docs: v1.5.2 changelog, readme, architecture sanity pass

- CHANGELOG: add [1.5.2] section (parse_cache, ContextExtractor, Backend Plugin)
- README: v1.5.2 badge + What's New v1.5.2 section
- Architecture.md: v1.5.2 header, engine/ file list updated, 443 tests, Backend Plugin dispatch
- Document_Parsing.md: version header v1.5.0 -> v1.5.2
- pyproject.toml + __init__.py: 1.5.1 -> 1.5.2

Author: Flamehaven CI

CHANGELOG.md

@@ -7,6 +7,49 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ---
 
+## [1.5.2] - 2026-04-19
+
+### Added
+
+- **Parse Cache** (`engine/parse_cache.py`): mtime-based file extraction cache.
+  Algorithm absorbed from RAG-Anything `processor.py:_generate_cache_key()`.
+  Cache key = MD5(resolved_path + mtime + parser_config). Path-indexed reverse
+  map enables O(1) `invalidate()`. API: `get/put/invalidate/clear/stats`.
+  `extract_text(use_cache=True)` integrates transparently — no API change.
+  Score: **0.0 CLEAN**.
+
+- **ContextExtractor** (`engine/context_extractor.py`): Sliding-window chunk
+  context extractor for RAG result enrichment. Algorithm absorbed from
+  RAG-Anything `modalprocessors.py:ContextExtractor`. Given `chunk_text()`
+  output, `enrich_chunks()` adds a `context` key to each chunk containing
+  surrounding neighbour text. `ContextConfig`: `window_size`, `max_context_chars`,
+  `include_headings`. Zero external dependencies. Score: **0.0 CLEAN**.
+
+- **Backend Plugin Architecture** (`engine/format_backends.py`): Format-family
+  backends absorbed from Docling `abstract_backend.py` pattern.
+  `AbstractFormatBackend` ABC with `supported_extensions` + `extract()`.
+  `BackendRegistry` maps extensions to backend classes; new formats register
+  without modifying the dispatcher. 11 concrete backends:
+  `PDFBackend`, `DOCXBackend`, `DOCBackend`, `XLSXBackend`, `PPTXBackend`,
+  `RTFBackend`, `HTMLBackend`, `VTTBackend`, `LaTeXBackend`, `CSVBackend`,
+  `ImageBackend`, `PlainTextBackend`. Score: **12.2 clean**.
+
+### Refactored
+
+- **`engine/file_parser.py`** (75 lines, was 340): Rewritten as pure registry
+  dispatcher — `_dispatch()` resolves backend via `BackendRegistry.get(ext)`
+  then calls `backend.extract()`. Cyclomatic complexity 13 → 3.
+  `function_clone_cluster` (5 structurally similar `_extract_*` functions)
+  eliminated by moving each into its own Backend class. Score: **3.0 CLEAN**.
+
+### Tests
+
+- `tests/test_phase1_parse_cache_context.py`: 33 tests (parse_cache + ContextExtractor).
+- `tests/test_phase2_format_backends.py`: 50 tests (registry + backends + helpers).
+- Combined: **83 tests**, all passing. AI-Slop-Detector critical deficits: 0.
+
+---
+
 ## [1.5.1] - 2026-04-18
 
 ### Removed

README.md

@@ -7,7 +7,7 @@
 ### Self-hosted RAG search engine. Production-ready in 3 minutes.
 
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
-[![Version](https://img.shields.io/badge/version-1.5.1-blue.svg)](CHANGELOG.md)
+[![Version](https://img.shields.io/badge/version-1.5.2-blue.svg)](CHANGELOG.md)
 [![Python](https://img.shields.io/badge/python-3.8+-blue.svg)](https://www.python.org/)
 [![Docker](https://img.shields.io/badge/docker-ready-brightgreen.svg)](https://hub.docker.com/r/flamehaven/filesearch)
 
@@ -60,6 +60,14 @@ Open source & MIT licensed</p>
 - **⚡ Ultra-Fast Vectors** - DSP v2.0 algorithm generates embeddings in <1ms without ML frameworks
 - **🎯 Source Attribution** - Every answer includes links back to source documents
 
+### What's New in v1.5.2 (patch)
+
+- **Parse Cache** (`engine/parse_cache.py`) — mtime-based file parse cache; `extract_text(use_cache=True)` skips re-parsing unchanged files
+- **ContextExtractor** (`engine/context_extractor.py`) — sliding-window chunk context enrichment for RAG pipelines (`enrich_chunks()`)
+- **Backend Plugin Architecture** (`engine/format_backends.py`) — 11 `AbstractFormatBackend` subclasses + `BackendRegistry`; new formats plug in without touching the dispatcher
+- **`file_parser.py`** refactored to 75 lines (was 340); cyclomatic complexity 13 → 3
+- 83 new tests; AI-Slop-Detector critical deficits: 0
+
 ### What's New in v1.5.1 (patch)
 
 - **Dead code removed** — `embedding_generator_legacy.py` deleted (306-line duplicate, unused)
@@ -485,7 +493,7 @@ Use the links below to jump to the most relevant guide.
 | [Document Parsing](docs/wiki/Document_Parsing.md) | Supported formats, internal parsers, RAG chunking |
 | [Framework Integrations](docs/wiki/Framework_Integrations.md) | LangChain, LlamaIndex, Haystack, CrewAI adapters |
 | [API Reference](docs/wiki/API_Reference.md) | REST endpoints, payloads, rate limits |
-| [Architecture](docs/wiki/Architecture.md) | How all layers fit together (v1.5.1) |
+| [Architecture](docs/wiki/Architecture.md) | How all layers fit together (v1.5.2) |
 | [Configuration Reference](docs/wiki/Configuration.md) | Full list of environment variables and config fields |
 | [Production Deployment](docs/wiki/Production_Deployment.md) | Docker, systemd, reverse proxy, scaling tips |
 | [Troubleshooting](docs/wiki/Troubleshooting.md) | Step-by-step debugging playbook |

docs/wiki/Architecture.md

@@ -1,14 +1,15 @@
 # Architecture Overview
 
 Flamehaven FileSearch balances simplicity with production-grade safeguards. This
-document describes the moving parts as of **v1.5.1**, featuring:
+document describes the moving parts as of **v1.5.2**, featuring:
 - **Gravitas DSP Engine** (v1.3.1+)
 - **Multimodal Search** (v1.4.0+)
 - **pgvector with HNSW** (v1.4.0+)
 - **Circuit Breaker & Health Checks** (v1.4.1+)
 - **ABC base classes, ruff CI, Windows filename fix** (v1.4.2)
 - **Universal Document Parser, Internal Chunker, Framework Integrations** (v1.5.0)
 - **Dead code removal, critical complexity fixes, 360-test suite** (v1.5.1)
+- **Parse Cache, ContextExtractor, Backend Plugin Architecture** (v1.5.2)
 
 ---
 
@@ -85,7 +86,7 @@ The new **Chronos-Grid** integration handles high-speed vector storage and simil
 
 ## 6. Testing & Quality (v1.4.2)
 
-- **Test Framework**: `pytest` — 360 tests collected, all passing.
+- **Test Framework**: `pytest` — 443 tests collected, all passing (360 + 83 new).
 - **Lint**: `black` (format) + `ruff` (lint/unused imports) — both enforced in CI.
 - **Validation**: `validators.py` enforces security policies (Filename 200-char max, FileSize, SearchQuery XSS/SQLi checks).
 - **SIDRCE Certification**: Omega 0.9894 (S++) — AI-Slop-Detector P0-P5 clean.
@@ -118,19 +119,32 @@ The `engine/` sub-package contains the full parsing stack:
 
 ```
 engine/
-  file_parser.py      — Dispatcher: routes by extension, tries parsers in priority order
-  format_parsers.py   — Internal parsers: HTML, WebVTT, LaTeX, CSV, Image OCR
-  text_chunker.py     — Structure-aware + token-aware RAG chunker (stdlib only)
-  embedding_generator.py  — DSP v2.0 vectorizer
-  chronos_grid.py     — Vector index + metadata store
-  gravitas_pack.py    — Metadata compression
-  intent_refiner.py   — Query analysis + search mode selection
+  file_parser.py       — Dispatcher: BackendRegistry.get(ext) -> backend.extract()
+  format_backends.py   — 11 AbstractFormatBackend classes + BackendRegistry (v1.5.2)
+  format_parsers.py    — Internal parsers: HTML, WebVTT, LaTeX, CSV, Image OCR
+  parse_cache.py       — mtime-based parse result cache (v1.5.2)
+  context_extractor.py — RAG chunk context window extractor (v1.5.2)
+  text_chunker.py      — Structure-aware + token-aware RAG chunker (stdlib only)
+  embedding_generator.py   — DSP v2.0 vectorizer
+  chronos_grid.py      — Vector index + metadata store
+  gravitas_pack.py     — Metadata compression
+  intent_refiner.py    — Query analysis + search mode selection
 ```
 
-**Extraction priority** (for each file):
-1. Per-format internal parser (HTML/VTT/LaTeX/CSV via `format_parsers.py`)
-2. Optional heavy parser (pymupdf, python-docx, openpyxl, python-pptx, striprtf)
-3. Plain UTF-8 read (last resort for unknown text formats)
+**Extraction dispatch** (v1.5.2 — Backend Plugin):
+
+Each file extension resolves to an `AbstractFormatBackend` subclass via
+`BackendRegistry`. New formats plug in by subclassing and registering — no
+changes to `file_parser.py` required.
+
+**Extraction stack per backend:**
+- PDF: `PDFBackend` — pymupdf → pypdf fallback
+- DOCX/DOC: `DOCXBackend` / `DOCBackend` — python-docx + antiword
+- XLSX: `XLSXBackend` — openpyxl multi-sheet
+- PPTX: `PPTXBackend` — python-pptx text + tables
+- HTML/VTT/LaTeX/CSV: stdlib-only backends (zero extra deps)
+- Images: `ImageBackend` — pytesseract ([vision] extra)
+- Unknown: `PlainTextBackend` — UTF-8 fallback
 
 **Content-based embedding** (v1.5.0): The first 2000 characters of extracted
 content are used to generate the vector embedding, replacing the previous

docs/wiki/Document_Parsing.md

@@ -1,4 +1,4 @@
-# Document Parsing (v1.5.0)
+# Document Parsing (v1.5.2)
 
 FLAMEHAVEN FileSearch extracts plain UTF-8 text from 34 file extensions using
 a layered stack of internal and optional parsers. No external document-AI

flamehaven_filesearch/__init__.py

@@ -6,7 +6,7 @@
 Fast, simple, and transparent file search for developers
 """
 
-__version__ = "1.5.1"
+__version__ = "1.5.2"
 __author__ = "FLAMEHAVEN"
 __license__ = "MIT"
 

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "flamehaven-filesearch"
-version = "1.5.1"
+version = "1.5.2"
 description = "FLAMEHAVEN FileSearch - Open source semantic document search with API authentication powered by Google Gemini"
 readme = "README.md"
 requires-python = ">=3.8"