Merge branch 'release/0.3.0'

.github/workflows/ci.yml

@@ -17,8 +17,10 @@ env:
 
 jobs:
   lint:
-    name: Linting
+    name: Lint
     runs-on: ubuntu-latest
+    permissions:
+      contents: read
     steps:
       -
         uses: actions/checkout@v3
@@ -41,8 +43,10 @@ jobs:
         uses: pre-commit/[email protected]
 
   test:
-    name: Python ${{ matrix.python-version }}
+    name: Test (Python ${{ matrix.python-version }})
     runs-on: ubuntu-latest
+    permissions:
+      contents: read
     needs:
       - lint
     strategy:
@@ -89,10 +93,12 @@ jobs:
           docker compose --file ${GITHUB_WORKSPACE}/.docker/docker-compose.ci-test.yml down
 
   build:
-    name: Build project
+    name: Build
     runs-on: ubuntu-latest
+    permissions:
+      contents: read
     needs:
-      - test
+      - lint
     steps:
       -
         uses: actions/checkout@v3
@@ -118,14 +124,51 @@ jobs:
             if-no-files-found: error
             retention-days: 7
 
-  publish:
-    name: Publish project
+  create-release:
+    name: Release
+    runs-on: ubuntu-latest
+    if: startsWith(github.ref, 'refs/tags/')
+    permissions:
+      contents: write
+    needs:
+      - build
+      - test
+    steps:
+      -
+        uses: actions/checkout@v3
+      -
+        uses: actions/download-artifact@v3
+        with:
+          name: artifacts
+          path: dist
+      -
+        name: Get latest release info
+        id: query-release-info
+        uses: release-flow/keep-a-changelog-action@v2
+        with:
+          command: query
+          version: ${{ github.ref_name }}
+      -
+        name: Display release info
+        run: |
+          echo "$Version: ${{ steps.query-release-info.outputs.version }}"
+          echo "$Date: ${{ steps.query-release-info.outputs.release-date }}"
+          echo "${{ steps.query-release-info.outputs.release-notes }}"
+      -
+        uses: ncipollo/release-action@v1
+        with:
+          artifacts: "dist/*.tar.gz,dist/*.whl"
+          body: ${{ steps.query-release-info.outputs.release-notes }}
+
+  pypi-publish:
+    name: Publish
     runs-on: ubuntu-latest
     if: startsWith(github.ref, 'refs/tags/')
     permissions:
       id-token: write  # IMPORTANT: this permission is mandatory for trusted publishing
     needs:
       - build
+      - test
     steps:
       -
         uses: actions/download-artifact@v3
@@ -134,4 +177,4 @@ jobs:
           path: dist
       -
         name: Publish build to PyPI
-        uses: pypa/[email protected].6
+        uses: pypa/[email protected].7

.pre-commit-config.yaml

@@ -27,7 +27,7 @@ repos:
       - id: check-case-conflict
       - id: detect-private-key
   - repo: https://github.com/pre-commit/mirrors-prettier
-    rev: 'v2.7.1'
+    rev: 'v3.0.0'
     hooks:
       - id: prettier
         types_or:
@@ -37,10 +37,10 @@ repos:
         exclude: "(^Pipfile\\.lock$)"
   # Python hooks
   - repo: https://github.com/charliermarsh/ruff-pre-commit
-    rev: 'v0.0.265'
+    rev: 'v0.0.278'
     hooks:
       - id: ruff
   - repo: https://github.com/psf/black
-    rev: 23.3.0
+    rev: 23.7.0
     hooks:
       - id: black

CHANGELOG.md

@@ -5,6 +5,18 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.3.0] - 2023-07-19
+
+### Added
+
+- Restricted action permissions to minimal requirements to function
+- Github CI also now creates a Github release with sdist, wheel and changelog
+- Additional classifiers to the project on PyPI
+
+### Fixed
+
+- Handling of ISO-8061 dates with fractional seconds, which Python doesn't support natively
+
 ## [0.2.0] - 2023-06-26
 
 ### Fixed

TODO.md

@@ -1,5 +1,5 @@
 # TODO
 
-- Create a mixin for some fields which seem often paired together, such as `dc:created` and `dc:modified`
 - Testing with sample email documents
 - Testing with the sample image documents
+- Add Github pages with more detailed documentation

pyproject.toml

@@ -15,15 +15,22 @@ authors = [
 ]
 classifiers = [
   "Development Status :: 4 - Beta",
+  "License :: OSI Approved :: GNU General Public License v3 (GPLv3)",
+  "Operating System :: OS Independent",
+  "Intended Audience :: Developers",
+  "Environment :: Web Environment",
   "Programming Language :: Python",
+  "Programming Language :: Python :: 3",
+  "Programming Language :: Python :: 3 :: Only",
   "Programming Language :: Python :: 3.8",
   "Programming Language :: Python :: 3.9",
   "Programming Language :: Python :: 3.10",
   "Programming Language :: Python :: 3.11",
+  "Programming Language :: Python :: 3.12",
   "Programming Language :: Python :: Implementation :: CPython",
-  "Programming Language :: Python :: Implementation :: PyPy",
+  "Programming Language :: Python :: Implementation :: PyPy"
 ]
-dependencies = ["httpx"]
+dependencies = ["httpx ~= 0.24"]
 
 [project.urls]
 Documentation = "https://github.com/stumpylog/tika-rest-client#readme"

src/tika_client/__about__.py

@@ -1 +1 @@
-__version__ = "0.2.0"
+__version__ = "0.3.0"

src/tika_client/data_models.py

@@ -1,4 +1,7 @@
+import logging
+import re
 from datetime import datetime
+from datetime import timedelta
 from enum import Enum
 from typing import Dict
 from typing import List
@@ -7,6 +10,9 @@
 
 # Based on https://cwiki.apache.org/confluence/display/TIKA/Metadata+Overview
 
+logger = logging.getLogger("tika-client.data")
+_FRACTION_REGEX = re.compile("(.*)([\\.,][0-9]+)(.*)")
+
 
 class TikaKey(str, Enum):
     Parsers = "X-TIKA:Parsed-By"
@@ -96,8 +102,34 @@ def get_optional_datetime(self, key: Union[TikaKey, DublinCoreKey, XmpKey, str])
         """
         if key not in self.data:  # pragma: no cover
             return None
+
+        date_str: str = self.data[key]
+
+        # Handle fractional seconds
+        frac = _FRACTION_REGEX.match(date_str)
+        if frac is not None:
+            logger.info("Located fractional seconds")
+            delta = timedelta(seconds=float(frac.group(2)))
+            date_str = frac.group(1)
+            # Attempt to include the timezone info still
+            if frac.group(3) is not None:
+                date_str += frac.group(3)
+        else:
+            delta = timedelta()
+
         # Handle Zulu time as UTC
-        return datetime.fromisoformat(self.data[key].replace("Z", "+00:00"))
+        if "Z" in date_str:
+            date_str = date_str.replace("Z", "+00:00")
+
+        # Assume UTC if it is not set
+        if "+" not in date_str:
+            date_str += "+00:00"
+
+        try:
+            return datetime.fromisoformat(date_str) + delta
+        except ValueError as e:
+            logger.error(f"{e} during datetime parsing")
+        return None
 
     def get_optional_string(self, key: Union[TikaKey, DublinCoreKey, XmpKey, str]) -> Optional[str]:
         if key not in self.data:

tests/samples/README.md

@@ -1,2 +1,5 @@
 sample.jpg - https://unsplash.com/photos/8OyKWQgBsKQ
 sample.png - https://unsplash.com/photos/iar-afB0QQw
+
+microsoft-sample.docx - produced by Microsoft Office
+sample-libre-office.odt - produced by LibreOffice Writer 7.5.12

tests/test_file_formats.py

@@ -0,0 +1,31 @@
+from datetime import datetime
+from datetime import timezone
+
+import magic
+
+from tests.conftest import SAMPLE_DIR
+from tika_client.client import TikaClient
+
+
+class TestLibreOfficeFormats:
+    def test_parse_libre_office_writer_document(self, tika_client: TikaClient):
+        test_file = SAMPLE_DIR / "sample-libre-office.odt"
+        resp = tika_client.tika.as_html.from_file(test_file, magic.from_file(str(test_file), mime=True))
+
+        assert resp.type == "application/vnd.oasis.opendocument.text"
+        assert (
+            "<body><p>This is a document created by LibreOffice Writer 7.5.12, on July 19th, 2023</p>\n</body>"
+            in resp.content
+        )
+        assert resp.content_length == 11149
+        assert resp.created is not None
+        assert resp.created == datetime(
+            year=2023,
+            month=7,
+            day=19,
+            hour=11,
+            minute=30,
+            second=44,
+            microsecond=719000,
+            tzinfo=timezone.utc,
+        )