1,
-        "created_by": "notepad-cleanup dedup",
-        "creation_date": datetime.now().isoformat(),
+        "created_by": "notepad-cleanup",
+        "creation_timestamp": now_ts,
+        "creation_date": now.isoformat(),
         "link": {
             "original_path": str(link_path),
-            "target_path": str(target_path.resolve()),
+            "target_path": str(resolved_target),
             "type": "dazzlelink",
             "relative_path": False,
         },
         "target": {
-            "exists": target_path.exists(),
-            "size": target_path.stat().st_size if target_path.exists() else 0,
+            "exists": target_exists,
+            "type": "file",
+            "size": target_size,
+            "extension": target_ext,
+            "timestamps": target_timestamps,
+        },
+        "config": {
+            "default_mode": "open",
+            "platform": _os.name,
         },
         "context": {
             "reason": "dedup_exact_match",
@@ -1127,6 +1162,45 @@ def write_link_manifest(results: list, output_dir: Path):
     return manifest_path
 
 
+LINK_MANIFEST_FILENAME = "_dedup_links.json"
+
+
+def load_link_manifest(session_dir: Path) -> dict:
+    """Load the dedup link manifest from a session directory.
+
+    Returns the parsed manifest dict, or an empty structure if no manifest exists.
+    The returned dict always has a "links" key (list of link entries).
+    """
+    manifest_path = Path(session_dir) / LINK_MANIFEST_FILENAME
+    if not manifest_path.exists():
+        return {"links": []}
+    try:
+        data = json.loads(manifest_path.read_text(encoding="utf-8"))
+        if "links" not in data:
+            data["links"] = []
+        return data
+    except (json.JSONDecodeError, OSError):
+        return {"links": []}
+
+
+def get_linked_paths(session_dir: Path) -> dict:
+    """Return a mapping of linked new_path -> canonical_path for successful links.
+
+    Keys are resolved Path objects (the window*/tab*.txt files that were linked).
+    Values are resolved Path objects pointing to the canonical (provenance root) file.
+    Returns empty dict if no manifest or no successful links.
+    """
+    manifest = load_link_manifest(session_dir)
+    linked = {}
+    for entry in manifest.get("links", []):
+        if not entry.get("success"):
+            continue
+        new_path = Path(entry["new_path"]).resolve()
+        canonical_path = Path(entry["canonical_path"]).resolve()
+        linked[new_path] = canonical_path
+    return linked
+
+
 # --- Diff tools ---
 
 # Known diff tools in preference order (name, command pattern).