python-hwpx 2.24.0__tar.gz → 2.29.1__tar.gz

{python_hwpx-2.24.0/src/python_hwpx.egg-info → python_hwpx-2.29.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: python-hwpx
-Version: 2.24.0
+Version: 2.29.1
 Summary: 한글 없이 HWPX 문서를 열고, 편집하고, 생성하고, 검증하는 Python 자동화 라이브러리
 Author: python-hwpx Maintainers
 License-Expression: Apache-2.0
@@ -20,7 +20,7 @@ Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
 License-File: NOTICE
-Requires-Dist: lxml<6,>=4.9
+Requires-Dist: lxml<7,>=4.9
 Provides-Extra: visual
 Requires-Dist: pymupdf>=1.24; extra == "visual"
 Requires-Dist: pillow>=10.0; extra == "visual"
@@ -33,6 +33,8 @@ Requires-Dist: twine>=4.0; extra == "dev"
 Requires-Dist: pytest>=7.4; extra == "dev"
 Provides-Extra: test
 Requires-Dist: build>=1.0; extra == "test"
+Requires-Dist: numpy>=1.26; extra == "test"
+Requires-Dist: pillow>=10.0; extra == "test"
 Requires-Dist: pytest>=7.4; extra == "test"
 Requires-Dist: pytest-cov>=5.0; extra == "test"
 Provides-Extra: typecheck
@@ -202,6 +204,15 @@ pip install -e ".[dev]"
 pytest
 ```
 
+## 감사의 말
+
+아래 공개 표준·프로젝트에 빚지고 있습니다.
+
+- **[OWPML — 개방형 워드프로세서 마크업 언어 (KS X 6101)](https://www.kssn.net/search/stddetail.do?itemNo=K001010119985)** — HWPX가 기반하는 한국 산업 표준
+- **[hancom-io/hwpx-owpml-model](https://github.com/hancom-io/hwpx-owpml-model)** — OWPML 요소 구조 참조 모델 · **[neolord0/hwpxlib](https://github.com/neolord0/hwpxlib)** — 오라클 샘플 코퍼스
+- **[edwardkim/rhwp](https://github.com/edwardkim/rhwp)** — 멱등성·검증 게이트 설계 영감
+- **범정부오피스** — 공무 문서 편집 워크플로 아이디어
+
 ## License
 
 Apache License 2.0. See LICENSE and NOTICE.

{python_hwpx-2.24.0 → python_hwpx-2.29.1}/README.md

@@ -160,6 +160,15 @@ pip install -e ".[dev]"
 pytest
 ```
 
+## 감사의 말
+
+아래 공개 표준·프로젝트에 빚지고 있습니다.
+
+- **[OWPML — 개방형 워드프로세서 마크업 언어 (KS X 6101)](https://www.kssn.net/search/stddetail.do?itemNo=K001010119985)** — HWPX가 기반하는 한국 산업 표준
+- **[hancom-io/hwpx-owpml-model](https://github.com/hancom-io/hwpx-owpml-model)** — OWPML 요소 구조 참조 모델 · **[neolord0/hwpxlib](https://github.com/neolord0/hwpxlib)** — 오라클 샘플 코퍼스
+- **[edwardkim/rhwp](https://github.com/edwardkim/rhwp)** — 멱등성·검증 게이트 설계 영감
+- **범정부오피스** — 공무 문서 편집 워크플로 아이디어
+
 ## License
 
 Apache License 2.0. See LICENSE and NOTICE.

{python_hwpx-2.24.0 → python_hwpx-2.29.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "python-hwpx"
-version = "2.24.0"
+version = "2.29.1"
 description = "한글 없이 HWPX 문서를 열고, 편집하고, 생성하고, 검증하는 Python 자동화 라이브러리"
 readme = { file = "README.md", content-type = "text/markdown" }
 license = "Apache-2.0"
@@ -25,7 +25,7 @@ classifiers = [
     "Topic :: Text Processing :: Markup :: XML",
 ]
 dependencies = [
-    "lxml>=4.9,<6",
+    "lxml>=4.9,<7",
 ]
 
 [project.optional-dependencies]
@@ -45,6 +45,8 @@ dev = [
 ]
 test = [
     "build>=1.0",
+    "numpy>=1.26",
+    "pillow>=10.0",
     "pytest>=7.4",
     "pytest-cov>=5.0",
 ]
@@ -59,6 +61,7 @@ Documentation = "https://airmang.github.io/python-hwpx/"
 Issues = "https://github.com/airmang/python-hwpx/issues"
 
 [project.scripts]
+hwpx = "hwpx.agent.cli:main"
 hwpx-unpack = "hwpx.tools.archive_cli:unpack_main"
 hwpx-pack = "hwpx.tools.archive_cli:pack_main"
 hwpx-validate = "hwpx.tools.validator:main"
@@ -100,11 +103,19 @@ testpaths = ["tests"]
 
 [tool.mypy]
 python_version = "3.10"
+# Explicit gradual-migration allowlist: transitive imports enter the gate only
+# after they are added here and made clean themselves.
+follow_imports = "skip"
 files = [
     "src/hwpx/document.py",
     "src/hwpx/oxml/document.py",
     "src/hwpx/opc/package.py",
     "src/hwpx/opc/relationships.py",
+    "src/hwpx/oxml/paragraph.py",
+    "src/hwpx/oxml/run.py",
+    "src/hwpx/oxml/table.py",
+    "src/hwpx/quality/policy.py",
+    "src/hwpx/quality/report.py",
     "src/hwpx/tools/archive_cli.py",
     "src/hwpx/tools/package_validator.py",
     "src/hwpx/tools/page_guard.py",
@@ -126,6 +137,11 @@ ignore_errors = true
 include = [
     "src/hwpx/opc/package.py",
     "src/hwpx/opc/relationships.py",
+    "src/hwpx/oxml/paragraph.py",
+    "src/hwpx/oxml/run.py",
+    "src/hwpx/oxml/table.py",
+    "src/hwpx/quality/policy.py",
+    "src/hwpx/quality/report.py",
     "src/hwpx/tools/archive_cli.py",
     "src/hwpx/tools/package_validator.py",
     "src/hwpx/tools/page_guard.py",

python_hwpx-2.29.1/src/hwpx/agent/__init__.py

@@ -0,0 +1,99 @@
+"""Versioned semantic contracts for the HWPX agent document interface."""
+
+from .model import (
+    AGENT_BATCH_RESULT_SCHEMA,
+    AGENT_BATCH_SCHEMA,
+    AGENT_CATALOG_SCHEMA,
+    AGENT_COMMAND_SCHEMA,
+    AGENT_ERROR_SCHEMA,
+    AGENT_NODE_SCHEMA,
+    AgentBatchResult,
+    AgentContractError,
+    AgentError,
+    AgentNode,
+    agent_contract_manifest,
+    validate_agent_batch,
+    validate_agent_command,
+)
+from .catalog import agent_catalog, agent_json_schemas, catalog_hash, human_help, node_help
+from .commands import apply_document_commands
+from .document import HwpxAgentDocument, NodeRecord
+from .path import PathSegment, SemanticPath, canonicalize_path, parse_path
+from .query import QueryResult, SemanticSelector, parse_selector
+from .blueprint import (
+    BLUEPRINT_CATALOG_SCHEMA,
+    BLUEPRINT_REPLAY_RESULT_SCHEMA,
+    BLUEPRINT_REPLAY_SCHEMA,
+    BLUEPRINT_SCHEMA,
+    BlueprintBundle,
+    BlueprintDumpResult,
+    BlueprintReplayResult,
+    ReplayPlan,
+    blueprint_catalog,
+    blueprint_catalog_hash,
+    blueprint_hash,
+    blueprint_human_help,
+    blueprint_json_schemas,
+    blueprint_limits,
+    build_blueprint_bundle,
+    dump_document_blueprint,
+    read_blueprint_bundle,
+    repack_blueprint_bundle,
+    replay_document_blueprint,
+    validate_blueprint_manifest,
+    validate_replay_request,
+    write_blueprint_bundle,
+)
+
+__all__ = [
+    "AGENT_BATCH_RESULT_SCHEMA",
+    "AGENT_BATCH_SCHEMA",
+    "AGENT_CATALOG_SCHEMA",
+    "AGENT_COMMAND_SCHEMA",
+    "AGENT_ERROR_SCHEMA",
+    "AGENT_NODE_SCHEMA",
+    "BLUEPRINT_CATALOG_SCHEMA",
+    "BLUEPRINT_REPLAY_RESULT_SCHEMA",
+    "BLUEPRINT_REPLAY_SCHEMA",
+    "BLUEPRINT_SCHEMA",
+    "AgentBatchResult",
+    "AgentContractError",
+    "AgentError",
+    "AgentNode",
+    "BlueprintReplayResult",
+    "ReplayPlan",
+    "BlueprintBundle",
+    "BlueprintDumpResult",
+    "HwpxAgentDocument",
+    "NodeRecord",
+    "PathSegment",
+    "QueryResult",
+    "SemanticPath",
+    "SemanticSelector",
+    "agent_catalog",
+    "agent_json_schemas",
+    "agent_contract_manifest",
+    "apply_document_commands",
+    "blueprint_catalog",
+    "blueprint_catalog_hash",
+    "blueprint_hash",
+    "blueprint_human_help",
+    "blueprint_json_schemas",
+    "blueprint_limits",
+    "build_blueprint_bundle",
+    "canonicalize_path",
+    "catalog_hash",
+    "human_help",
+    "node_help",
+    "parse_path",
+    "parse_selector",
+    "dump_document_blueprint",
+    "read_blueprint_bundle",
+    "repack_blueprint_bundle",
+    "replay_document_blueprint",
+    "validate_agent_batch",
+    "validate_agent_command",
+    "validate_blueprint_manifest",
+    "validate_replay_request",
+    "write_blueprint_bundle",
+]

python_hwpx-2.29.1/src/hwpx/agent/blueprint/__init__.py

@@ -0,0 +1,58 @@
+"""Typed semantic HWPX blueprint contracts."""
+
+from .catalog import (
+    blueprint_catalog,
+    blueprint_catalog_hash,
+    blueprint_human_help,
+    blueprint_json_schemas,
+)
+from .bundle import (
+    BlueprintBundle,
+    build_blueprint_bundle,
+    read_blueprint_bundle,
+    repack_blueprint_bundle,
+    write_blueprint_bundle,
+)
+from .dump import BlueprintDumpResult, dump_document_blueprint
+from .mapping import ReplayPlan
+from .replay import replay_document_blueprint
+from .model import (
+    BLUEPRINT_CATALOG_SCHEMA,
+    BLUEPRINT_REPLAY_RESULT_SCHEMA,
+    BLUEPRINT_REPLAY_SCHEMA,
+    BLUEPRINT_SCHEMA,
+    BlueprintReplayResult,
+    blueprint_hash,
+    blueprint_limits,
+    canonical_manifest_bytes,
+    validate_blueprint_manifest,
+    validate_replay_request,
+    with_blueprint_hash,
+)
+
+__all__ = [
+    "BLUEPRINT_CATALOG_SCHEMA",
+    "BLUEPRINT_REPLAY_RESULT_SCHEMA",
+    "BLUEPRINT_REPLAY_SCHEMA",
+    "BLUEPRINT_SCHEMA",
+    "BlueprintBundle",
+    "BlueprintDumpResult",
+    "BlueprintReplayResult",
+    "ReplayPlan",
+    "blueprint_catalog",
+    "blueprint_catalog_hash",
+    "blueprint_hash",
+    "blueprint_human_help",
+    "blueprint_json_schemas",
+    "blueprint_limits",
+    "build_blueprint_bundle",
+    "canonical_manifest_bytes",
+    "dump_document_blueprint",
+    "read_blueprint_bundle",
+    "repack_blueprint_bundle",
+    "replay_document_blueprint",
+    "validate_blueprint_manifest",
+    "validate_replay_request",
+    "with_blueprint_hash",
+    "write_blueprint_bundle",
+]

python_hwpx-2.29.1/src/hwpx/agent/blueprint/bundle.py

@@ -0,0 +1,282 @@
+# SPDX-License-Identifier: Apache-2.0
+"""Deterministic and fail-closed ``.hwpxbp`` bundle I/O."""
+
+from __future__ import annotations
+
+import hashlib
+import io
+import json
+import os
+import stat
+import tempfile
+import zipfile
+from collections.abc import Mapping
+from dataclasses import dataclass
+from pathlib import Path, PurePosixPath
+from typing import Any
+
+from ..model import AgentContractError
+from .model import (
+    ASSET_PATH_PATTERN,
+    MAX_ASSETS,
+    MAX_ASSET_BYTES,
+    MAX_MANIFEST_BYTES,
+    MAX_TOTAL_ASSET_BYTES,
+    canonical_manifest_bytes,
+    validate_blueprint_manifest,
+    with_blueprint_hash,
+)
+
+MANIFEST_PATH = "blueprint.json"
+MAX_BUNDLE_BYTES = MAX_MANIFEST_BYTES + MAX_TOTAL_ASSET_BYTES + 2 * 1024 * 1024
+MAX_COMPRESSION_RATIO = 100
+
+ALLOWED_MEDIA_TYPES: dict[str, tuple[str, ...]] = {
+    "image/png": ("png",),
+    "image/jpeg": ("jpg", "jpeg"),
+    "image/gif": ("gif",),
+    "image/bmp": ("bmp",),
+    "image/tiff": ("tif", "tiff"),
+    "image/webp": ("webp",),
+}
+
+
+@dataclass(frozen=True, slots=True)
+class BlueprintBundle:
+    """A fully validated manifest and detached content-addressed assets."""
+
+    manifest: Mapping[str, Any]
+    assets: Mapping[str, bytes]
+    bundle_sha256: str
+    size: int
+
+
+def _error(code: str, message: str, *, target: str = "bundle") -> AgentContractError:
+    return AgentContractError(code, message, target=target)
+
+
+def _json_without_duplicate_keys(data: bytes) -> dict[str, Any]:
+    def pairs_hook(pairs: list[tuple[str, Any]]) -> dict[str, Any]:
+        result: dict[str, Any] = {}
+        for key, value in pairs:
+            if key in result:
+                raise _error("invariant_violation", f"duplicate JSON field: {key}", target=MANIFEST_PATH)
+            result[key] = value
+        return result
+
+    try:
+        value = json.loads(data.decode("utf-8"), object_pairs_hook=pairs_hook)
+    except AgentContractError:
+        raise
+    except (UnicodeDecodeError, json.JSONDecodeError) as exc:
+        raise _error("invalid_syntax", "blueprint.json is not strict UTF-8 JSON", target=MANIFEST_PATH) from exc
+    if not isinstance(value, dict):
+        raise _error("invalid_syntax", "blueprint.json must contain one object", target=MANIFEST_PATH)
+    return value
+
+
+def _validate_entry_name(name: str) -> None:
+    if not name or "\\" in name or "\x00" in name:
+        raise _error("verification_failed", "bundle entry path is not normalized", target=name or "bundle")
+    path = PurePosixPath(name)
+    if path.is_absolute() or name.startswith("/") or any(part in {"", ".", ".."} for part in path.parts):
+        raise _error("verification_failed", "bundle entry path is unsafe", target=name)
+    if name != MANIFEST_PATH and not ASSET_PATH_PATTERN.fullmatch(name):
+        raise _error("unsupported_content", "bundle contains an unknown or forbidden entry", target=name)
+
+
+def _validate_zip_info(info: zipfile.ZipInfo) -> None:
+    _validate_entry_name(info.filename)
+    if info.is_dir():
+        raise _error("unsupported_content", "directory entries are forbidden", target=info.filename)
+    mode = (info.external_attr >> 16) & 0xFFFF
+    if mode and stat.S_ISLNK(mode):
+        raise _error("unsupported_content", "symlink entries are forbidden", target=info.filename)
+    if info.flag_bits & 0x1:
+        raise _error("unsupported_content", "encrypted bundle entries are forbidden", target=info.filename)
+    if info.compress_type not in {zipfile.ZIP_STORED, zipfile.ZIP_DEFLATED}:
+        raise _error("unsupported_content", "bundle compression method is forbidden", target=info.filename)
+    limit = MAX_MANIFEST_BYTES if info.filename == MANIFEST_PATH else MAX_ASSET_BYTES
+    if info.file_size > limit:
+        raise _error("resource_limit", "bundle entry exceeds its byte limit", target=info.filename)
+    if info.file_size and info.compress_size == 0:
+        raise _error("resource_limit", "bundle entry has an invalid compression ratio", target=info.filename)
+    if info.compress_size and info.file_size / info.compress_size > MAX_COMPRESSION_RATIO:
+        raise _error("resource_limit", "bundle entry exceeds the decompression-ratio limit", target=info.filename)
+
+
+def _sniff_media_type(data: bytes) -> str | None:
+    if data.startswith(b"\x89PNG\r\n\x1a\n"):
+        return "image/png"
+    if data.startswith(b"\xff\xd8\xff"):
+        return "image/jpeg"
+    if data.startswith((b"GIF87a", b"GIF89a")):
+        return "image/gif"
+    if data.startswith(b"BM"):
+        return "image/bmp"
+    if data.startswith((b"II*\x00", b"MM\x00*")):
+        return "image/tiff"
+    if len(data) >= 12 and data[:4] == b"RIFF" and data[8:12] == b"WEBP":
+        return "image/webp"
+    return None
+
+
+def _validated_assets(
+    manifest: Mapping[str, Any], assets: Mapping[str, bytes]
+) -> dict[str, bytes]:
+    declared = {str(item["assetPath"]): dict(item) for item in manifest["resources"]}
+    if set(assets) != set(declared):
+        raise _error(
+            "invariant_violation",
+            "bundle assets do not exactly match manifest resources",
+            target="resources",
+        )
+    if len(assets) > MAX_ASSETS:
+        raise _error("resource_limit", "asset count exceeds limit", target="resources")
+    total = 0
+    detached: dict[str, bytes] = {}
+    for path in sorted(assets):
+        _validate_entry_name(path)
+        payload = assets[path]
+        if not isinstance(payload, bytes):
+            raise _error("invalid_syntax", "asset payload must be bytes", target=path)
+        total += len(payload)
+        if len(payload) > MAX_ASSET_BYTES or total > MAX_TOTAL_ASSET_BYTES:
+            raise _error("resource_limit", "asset bytes exceed limit", target=path)
+        record = declared[path]
+        digest = "sha256:" + hashlib.sha256(payload).hexdigest()
+        if digest != record["sha256"] or len(payload) != record["size"]:
+            raise _error("verification_failed", "asset hash or size mismatch", target=path)
+        media_type = str(record["mediaType"])
+        suffix = path.rsplit(".", 1)[-1]
+        if media_type not in ALLOWED_MEDIA_TYPES or suffix not in ALLOWED_MEDIA_TYPES[media_type]:
+            raise _error("unsupported_content", "asset extension/MIME is not allow-listed", target=path)
+        if _sniff_media_type(payload) != media_type:
+            raise _error("verification_failed", "asset bytes do not match declared MIME", target=path)
+        detached[path] = bytes(payload)
+    return detached
+
+
+def _zip_info(name: str) -> zipfile.ZipInfo:
+    info = zipfile.ZipInfo(name, date_time=(1980, 1, 1, 0, 0, 0))
+    info.compress_type = zipfile.ZIP_STORED
+    info.create_system = 0
+    info.external_attr = 0
+    info.internal_attr = 0
+    info.flag_bits = 0
+    return info
+
+
+def build_blueprint_bundle(
+    manifest: Mapping[str, Any], assets: Mapping[str, bytes] | None = None
+) -> bytes:
+    """Return canonical bundle bytes for an already hashed manifest."""
+
+    checked = validate_blueprint_manifest(manifest)
+    checked_assets = _validated_assets(checked, assets or {})
+    manifest_bytes = canonical_manifest_bytes(checked, include_hash=True)
+    stream = io.BytesIO()
+    with zipfile.ZipFile(stream, "w", allowZip64=False) as archive:
+        archive.writestr(_zip_info(MANIFEST_PATH), manifest_bytes)
+        for name, payload in checked_assets.items():
+            archive.writestr(_zip_info(name), payload)
+    result = stream.getvalue()
+    if len(result) > MAX_BUNDLE_BYTES:
+        raise _error("resource_limit", "blueprint bundle exceeds total byte limit")
+    return result
+
+
+def write_blueprint_bundle(
+    destination: str | os.PathLike[str],
+    manifest: Mapping[str, Any],
+    assets: Mapping[str, bytes] | None = None,
+    *,
+    overwrite: bool = False,
+) -> BlueprintBundle:
+    """Atomically write one deterministic bundle and return its validated view."""
+
+    path = Path(destination)
+    if path.suffix.casefold() != ".hwpxbp":
+        raise _error("invalid_syntax", "blueprint output must use .hwpxbp", target="output")
+    if path.exists() and not overwrite:
+        raise _error("identity_collision", "blueprint output already exists", target=str(path))
+    data = build_blueprint_bundle(manifest, assets)
+    path.parent.mkdir(parents=True, exist_ok=True)
+    descriptor, temp_name = tempfile.mkstemp(prefix=f".{path.name}.", suffix=".tmp", dir=path.parent)
+    try:
+        with os.fdopen(descriptor, "wb") as handle:
+            handle.write(data)
+            handle.flush()
+            os.fsync(handle.fileno())
+        if path.exists() and not overwrite:
+            raise _error("identity_collision", "blueprint output already exists", target=str(path))
+        os.replace(temp_name, path)
+    except Exception:
+        try:
+            os.unlink(temp_name)
+        except FileNotFoundError:
+            pass
+        raise
+    return read_blueprint_bundle(data)
+
+
+def read_blueprint_bundle(source: str | os.PathLike[str] | bytes) -> BlueprintBundle:
+    """Validate a complete bundle in memory before returning any materialized data."""
+
+    data = bytes(source) if isinstance(source, bytes) else Path(source).read_bytes()
+    if len(data) > MAX_BUNDLE_BYTES:
+        raise _error("resource_limit", "blueprint bundle exceeds total byte limit")
+    try:
+        with zipfile.ZipFile(io.BytesIO(data), "r") as archive:
+            infos = archive.infolist()
+            names = [info.filename for info in infos]
+            if len(names) != len(set(names)):
+                raise _error("invariant_violation", "bundle contains duplicate entry paths")
+            if names.count(MANIFEST_PATH) != 1:
+                raise _error("invariant_violation", "bundle must contain exactly one blueprint.json")
+            if len(infos) > MAX_ASSETS + 1:
+                raise _error("resource_limit", "bundle entry count exceeds limit")
+            for info in infos:
+                _validate_zip_info(info)
+            if sum(info.file_size for info in infos if info.filename != MANIFEST_PATH) > MAX_TOTAL_ASSET_BYTES:
+                raise _error("resource_limit", "bundle assets exceed total byte limit", target="resources")
+            manifest_data = archive.read(MANIFEST_PATH)
+            manifest = validate_blueprint_manifest(_json_without_duplicate_keys(manifest_data))
+            asset_names = sorted(name for name in names if name != MANIFEST_PATH)
+            assets = {name: archive.read(name) for name in asset_names}
+    except AgentContractError:
+        raise
+    except (OSError, zipfile.BadZipFile, zipfile.LargeZipFile, RuntimeError) as exc:
+        raise _error("invalid_syntax", "invalid .hwpxbp ZIP container") from exc
+    checked_assets = _validated_assets(manifest, assets)
+    return BlueprintBundle(
+        manifest=manifest,
+        assets=checked_assets,
+        bundle_sha256="sha256:" + hashlib.sha256(data).hexdigest(),
+        size=len(data),
+    )
+
+
+def repack_blueprint_bundle(
+    source: str | os.PathLike[str] | bytes,
+    destination: str | os.PathLike[str],
+    manifest: Mapping[str, Any],
+    *,
+    overwrite: bool = False,
+) -> BlueprintBundle:
+    """Safely repack edited typed JSON while preserving only validated assets."""
+
+    original = read_blueprint_bundle(source)
+    rehashed = with_blueprint_hash(manifest)
+    return write_blueprint_bundle(destination, rehashed, original.assets, overwrite=overwrite)
+
+
+__all__ = [
+    "ALLOWED_MEDIA_TYPES",
+    "BlueprintBundle",
+    "MANIFEST_PATH",
+    "build_blueprint_bundle",
+    "read_blueprint_bundle",
+    "repack_blueprint_bundle",
+    "write_blueprint_bundle",
+]