hadith-detector 0.1.0__tar.gz

hadith_detector-0.1.0/.gitattributes

@@ -0,0 +1,10 @@
+* text=auto
+*.py text eol=lf
+*.toml text eol=lf
+*.yml text eol=lf
+*.md text eol=lf
+*.csv text eol=lf
+*.json text eol=lf
+*.lock text eol=lf
+# Preserve checksum-pinned upstream notices byte-for-byte on every platform.
+tools/bundle-notices/** -text

hadith_detector-0.1.0/.gitignore

@@ -0,0 +1,72 @@
+# === Python ignores ===
+__pycache__/
+*.pyc
+*.pyo
+*.pyd
+*.pkl
+*.db
+*.sqlite3
+
+# Virtual environment folders
+venv/
+env/
+.venv/
+.env/
+ENV/
+*.env
+.env
+
+# Logs
+*.log
+logs/*
+!logs/.gitkeep
+
+# IDE settings
+.idea/
+
+# OS files
+.DS_Store
+Thumbs.db
+
+# Secret / private keys
+*secret*.json
+*key*.json
+
+# Deployment / containers
+*.pid
+*.sock
+*.crt
+*.pem
+
+# Coverage / test artifacts
+*.coverage
+htmlcov/
+
+# Cache folders
+.cache/
+.ruff_cache/
+.mypy_cache/
+dist/
+build/
+
+# Docker-related
+docker-compose.override.yml
+
+# Data
+src/hadith_detector/_data/
+data/*
+!data/.gitkeep
+!data/README.md
+.env.*
+!.env.example
+.ipynb_checkpoints/
+.pytest_cache/
+.coverage*
+*.egg-info/
+
+# Local research (preserved, not part of the supported package)
+notebooks/*
+!notebooks/phase6_review.ipynb
+!notebooks/phase6_research_review.ipynb
+!notebooks/phase6_review_50.ipynb
+others/*

hadith_detector-0.1.0/.python-version

@@ -0,0 +1 @@
+3.14

hadith_detector-0.1.0/CHANGELOG.md

@@ -0,0 +1,96 @@
+# Changelog
+
+Release notes are maintained manually. Versions use Git tags and 0.x development
+conventions described in [CONTRIBUTING.md](CONTRIBUTING.md).
+
+## Unreleased
+
+### Fixed
+
+- Release-readiness probes execute the current README examples without depending
+  on variable names from an earlier draft; documentation links also work on PyPI.
+- Clean-wheel CI checks now install from hash-verified locked dependency wheels,
+  without relying on uv registry metadata left in a developer's cache.
+
+### Added
+
+- One bundled Bukhari/Muslim corpus with prebuilt MARISA v2 resources, optional
+  `index_dir` defaults, offline staging and fail-closed distribution builds (Phase 7).
+- Manual, non-publishing release-readiness workflow for actual bundled candidates
+  on Windows/Linux and Python 3.11–3.14, plus a professional README draft.
+- Finalized the 55-input Phase 6 technical pilot with 382 completed human judgments,
+  reproducible offline scoring, evidence validation and separate release-description wording.
+- Expanded the technical pilot to 55 frozen inputs, including all 20 commentary
+  paragraphs, with four-group reporting and explicitly approved, lossless review migration.
+- Frozen 50-input technical pilot with source-linked comparisons, 62 deduplicated
+  passage pairs and separate automatic review progress; full intent research remains optional.
+- Interactive Phase 6 ipywidgets review notebook with exact-offset Arabic previews,
+  lazy supporting evidence and atomic, resumable CSV progress saves.
+- Repository-only Phase 6 input freezing, six-configuration evidence capture,
+  reversible review CSV, candidate-unwanted-alert inventory and offline replay.
+- Pinned natural-source provenance and input-first draft spans. Original intent labels
+  remain provisional; the smaller technical pilot now has approved correspondence review.
+
+- Instance and convenience `annotate()` methods returning safely escaped HTML
+  fragments with fixed marks, original offsets and best supporting costs (Phase 5B).
+- Text/offset round trips, injection, exact/fuzzy parity and annotation concurrency checks.
+- Carriage-return preservation and explicit annotation-only rejection of NUL/lone surrogates.
+
+- Frozen `DetectorConfig`, directory constructor, read-only resolved `index_dir`,
+  and convenience `detect()`/`get_hadith()` with a synchronized single-entry cache.
+- Validated result `from_dict()` methods and isolated JSON metadata round trips.
+- Per-instance reentrant locking: concurrent detection and lookup calls execute sequentially.
+- Public API, cache/concurrency and deserialization regression checks (Phase 5A).
+
+- Prebuilt MARISA v2 vocabulary/trigram lookups with explicit preserved IDs,
+  complete resource validation and deterministic offline construction.
+- Native-resource corruption tests, supported-version and offline installation checks.
+
+- Production fuzzy detection with simultaneous character-edit and internal source-word
+  omission budgets, complete anchor retrieval and two-pass occurrence recovery.
+- Per-occurrence costs, canonical word alignments and omitted-token models; JSON result
+  dictionaries now include these fields, including zero-cost traces on exact results.
+
+- Independent exhaustive fuzzy alignment oracle and tiny direct path enumeration.
+- Complete first-word, guaranteed-anchor and trigram-hybrid retrieval experiments,
+  with deterministic word traces, internal source omissions and combined budgets.
+- Repository-only RapidFuzz 3.14.6 comparison, isolated timing/memory tooling and
+  supported-version backend compatibility checks (Phase 4A).
+
+- Installable `hadith_detector` foundation with metadata-derived version and typing marker.
+- Python 3.11–3.14 policy, uv lockfile, Ruff, strict mypy, pytest, and build validation.
+- CI covering Linux and Windows, including clean wheel and sdist installation checks.
+- Project status, phased roadmap, contribution workflow, and data artifact policy.
+- Pinned QuranLab source lock and repository-only fetch/build/verify workflow.
+- Shared versioned Arabic tokenization with original spans and matching barriers.
+- Canonical JSONL artifacts, build manifests, warning reports, and reproducibility checks.
+- Synthetic preprocessing CI and full-corpus acceptance evidence.
+- Repository-only exhaustive exact-match oracle and positional/2-/3-/5-word prototypes.
+- Direct substring-enumeration checks, deterministic query generation and isolated memory/latency benchmarks.
+
+- Compact versioned trigram resources with complete unigram fallback, checksums,
+  structural validation, and deterministic offline construction.
+- Provisional `HadithDetector.from_directory()`, exact detection and source lookup;
+  typed serializable results with all supporting references and occurrences.
+- Runtime/oracle comparison, corruption tests, metadata isolation and clean-install checks.
+
+### Changed
+
+- Detection settings now inherit defaults when omitted or `None`; explicit zero overrides.
+- The provisional constructor now takes a directory; `from_directory()` remains supported.
+
+- Require `marisa-trie==1.4.1` alongside RapidFuzz. Index format v2 replaces v1;
+  rebuild into a new directory from the prepared corpus and preserve v1 for rollback.
+- Accept increased one-time validated initialization for substantially lower memory
+  in long-lived instances. Phase 5A now serializes shared-instance calls.
+
+- RapidFuzz is the sole required production distance engine (`>=3.14.6,<4`);
+  Python distance implementations remain repository-only testing/benchmark tools.
+
+- Dependency policy now requires offline operation while allowing evidence-based
+  adoption of maintained dependencies.
+
+- Replaced the unused application scaffold with a library using no runtime dependencies.
+
+Annotation and bundled resources remain planned. Runtime distributions
+contain neither the corpus nor preprocessing dependencies.

hadith_detector-0.1.0/CONTRIBUTING.md

@@ -0,0 +1,127 @@
+# Contributing
+
+## Environment and checks
+
+Use Python 3.14 locally and uv (CI pins uv 0.11.26). The supported test matrix is
+Python 3.11–3.14 on Linux and 3.14 on Windows. Install the checkout before importing;
+do not set `PYTHONPATH` or add the source directory to test paths.
+
+```sh
+uv sync --locked
+uv run ruff check .
+uv run ruff format --check .
+uv run mypy src/hadith_detector
+uv run pytest
+```
+
+Default resource use and production builds now require offline staging of the approved
+bundle. Packaging tests prepare synthetic bundles in isolated source trees, so normal
+CI does not fetch the corpus. For real wheel/sdist builds, stage verified resources
+first and follow [release readiness](docs/release-readiness.md). Its manual workflow
+tests actual bundled candidates on both Linux and Windows, Python 3.11–3.14.
+
+In PowerShell, if wildcard expansion for twine is unavailable, use
+`Get-ChildItem dist -File | ForEach-Object { uv run twine check $_.FullName }`.
+Packaging tests invoke uv, resolve isolated build dependencies, build a wheel from
+an sdist without Git metadata, and install both wheels in fresh environments.
+Preparation downloads compatible runtime dependency wheels listed in `uv.lock`
+and verifies their SHA-256 hashes. Both clean installations then use that local
+wheelhouse with `--offline --no-index --no-cache`, so a warm registry cache cannot
+hide missing dependencies. Installed-library probes also block network access.
+Build/dependency preparation may access the network; offline installation and
+runtime imports do not.
+Use `uv run pytest -m "not packaging"` for the fast foundation checks.
+
+Phase 2 preprocessing uses `uv sync --locked --group data`. Run its tests with
+`uv run --frozen --group data pytest`; the default environment skips the synthetic
+pipeline tests when preprocessing dependencies are absent. Shared normalization
+tests run on every supported Python. Full builds require Python 3.14. Ordinary CI
+uses synthetic Parquet data and does not fetch the real corpus. See the
+[data-preparation guide](docs/data-preparation.md) for full-corpus validation.
+
+Index experiments use `uv sync --locked --group benchmark`. Run the synthetic
+worker smoke test with `uv run --frozen --group benchmark pytest tests/test_exact_benchmark.py`.
+The exhaustive oracle, retrieval, serialization and randomized reference tests
+need no benchmark dependency; only the subprocess memory-measurement smoke test
+needs psutil. Full-corpus performance experiments are local and never run in CI.
+See [benchmark usage](docs/index-benchmark.md).
+
+Phase 3B runtime and synthetic index-builder tests need no optional dependencies
+and run across the supported Python matrix. Full-corpus index construction uses
+Python 3.14 and the verified prepared artifacts; see [runtime resources](docs/runtime-index.md).
+Fresh-process validation measurements use the separate `benchmark` group.
+
+Phase 4B requires RapidFuzz at runtime; the benchmark group pins 3.14.6 for reproduction.
+Run the repository's synthetic
+oracle/retrieval/backend checks with `uv run --frozen --group benchmark pytest tests/test_fuzzy_benchmark.py`.
+The normal environment skips RapidFuzz parity and subprocess tests if their optional
+tools are absent. See [fuzzy experiments](docs/fuzzy-benchmark.md) for full-corpus commands.
+Dependencies are assessed on their merits; offline operation is required, while
+zero runtime dependencies is not a permanent project rule. The maintainer approved
+RapidFuzz as the sole production distance engine and authorized Phase 4B. Python
+distance implementations stay in repository tests/benchmarks. Further phase changes
+require review and approval.
+
+Ruff handles linting and formatting (100-character lines, Python 3.11 syntax).
+Mypy checks package code in strict mode. Regenerate `uv.lock` with `uv lock` only
+when dependencies change, then rerun checks. No pre-commit hook is required.
+
+## Review and Git workflow
+
+Work in independently reviewable phases. Update documentation and the Unreleased
+changelog with behavior changes. Run relevant checks before each commit, inspect
+the explicit staged diff, and run the complete suite before phase acceptance.
+Do not stage ignored research, raw downloads, large artifacts, or unrelated changes.
+
+Phase 1 commit boundaries:
+
+1. `chore: establish installable library foundation`
+2. `test: validate package builds and add CI`
+3. `docs: document project status and phased roadmap`
+
+Phase 2 proposed commit boundaries:
+
+1. `data: pin QuranLab sources and document provenance`
+2. `feat: add mapped Arabic normalization and tokenization`
+3. `feat: add reproducible corpus preparation`
+4. `docs: record full-corpus preparation validation`
+
+Phase 3A proposed commit boundaries:
+
+1. `test: establish exhaustive exact-match oracle`
+2. `perf: benchmark positional and n-gram indexes`
+3. `docs: record index benchmark and recommendation`
+
+Phase 3B proposed commit boundaries:
+
+1. `feat: add compact offline index resources`
+2. `feat: implement provisional exact detector`
+3. `docs: record runtime validation and API usage`
+
+Commit and push when requested. Preserve existing history and origin; never
+force-push. Remote changes, destructive Git operations, release tags, public
+releases, and PyPI publication require explicit approval. Stop after each phase
+for review before implementing the next.
+
+## Versions and releases
+
+Setuptools-scm derives versions from Git; fetch full history and tags in CI.
+Untagged builds are development versions, not a stable `0.1.0`. Import
+`hadith_detector.__version__` to read installed distribution metadata; runtime
+does not need Git or setuptools-scm. Build from a checkout or an official sdist;
+arbitrary source copies without Git or sdist metadata are unsupported.
+
+Reserve `v0.1.0` for the first approved usable release. Milestones need no tags.
+During 0.x development, minor versions introduce features or breaking changes;
+patch versions contain compatible fixes. Document incompatibilities and manually
+move Unreleased notes into a dated release section when releasing.
+
+Record schemas, normalization, and index formats have independent versions.
+Dataset builds identify pinned upstream revision, preprocessing code fingerprint,
+configuration, and checksums. Corpus rights are separate from MIT; source-specific
+terms and attribution are documented for the approved Arabic configurations.
+No publishing workflow is enabled in Phase 1.
+
+Local notebooks and evaluation files are preserved but are not supported package
+code. Use separate environments for their exploratory dependencies; the new
+foundation environment intentionally excludes the former web/research dependencies.

hadith_detector-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Mohamed Rayen Fekih
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

hadith_detector-0.1.0/MANIFEST.in

@@ -0,0 +1,24 @@
+include README.md CONTRIBUTING.md CHANGELOG.md LICENSE pyproject.toml
+include setup.py
+include src/hadith_detector/_bundle.json
+recursive-include src/hadith_detector/_data *
+recursive-include src/hadith_detector *.py py.typed
+recursive-include tests *.py
+recursive-include docs *.md
+# This report records final archive checksums; keep it outside the archives it describes.
+exclude docs/phase7-validation.md
+recursive-include tools *.py *.json
+recursive-include tools/bundle-notices *.md *.txt *.csv
+include uv.lock
+include data/README.md
+prune .github
+prune .vscode
+prune .cache
+prune .ruff_cache
+prune .mypy_cache
+prune .pytest_cache
+prune .venv
+prune notebooks
+prune others
+prune logs
+global-exclude *.py[cod] .gitkeep

hadith_detector-0.1.0/PKG-INFO

@@ -0,0 +1,170 @@
+Metadata-Version: 2.4
+Name: hadith-detector
+Version: 0.1.0
+Summary: Deterministic offline Arabic textual matching against Sahih Bukhari and Sahih Muslim
+Author: Mohamed Rayen Fekih
+License-Expression: MIT
+Project-URL: Repository, https://github.com/RayenFekih/hadith-detector
+Project-URL: Issues, https://github.com/RayenFekih/hadith-detector/issues
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Typing :: Typed
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: rapidfuzz<4,>=3.14.6
+Requires-Dist: marisa-trie==1.4.1
+Dynamic: license-file
+
+# Hadith Detector
+
+**Detect Arabic Hadith text and find its sources. Offline, in Python.**
+
+[![Python 3.11–3.14](https://img.shields.io/badge/Python-3.11%E2%80%933.14-3776AB?logo=python&logoColor=white)](https://github.com/RayenFekih/hadith-detector/blob/main/pyproject.toml)
+[![CI](https://github.com/RayenFekih/hadith-detector/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/RayenFekih/hadith-detector/actions/workflows/ci.yml)
+[![Code license: MIT](https://img.shields.io/badge/Code_license-MIT-blue.svg)](https://github.com/RayenFekih/hadith-detector/blob/main/LICENSE)
+
+`hadith-detector` is a Python library that finds Arabic passages matching Hadith texts and returns their supporting sources. It works with short quotations or longer documents, supports exact and configurable fuzzy matching, and runs locally without an API key or an internet connection after installation.
+
+The current bundled corpus covers **Sahih al-Bukhari and Sahih Muslim**. Support for additional Hadith collections is planned.
+
+## Installation
+
+```bash
+pip install hadith-detector
+```
+
+Or with [uv](https://docs.astral.sh/uv/):
+
+```bash
+uv add hadith-detector
+```
+
+Requires **Python 3.11–3.14**. The package includes the Hadith texts and a prebuilt search index, so there is no separate corpus download or index-building step. Detection works offline after installation.
+
+## Quick start
+
+Detect a Hadith quotation and highlight it in the original Arabic text:
+
+```python
+from hadith_detector import detect, annotate
+
+text = "قال رسول الله ﷺ: إنما الأعمال بالنيات وإنما لكل امرئ ما نوى."
+
+matches = detect(text)
+print(matches[0].text)
+print(matches[0].references[0].record.hadith_key)
+
+html = annotate(text)
+print(html)  # Escaped HTML fragment with <mark> around detected passages
+```
+
+`detect()` returns matched passages, their original character offsets, and supporting Hadith references. A passage may have multiple supporting references. `annotate()` returns an HTML fragment that highlights matches without changing the original Arabic wording.
+
+## Features
+
+- **Exact Arabic text matching**, with normalization of supported Arabic letter forms and diacritics.
+- **Optional fuzzy matching** for limited spelling differences and omitted words from the source.
+- **Source references** with complete original records, occurrences, offsets, and alignment details.
+- **Multiple detections** within the same paragraph or document.
+- **HTML highlighting** of detected passages using `annotate()`.
+- **Offline and deterministic** operation with a bundled index; no LLM or external service required.
+
+## Configuration
+
+The default settings are intentionally strict:
+
+| Parameter | Default | What it controls |
+| --- | ---: | --- |
+| `min_match_words` | `5` | Minimum number of matched input words. Lower values also detect more short, common expressions; higher values can miss short quotations. |
+| `max_errors` | `0` | Total number of character edits allowed across the matched words. |
+| `max_missing_words` | `0` | Number of internal words in the source passage that may be absent from the input. |
+
+```python
+from hadith_detector import DetectorConfig, HadithDetector
+
+detector = HadithDetector(
+    config=DetectorConfig(
+        min_match_words=5,
+        max_errors=1,
+        max_missing_words=1,
+    )
+)
+
+matches = detector.detect("إنما الأعمال بالنياث وإنما لكل امرئ ما نوى")
+
+# Override the settings for one call without changing the configuration.
+exact_matches = detector.detect(
+    "إنما الأعمال بالنيات وإنما لكل امرئ ما نوى", max_errors=0, max_missing_words=0
+)
+```
+
+For long Arabic documents, consider a higher `min_match_words` if short incidental matches create too much noise. Enable fuzzy budgets only when your input may contain spelling variations or source-word omissions. Broader settings can increase both detections and processing time. There is no universally best configuration; test the trade-offs on your own texts.
+
+## More usage examples
+
+**Retrieve a complete source record**
+
+```python
+from hadith_detector import get_hadith
+
+record = get_hadith("bukhari:1")
+print(record.text)
+```
+
+**Highlight matches in plain text**
+
+```python
+from hadith_detector import annotate
+
+html = annotate("ورد في النص: إنما الأعمال بالنيات وإنما لكل امرئ ما نوى")
+```
+
+`annotate()` produces an escaped HTML fragment with matched passages wrapped in `<mark>` elements. It is designed for plain-text input, not for editing an existing HTML document.
+
+## Important notes
+
+- **Generic phrases can trigger incidental matches.** For example, a phrase such as `هو عبد الله` may also appear in Hadith records even when it occurs in an unrelated article. Likewise, narration formulas such as `عن النبي صلى الله عليه وسلم قال` may match without being meaningful standalone quotations. Shorter `min_match_words` values increase this risk, and permissive fuzzy budgets can introduce additional matches. **Tune `min_match_words`, `max_errors`, and `max_missing_words` for your text and use case**, especially for long documents; increasing the minimum can reduce short incidental matches but may miss genuine short quotations.
+- **A match means textual support, not quotation intent.** Common expressions, honorifics, Quranic wording embedded in narrations, and narrator-chain fragments can also match. Do not automatically treat every detection as a deliberate Hadith quotation.
+- **Review citations and source context.** Before publishing religious content or relying on the returned citation, inspect the original narration, relevant supporting occurrences, and data provenance. The library does not independently authenticate Hadith or interpret their meaning.
+- **No match does not mean no Hadith exists.** Coverage is currently limited to the bundled Sahihayn collection, and configured matching thresholds may exclude short or modified quotations.
+- **The dataset is preserved as supplied.** Source data can include duplicate narration texts, missing URLs, or damaged characters. Check that it meets your application’s accuracy and provenance requirements.
+
+## Data
+
+The package currently includes **14,940 Arabic Hadith records**:
+
+| Collection | Records |
+| --- | ---: |
+| Sahih al-Bukhari | 7,580 |
+| Sahih Muslim | 7,360 |
+
+The data comes from the [QuranLab Hadith dataset on Hugging Face](https://huggingface.co/datasets/quranlab/hadith), using its `bukhari-ar` and `muslim-ar` configurations. Original Arabic records, source identities, and accompanying metadata are retained. **Additional Hadith books are planned for future versions.**
+
+The library's **code is MIT-licensed**; the underlying Hadith data has separate source terms and notices. Consult the [dataset provenance and redistribution notes](https://github.com/RayenFekih/hadith-detector/blob/main/tools/bundle-notices/PROVENANCE.md) before reusing or redistributing the data independently.
+
+## Use cases
+
+- **Arabic documents and articles:** Locate matching Hadith passages inside longer text.
+- **Research and reference lookup:** Find supporting collections and inspect the original narration.
+- **Educational and editorial tools:** Help reviewers locate quotations and check source wording.
+- **Web applications:** Highlight text matches and show citations or source context.
+- **AI and LLM applications:** Check generated Hadith wording, retrieve source passages, and provide grounding evidence for citation workflows. Human review remains important.
+
+## Documentation and community
+
+- [Public API and result structure](https://github.com/RayenFekih/hadith-detector/blob/main/docs/public-api.md)
+- [Matching and fuzzy behavior](https://github.com/RayenFekih/hadith-detector/blob/main/docs/fuzzy-runtime.md)
+- [HTML annotation](https://github.com/RayenFekih/hadith-detector/blob/main/docs/html-annotation.md)
+- [Evaluation and limitations](https://github.com/RayenFekih/hadith-detector/blob/main/docs/phase6-evaluation.md)
+- [Contributing](https://github.com/RayenFekih/hadith-detector/blob/main/CONTRIBUTING.md)
+- [Changelog](https://github.com/RayenFekih/hadith-detector/blob/main/CHANGELOG.md)
+- [Report a bug or request a feature](https://github.com/RayenFekih/hadith-detector/issues)
+
+Contributions, issue reports, reproducible examples, and suggestions for additional Hadith sources are welcome.