Add METS metadata parsing experiment (DOROP-20)

experiments/dorop-20/Dockerfile

@@ -0,0 +1,58 @@
+FROM python:3.12-slim-bookworm AS base
+
+ARG POETRY_VERSION=1.8.4
+ARG UID=1000
+ARG GID=1000
+
+ENV PYTHONPATH="/app"
+
+RUN groupadd -g ${GID} -o app
+RUN useradd -m -d /app -u ${UID} -g ${GID} -o -s /bin/bash app
+
+RUN apt-get update -yqq && apt-get install -yqq --no-install-recommends \
+  python3-dev \ 
+  build-essential \ 
+  pkg-config \
+  vim-tiny \
+  curl \
+  unzip
+
+WORKDIR /app
+
+ENV PYTHONPATH="/app"
+
+CMD ["tail", "-f", "/dev/null"]
+
+FROM base AS poetry
+
+RUN pip install poetry==${POETRY_VERSION}
+
+ENV PYTHONUNBUFFERED=1\
+  PIP_NO_CACHE_DIR=off \
+  PIP_DISABLE_PIP_VERSION_CHECK=on \
+  PIP_DEFAULT_TIMEOUT=100 \
+  POETRY_NO_INTERACTION=1 \
+  POETRY_VIRTUALENVS_CREATE=1 \
+  POETRY_VIRTUALENVS_IN_PROJECT=1 \
+  POETRY_CACHE_DIR=/tmp/poetry_cache
+
+FROM poetry AS build
+
+COPY pyproject.toml poetry.lock README.md ./
+
+RUN poetry export --without dev -f requirements.txt --output requirements.txt
+
+FROM poetry AS development
+RUN apt-get update -yqq && apt-get install -yqq --no-install-recommends \
+  git
+
+USER app
+
+FROM base AS production
+
+COPY --chown=${UID}:${GID} . /app
+COPY --chown=${UID}:${GID} --from=build "/app/requirements.txt" /app/requirements.txt
+
+RUN pip install -r /app/requirements.txt
+
+USER app

experiments/dorop-20/README.md

@@ -0,0 +1,7 @@
+# Metadata extraction experiment
+
+## Tests
+
+```sh
+docker compose run poetry run pytest
+```

experiments/dorop-20/compose.yml

@@ -0,0 +1,16 @@
+services:
+  app:
+    build: 
+      context: .
+      target: development
+      dockerfile: Dockerfile
+      args:
+        UID: ${UID:-1000}
+        GID: ${GID:-1000}
+        DEV: ${DEV:-false}
+        POETRY_VERSION: ${POETRY_VERSION:-1.8.4}
+    volumes:
+      - .:/app
+      - ../../tests/fixtures/test_submission_package:/app/tests/fixtures/test_submission_package
+    tty: true
+    stdin_open: true

experiments/dorop-20/metadata/element_adapter.py

@@ -0,0 +1,46 @@
+import xml.etree.ElementTree as ET
+from xml.etree.ElementTree import Element
+
+from metadata.exceptions import DataNotFoundError
+
+class ElementAdapter():
+
+    @classmethod
+    def from_string(cls, text: str, namespaces: dict[str, str]) -> "ElementAdapter":
+        return cls(ET.fromstring(text=text), namespaces)
+
+    def __init__(self, elem: Element, namespaces: dict[str, str]):
+        self.elem = elem
+        self.namespaces = namespaces
+
+    def find(self, path: str) -> "ElementAdapter":
+        result = self.elem.find(path, self.namespaces)
+        if result is None:
+            raise DataNotFoundError(f"No element found for path {path}")
+        return ElementAdapter(result, self.namespaces)
+
+    @property
+    def text(self) -> str:
+        result = self.elem.text
+        if result is None:
+            raise DataNotFoundError(f"No text found for {self.elem.tag}")
+        return result
+
+    def get(self, key: str) -> str:
+        result = self.elem.get(key)
+        if result is None:
+            raise DataNotFoundError(f"No value for attribute {key} found for {self.elem.tag}")
+        return result
+
+    def findall(self, path: str) -> "list[ElementAdapter]":
+        return [
+            ElementAdapter(elem, self.namespaces)
+            for elem in self.elem.findall(path, self.namespaces)
+        ]
+
+    @property
+    def tag(self) -> str:
+        return self.elem.tag
+
+    def get_children(self) -> "list[ElementAdapter]":
+        return [ElementAdapter(elem, self.namespaces) for elem in self.elem[:]]

experiments/dorop-20/metadata/exceptions.py

@@ -0,0 +1,5 @@
+class MetadataFileNotFoundError(Exception):
+    pass
+
+class DataNotFoundError(Exception):
+    pass

experiments/dorop-20/metadata/models.py

@@ -0,0 +1,81 @@
+from pathlib import Path
+from dataclasses import dataclass
+from datetime import datetime
+from enum import Enum
+
+from pydantic import BaseModel
+
+class RecordStatus(Enum):
+    STAGE = "stage"
+    STORE = "store"
+
+@dataclass
+class Actor():
+    address: str
+    role: str
+
+@dataclass
+class PreservationEvent():
+    identifier: str
+    type: str
+    datetime: datetime
+    detail: str
+    actor: Actor
+
+class CommonMetadata(BaseModel):
+    title: str
+    author: str
+    publication_date: datetime
+    subjects: list[str]
+
+class AssetFileUse(Enum):
+    ACCESS = "ACCESS"
+    SOURCE = "SOURCE"
+
+class FileMetadataFileType(Enum):
+    TECHNICAL = "TECHNICAL"
+
+@dataclass
+class FileMetadataFile:
+    id: str
+    type: FileMetadataFileType
+    path: Path
+
+@dataclass
+class AssetFile:
+    id: str
+    use: AssetFileUse
+    path: Path
+    metadata_file: FileMetadataFile
+
+@dataclass
+class Asset:
+    id: str
+    events: list[PreservationEvent]
+    files: list[AssetFile]
+
+class StructMapType(Enum):
+    PHYSICAL = "PHYSICAL"
+    LOGICAL = "LOGICAL"
+
+@dataclass
+class StructMapItem():
+    order: int
+    label: str
+    asset_id: str
+
+@dataclass
+class StructMap():
+    id: str
+    type: StructMapType
+    items: list[StructMapItem]
+
+@dataclass
+class RepositoryItem():
+    id: str
+    record_status: RecordStatus
+    rights: str | None
+    events: list[PreservationEvent]
+    common_metadata: CommonMetadata
+    struct_map: StructMap
+    assets: list[Asset]