photo-stack-finder 0.1.7__tar.gz → 0.1.8__tar.gz

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/PKG-INFO

@@ -1,10 +1,10 @@
 Metadata-Version: 2.4
 Name: photo-stack-finder
-Version: 0.1.7
+Version: 0.1.8
 Summary: Photo deduplication using perceptual hashing and sequence detection
 Author: Geoff Barrett
 Maintainer: Geoff Barrett
-License: AGPL-3.0-or-later
+License-Expression: AGPL-3.0-or-later
 Project-URL: Homepage, https://github.com/gbarrett28/photo_dedup
 Project-URL: Repository, https://github.com/gbarrett28/photo_dedup
 Project-URL: Issues, https://github.com/gbarrett28/photo_dedup/issues
@@ -12,7 +12,6 @@ Project-URL: Discussions, https://github.com/gbarrett28/photo_dedup/discussions
 Keywords: photo,deduplication,perceptual-hashing,image-processing
 Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: End Users/Desktop
-Classifier: License :: OSI Approved :: GNU Affero General Public License v3 or later (AGPLv3+)
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/pyproject.toml

@@ -7,11 +7,11 @@ orchestrator = ["static/*", "py.typed"]
 
 [project]
 name = "photo-stack-finder"
-version = "0.1.7"
+version = "0.1.8"
 description = "Photo deduplication using perceptual hashing and sequence detection"
 readme = "README.md"
 requires-python = ">=3.11"
-license = {text = "AGPL-3.0-or-later"}
+license = "AGPL-3.0-or-later"
 authors = [
     {name = "Geoff Barrett"}
 ]
@@ -22,7 +22,6 @@ keywords = ["photo", "deduplication", "perceptual-hashing", "image-processing"]
 classifiers = [
     "Development Status :: 4 - Beta",
     "Intended Audience :: End Users/Desktop",
-    "License :: OSI Approved :: GNU Affero General Public License v3 or later (AGPLv3+)",
     "Programming Language :: Python :: 3",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/orchestrator/__init__.py

@@ -1,7 +1,7 @@
-"""Orchestration layer for photo deduplication workflow.
+"""Orchestration layer for photo stack finding workflow.
 
 This module provides the web-based interface and pipeline orchestration for the
-photo deduplication system. It includes:
+photo stack finding system. It includes:
 
 - FastAPI web application (app)
 - Pipeline orchestration and execution (PipelineOrchestrator)

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/orchestrator/app.py

@@ -1,4 +1,4 @@
-"""FastAPI orchestration server for photo deduplication.
+"""FastAPI orchestration server for photo stack finding.
 
 This provides a web-based interface for configuring and running the pipeline.
 
@@ -209,7 +209,7 @@ async def lifespan_manager(app: FastAPI) -> AsyncIterator[None]:
 
 
 # Pass the new lifespan function to the FastAPI constructor
-app = FastAPI(title="Photo Dedup Orchestrator", lifespan=lifespan_manager)
+app = FastAPI(title="Photo Stack Finder Orchestrator", lifespan=lifespan_manager)
 
 
 @app.get("/")
@@ -876,9 +876,7 @@ async def websocket_progress(
 async def serve_review_identical() -> FileResponse:
     """Serve identical files review interface."""
     static_path = (
-        Path(CONFIG.orchestrator.STATIC_DIR)
-        if CONFIG.orchestrator.STATIC_DIR
-        else Path(__file__).parent / "static"
+        Path(CONFIG.orchestrator.STATIC_DIR) if CONFIG.orchestrator.STATIC_DIR else Path(__file__).parent / "static"
     )
     return FileResponse(static_path / "review_identical.html")
 
@@ -887,9 +885,7 @@ async def serve_review_identical() -> FileResponse:
 async def serve_review_sequences() -> FileResponse:
     """Serve sequences review interface."""
     static_path = (
-        Path(CONFIG.orchestrator.STATIC_DIR)
-        if CONFIG.orchestrator.STATIC_DIR
-        else Path(__file__).parent / "static"
+        Path(CONFIG.orchestrator.STATIC_DIR) if CONFIG.orchestrator.STATIC_DIR else Path(__file__).parent / "static"
     )
     return FileResponse(static_path / "review_sequences.html")
 
@@ -899,13 +895,12 @@ static_dir = Path(CONFIG.orchestrator.STATIC_DIR) if CONFIG.orchestrator.STATIC_
 if static_dir and static_dir.exists():
     app.mount("/static", StaticFiles(directory=str(static_dir)), name="static")
 
+
 # Serve review_common.js from static directory
 @app.get("/review_common.js")
 async def serve_review_common_js() -> FileResponse:
     """Serve review_common.js from static directory."""
     static_path = (
-        Path(CONFIG.orchestrator.STATIC_DIR)
-        if CONFIG.orchestrator.STATIC_DIR
-        else Path(__file__).parent / "static"
+        Path(CONFIG.orchestrator.STATIC_DIR) if CONFIG.orchestrator.STATIC_DIR else Path(__file__).parent / "static"
     )
     return FileResponse(static_path / "review_common.js", media_type="application/javascript")

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/orchestrator/build_pipeline.py

@@ -1,6 +1,6 @@
 """Pipeline construction using PipelineBuilder pattern.
 
-This module constructs the complete photo deduplication pipeline using the
+This module constructs the complete photo stack finding pipeline using the
 new port-based orchestration system.  All 8 stages are wired together via
 Channel connections based on their InputPort/OutputPort declarations.
 """
@@ -19,7 +19,6 @@ from utils import (
     ComputePerceptualHash,
     ComputePerceptualMatch,
     ComputeShaBins,
-    ComputeTemplates,
     ComputeTemplateSimilarity,
     ComputeVersions,
     InputPort,
@@ -31,18 +30,21 @@ from .pipeline_orchestrator import PipelineOrchestrator
 
 
 def build_pipeline(source_dir: Path) -> PipelineOrchestrator:
-    """Build the complete photo deduplication pipeline.
-
-    Constructs a pipeline graph with 8-9 stages connected via ports:
-    1. ComputeShaBins - Hash files and bin by SHA256
-    2. ComputeIdentical - Find byte-identical duplicates
-    3. ComputeTemplates - Bin photos by filename template
-    4. ComputeVersions - Detect version patterns in filenames
-    5. ComputeTemplateSimilarity - Match photos with similar templates
-    6. ComputeIndices - Find sequences with overlapping indices
-    7. ComputePerceptualHash - Compute perceptual hashes and bin
-    8. ComputePerceptualMatch - Match photos by perceptual hash similarity
-    9. ComputeBenchmarks - (Optional, controlled by CONFIG.benchmark.ENABLED)
+    """Build the complete photo stack finding pipeline.
+
+    Constructs a pipeline graph with 7-8 stages connected via ports:
+    1. ComputeShaBins - Hash files, bin by SHA256, extract templates
+    2. ComputeIdentical - Find byte-identical duplicates, output template bins
+    3. ComputeVersions - Detect version patterns in filenames
+    4. ComputeTemplateSimilarity - Match photos with similar templates
+    5. ComputeIndices - Find sequences with overlapping indices
+    6. ComputePerceptualHash - Compute perceptual hashes and bin
+    7. ComputePerceptualMatch - Match photos by perceptual hash similarity
+    8. ComputeBenchmarks - (Optional, controlled by CONFIG.benchmark.ENABLED)
+
+    Note: ComputeTemplates stage has been merged into ComputeIdentical.
+    Template extraction now happens during SHA binning in ComputeShaBins,
+    and template binning happens in ComputeIdentical's finalise() method.
 
     Args:
         source_dir: Root directory containing photos to process
@@ -60,17 +62,13 @@ def build_pipeline(source_dir: Path) -> PipelineOrchestrator:
         # SHA256 Hashing and Binning
         sha_bins_stage = ComputeShaBins(source_path=source_dir)
 
-        # Identical Files Detection
+        # Identical Files Detection (outputs template bins)
         identical_stage = ComputeIdentical()
         Channel(sha_bins_stage.sha_bins_o, identical_stage.sha_bins_i)
 
-        # Template Binning
-        templates_stage = ComputeTemplates()
-        Channel(identical_stage.nonidentical_o, templates_stage.nonidentical_photos_i)
-
-        # Version Detection
+        # Version Detection (receives template bins directly from ComputeIdentical)
         versions_stage = ComputeVersions()
-        Channel(templates_stage.template_bins_o, versions_stage.template_bins_i)
+        Channel(identical_stage.nonidentical_o, versions_stage.template_bins_i)
 
         # Template Similarity
         template_similarity_stage = ComputeTemplateSimilarity()

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/orchestrator/orchestrator_runner.py

@@ -68,25 +68,25 @@ def get_os_subdir() -> str:
 
     Returns:
         OS-specific subdirectory name:
-        - 'window' for Windows
+        - 'windows' for Windows
         - 'linux' for Linux
         - 'darwin' for macOS
         - Platform name for others
 
     Example:
-        work_dir = source_parent / "photo_dedup" / get_os_subdir()
-        # Windows: .../photo_dedup/window/
-        # Linux:   .../photo_dedup/linux/
-        # macOS:   .../photo_dedup/darwin/
+        work_dir = source_parent / "photo_stack_finder" / get_os_subdir()
+        # Windows: .../photo_stack_finder/windows/
+        # Linux:   .../photo_stack_finder/linux/
+        # macOS:   .../photo_stack_finder/darwin/
     """
     platform = sys.platform
     if platform.startswith("win"):
-        return "window"
+        return "windows"
     if platform.startswith("linux"):
         return "linux"
     if platform.startswith("darwin"):
         return "darwin"
-    return platform  # Fallback for other platforms
+    return platform  # Fallback for other platforms  # Fallback for other platforms
 
 
 @dataclass
@@ -468,7 +468,7 @@ class OrchestratorRunner:
             # Default work_dir if not provided: OS-specific subdirectory
             source_path = Path(CONFIG.paths.SOURCE_DIR)
             os_subdir = get_os_subdir()
-            CONFIG.paths.WORK_DIR = str(source_path.parent / "photo_dedup" / os_subdir)
+            CONFIG.paths.WORK_DIR = str(source_path.parent / "photo_stack_finder" / os_subdir)
 
         # Create work directory if it doesn't exist
         CONFIG.paths.work_dir.mkdir(parents=True, exist_ok=True)
@@ -483,6 +483,9 @@ class OrchestratorRunner:
         if "debug_mode" in config:
             CONFIG.processing.DEBUG_MODE = config["debug_mode"]
 
+        if "skip_byte_identical" in config:
+            CONFIG.processing.SKIP_BYTE_IDENTICAL = config["skip_byte_identical"]
+
         # Update gate thresholds if provided
         if config.get("gate_thresholds"):
             assert CONFIG.processing.GATE_THRESHOLDS is not None

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/orchestrator/static/orchestrator.html

@@ -120,6 +120,12 @@
                                         Debug Mode (sequential processing)
                                     </label>
                                 </div>
+                                <div class="form-group">
+                                    <label for="skip-byte-identical">
+                                        <input type="checkbox" id="skip-byte-identical" name="skip_byte_identical" checked>
+                                        Skip byte-identical detection (trust SHA256 uniqueness)
+                                    </label>
+                                </div>
                             </div>
 
                             <div class="option-group">

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/orchestrator/static/orchestrator.js

@@ -313,6 +313,7 @@ function populateForm(config) {
     document.getElementById('max-workers').value = config.max_workers || '';
     document.getElementById('batch-size').value = config.batch_size || '';
     document.getElementById('debug-mode').checked = config.debug_mode || false;
+    document.getElementById('skip-byte-identical').checked = config.skip_byte_identical !== false; // Default true
 
     document.getElementById('comparison-method').value = config.comparison_method || 'SSIM';
     document.getElementById('ssim-threshold').value = config.gate_thresholds?.SSIM || 0.95;
@@ -369,6 +370,7 @@ async function handleFormSubmit(event) {
         max_workers: maxWorkers ? parseInt(maxWorkers) : null,
         batch_size: batchSize ? parseInt(batchSize) : null,
         debug_mode: formData.get('debug_mode') === 'on',
+        skip_byte_identical: formData.get('skip_byte_identical') === 'on',
         comparison_method: formData.get('comparison_method'),
         gate_thresholds: {
             SSIM: parseFloat(formData.get('ssim_threshold')),

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/overlap_metrics/__init__.py

@@ -40,7 +40,7 @@ if version.parse(scipy.__version__) < version.parse(MINIMUM_SCIPY_VERSION):
 
 # Version info
 __version__ = "1.0.0"
-__author__ = "Photo Deduplication Team"
+__author__ = "Photo Stack Finder Team"
 __description__ = "Distribution separation and overlap metrics with pluggable estimators"
 
 # Default metric suite for compute_suite()

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/photo_compare/__init__.py

@@ -43,7 +43,7 @@ from .structural_methods import HOGMethod, MultiScaleSSIMMethod, SSIMMethod
 
 # Version information
 __version__ = "1.0.0"
-__author__ = "Photo Deduplication Team"
+__author__ = "Photo Stack Finder Team"
 __description__ = "Image similarity methods with integrated caching and factory pattern"
 
 

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/photo_stack_finder.egg-info/PKG-INFO

@@ -1,10 +1,10 @@
 Metadata-Version: 2.4
 Name: photo-stack-finder
-Version: 0.1.7
+Version: 0.1.8
 Summary: Photo deduplication using perceptual hashing and sequence detection
 Author: Geoff Barrett
 Maintainer: Geoff Barrett
-License: AGPL-3.0-or-later
+License-Expression: AGPL-3.0-or-later
 Project-URL: Homepage, https://github.com/gbarrett28/photo_dedup
 Project-URL: Repository, https://github.com/gbarrett28/photo_dedup
 Project-URL: Issues, https://github.com/gbarrett28/photo_dedup/issues
@@ -12,7 +12,6 @@ Project-URL: Discussions, https://github.com/gbarrett28/photo_dedup/discussions
 Keywords: photo,deduplication,perceptual-hashing,image-processing
 Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: End Users/Desktop
-Classifier: License :: OSI Approved :: GNU Affero General Public License v3 or later (AGPLv3+)
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/photo_stack_finder.egg-info/SOURCES.txt

@@ -55,7 +55,6 @@ src/utils/compute_perceptual_hash.py
 src/utils/compute_perceptual_match.py
 src/utils/compute_sha_bins.py
 src/utils/compute_template_similarity.py
-src/utils/compute_templates.py
 src/utils/compute_versions.py
 src/utils/config.py
 src/utils/data_io.py
@@ -75,6 +74,7 @@ src/utils/review_utils.py
 src/utils/sequence.py
 src/utils/sequence_clustering.py
 src/utils/template.py
+src/utils/template_parsing.py
 tests/test_app_endpoints_comprehensive.py
 tests/test_comparison_methods_comprehensive.py
 tests/test_compute_benchmarks_comprehensive.py
@@ -84,7 +84,6 @@ tests/test_compute_indices_comprehensive.py
 tests/test_compute_perceptual_hash_comprehensive.py
 tests/test_compute_perceptual_match_comprehensive.py
 tests/test_compute_sha_bins_comprehensive.py
-tests/test_compute_templates_comprehensive.py
 tests/test_compute_versions_comprehensive.py
 tests/test_orchestrator_comprehensive.py
 tests/test_photo_file_pixel_array.py

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/scripts/orchestrate.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
-"""Photo Deduplication Orchestrator - Web Interface Entry Point.
+"""Photo Stack Finderlication Orchestrator - Web Interface Entry Point.
 
-This provides a web-based interface for the photo deduplication pipeline.
+This provides a web-based interface for the photo stack finding pipeline.
 
 Usage:
     python orchestrate.py
@@ -38,10 +38,10 @@ def build_arg_parser() -> argparse.ArgumentParser:
         Configured ArgumentParser instance
     """
     parser = argparse.ArgumentParser(
-        description="Photo Dedup Orchestrator - Web Interface",
+        description="Photo Stack Finder Orchestrator - Web Interface",
         formatter_class=argparse.RawDescriptionHelpFormatter,
         epilog="""
-This is a web interface for the photo deduplication pipeline.
+This is a web interface for the photo stack finding pipeline.
 
 For the web interface, just run:
   - python orchestrate.py
@@ -277,11 +277,13 @@ def main() -> None:
 
     # Add platform-specific instructions
     if is_linux:
-        lines.extend([
-            "To access the web interface:",
-            f"  Open your browser to: {url}",
-            "",
-        ])
+        lines.extend(
+            [
+                "To access the web interface:",
+                f"  Open your browser to: {url}",
+                "",
+            ]
+        )
 
     lines.append("Press Ctrl+C to stop the server")
     print_banner("", lines)

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/utils/__init__.py

@@ -1,4 +1,4 @@
-"""Utilities for photo deduplication."""
+"""Utilities for photo stack finding."""
 
 from __future__ import annotations
 
@@ -26,7 +26,6 @@ from .compute_perceptual_hash import ComputePerceptualHash
 from .compute_perceptual_match import ComputePerceptualMatch
 from .compute_sha_bins import ComputeShaBins
 from .compute_template_similarity import ComputeTemplateSimilarity
-from .compute_templates import ComputeTemplates
 from .compute_versions import ComputeVersions
 
 # Config module exports
@@ -109,6 +108,9 @@ from .template import (
     partial_format,
 )
 
+# Template parsing module exports (extract_template not exported - internal use only)
+# INDEX_T exported via sequence module
+
 __all__ = [
     # Config
     "CONFIG",
@@ -136,7 +138,6 @@ __all__ = [
     "ComputePerceptualMatch",
     "ComputeShaBins",
     "ComputeTemplateSimilarity",
-    "ComputeTemplates",
     "ComputeVersions",
     # Configuration
     "Config",

{photo_stack_finder-0.1.7 → photo_stack_finder-0.1.8}/src/utils/comparison_gates.py

@@ -291,8 +291,8 @@ class AspectRatioGate(BaseGate):
         """
         # Get aspect ratios (may trigger lazy dimension extraction)
         # This may load pixels if dimensions not yet cached
-        ratio1 = img1.get_aspect_ratio()
-        ratio2 = img2.get_aspect_ratio()
+        ratio1 = img1._photo.aspect_ratio
+        ratio2 = img2._photo.aspect_ratio
 
         # Normalize to portrait orientation (AR < 1.0)
         # This allows matching landscape ↔ portrait photos
@@ -372,8 +372,14 @@ class GateSequence:
                     # MethodGate: check if pixels needed (cache miss detection)
                     # Only load pixels if we'll actually need them
                     # Note: cache keys are (method_name, rotation) tuples, always tuples even for rotation=0
-                    will_need_pixels1 = (photo1.id, 0) not in gate.cache and (gate._name, 0) not in photo1.cache
-                    will_need_pixels2 = (photo2.id, 0) not in gate.cache and (gate._name, 0) not in photo2.cache
+                    will_need_pixels1 = (photo1.id, 0) not in gate.cache and (
+                        gate._name,
+                        0,
+                    ) not in photo1.cache
+                    will_need_pixels2 = (photo2.id, 0) not in gate.cache and (
+                        gate._name,
+                        0,
+                    ) not in photo2.cache
 
                     if will_need_pixels1 or will_need_pixels2:
                         # Cache miss detected - load pixels if not already loaded
@@ -430,29 +436,24 @@ class GateSequence:
         # Flexible ImageData management: create contexts only for photos that don't have pre-created ImageData
         # Case 1: Both provided - use directly
         if ref_img is not None and cand_img is not None:
-            return self._compare_with_rotation_impl(
-                reference, candidate, ref_img, cand_img, short_circuit
-            )
+            return self._compare_with_rotation_impl(reference, candidate, ref_img, cand_img, short_circuit)
 
         # Case 2: Only ref provided - create context for candidate
         if ref_img is not None and cand_img is None:
             with candidate.image_data() as cand_img_ctx:
-                return self._compare_with_rotation_impl(
-                    reference, candidate, ref_img, cand_img_ctx, short_circuit
-                )
+                return self._compare_with_rotation_impl(reference, candidate, ref_img, cand_img_ctx, short_circuit)
 
         # Case 3: Only cand provided - create context for reference
         if ref_img is None and cand_img is not None:
             with reference.image_data() as ref_img_ctx:
-                return self._compare_with_rotation_impl(
-                    reference, candidate, ref_img_ctx, cand_img, short_circuit
-                )
+                return self._compare_with_rotation_impl(reference, candidate, ref_img_ctx, cand_img, short_circuit)
 
         # Case 4: Neither provided - create contexts for both (original pattern)
-        with reference.image_data() as ref_img_new, candidate.image_data() as cand_img_new:
-            return self._compare_with_rotation_impl(
-                reference, candidate, ref_img_new, cand_img_new, short_circuit
-            )
+        with (
+            reference.image_data() as ref_img_new,
+            candidate.image_data() as cand_img_new,
+        ):
+            return self._compare_with_rotation_impl(reference, candidate, ref_img_new, cand_img_new, short_circuit)
 
     def _compare_with_rotation_impl(
         self,
@@ -508,7 +509,13 @@ class GateSequence:
         # Second attempt: try 180° rotation ONLY if first failed
         if not overall_pass:
             ref_rotation_180 = ref_norm_rotation + 180
-            scores_180, overall_pass_180, final_gate_score_180, ref_pixels, cand_pixels = self._attempt_comparison(
+            (
+                scores_180,
+                overall_pass_180,
+                final_gate_score_180,
+                ref_pixels,
+                cand_pixels,
+            ) = self._attempt_comparison(
                 reference,
                 candidate,
                 ref_img,
@@ -539,7 +546,13 @@ class GateSequence:
         ref_pixels: npt.NDArray[np.uint8] | None,
         cand_pixels: npt.NDArray[np.uint8] | None,
         short_circuit: bool,
-    ) -> tuple[dict[str, float], bool, float, npt.NDArray[np.uint8] | None, npt.NDArray[np.uint8] | None]:
+    ) -> tuple[
+        dict[str, float],
+        bool,
+        float,
+        npt.NDArray[np.uint8] | None,
+        npt.NDArray[np.uint8] | None,
+    ]:
         """Attempt comparison at specific rotation angles.
 
         Args:
@@ -569,11 +582,17 @@ class GateSequence:
             elif isinstance(gate, MethodGate):
                 # Check if we need to load pixels (cache miss detection)
                 # Cache keys are always (method_name, rotation) tuples
-                will_need_ref_pixels = (reference.id, ref_rotation) not in gate.cache and (
+                will_need_ref_pixels = (
+                    reference.id,
+                    ref_rotation,
+                ) not in gate.cache and (
                     gate._name,
                     ref_rotation,
                 ) not in reference.cache
-                will_need_cand_pixels = (candidate.id, cand_rotation) not in gate.cache and (
+                will_need_cand_pixels = (
+                    candidate.id,
+                    cand_rotation,
+                ) not in gate.cache and (
                     gate._name,
                     cand_rotation,
                 ) not in candidate.cache