Separate Python packages (#50)

* start separating python packages * maybe CI * maybe docs * move import to geoarrow.c * with passing test * readme * fix a few references * maybe fix doctests * fix package name * fix install instructions * maybe fix doctest * move doctests to the end * maybe fix wheel name * maybe fix * actions * maybe fix doc * also ignore opt/ for coverage * maybe actually get doctests to run * maybe more portable doctest
geoarrow · Sep 7, 2023 · a50551d · a50551d
1 parent 73a5d09
commit a50551d
Show file tree

Hide file tree

Showing 35 changed files with 157 additions and 194 deletions.
diff --git a/.env b/.env
diff --git a/.github/workflows/python.yaml b/.github/workflows/python.yaml
@@ -32,30 +32,22 @@ jobs:
 
       - name: Install geoarrow
         run: |
-          pushd python
+          pushd python/geoarrow-c
           pip install .[test]
           popd
           pip list
 
       - name: Run tests
         run: |
-          pytest python/tests -v -s
-
-      - name: Run doctests
-        if: success() && matrix.python-version == '3.10'
-        run: |
-          pytest --pyargs geoarrow --doctest-modules
-          # No Cython docs yet
-          # pip install pytest-cython
-          # pytest --pyargs geoarrow --doctest-cython
+          pytest python/geoarrow-c/tests -v -s
 
       - name: Coverage
         if: success() && matrix.python-version == '3.10'
         run: |
           sudo apt-get install -y lcov
           pip uninstall --yes geoarrow
           pip install pytest-cov Cython
-          pushd python
+          pushd python/geoarrow-c
 
           # Build with Cython + gcc coverage options
           pip install -e .[test]
@@ -69,9 +61,10 @@ jobs:
           lcov \
             --capture --directory build \
             --exclude "/usr/*" \
+            --exclude "/opt/*" \
             --exclude "/Library/*" \
             --exclude "*/_lib.cpp" \
-            --exclude "*/src/geoarrow/geoarrow/*" \
+            --exclude "*/src/geoarrow/c/geoarrow/*" \
             --output-file=coverage.info
 
           lcov --list coverage.info
@@ -80,4 +73,17 @@ jobs:
         if: success() && matrix.python-version == '3.10'
         uses: codecov/codecov-action@v2
         with:
-          files: 'python/coverage.info,python/coverage.xml'
+          files: 'python/geoarrow-coverage.info,python/geoarrow-c/coverage.xml'
+
+      - name: Run doctests
+        if: success() && matrix.python-version == '3.10'
+        run: |
+          # Because of namespace packaging we have to add this here and
+          # rebuild to avoid confusig pytest
+          touch python/geoarrow-c/src/geoarrow/__init__.py
+          pip install python/geoarrow-c
+
+          pytest --pyargs geoarrow.c --doctest-modules
+          # No Cython docs yet
+          # pip install pytest-cython
+          # pytest --pyargs geoarrow --doctest-cython
diff --git a/README.md b/README.md
@@ -18,7 +18,7 @@ release of the geoarrow specification.
 ## Get started in Python
 
 ```python
-import geoarrow.pyarrow as ga
+import geoarrow.c.pyarrow as ga
 
 ga.point()
 # PointType(geoarrow.point)

diff --git a/ci/scripts/build-docs.sh b/ci/scripts/build-docs.sh
@@ -51,10 +51,10 @@ main() {
    # pip install . doesn't quite work with the setuptools available on the
    # ubuntu docker image...python -m build works I think because it sets up
    # a virtualenv
-   pushd python
+   pushd python/geoarrow-c
    rm -rf dist
    python3 -m build --wheel
-   pip3 install dist/geoarrow-*.whl
+   pip3 install dist/geoarrow*.whl
    popd
 
    pushd docs

diff --git a/docker-compose.yml b/docker-compose.yml
@@ -4,14 +4,7 @@ version: '3.5'
 services:
 
   docs:
-    image: ${REPO}:ubuntu-${GEOARROW_ARCH}
-    build:
-      context: .
-      cache_from:
-        - ${REPO}:ubuntu-${GEOARROW_ARCH}
-      dockerfile: ci/docker/ubuntu.dockerfile
-      args:
-        GEOARROW_ARCH: ${GEOARROW_ARCH}
+    image: ghcr.io/apache/arrow-nanoarrow:ubuntu
     volumes:
       - .:/geoarrow-c
     command: "/bin/bash /geoarrow-c/ci/scripts/build-docs.sh"
diff --git a/docs/source/conf.py b/docs/source/conf.py
@@ -14,7 +14,7 @@
 import sys
 import datetime
 
-import geoarrow
+import geoarrow.c
 
 sys.path.insert(0, os.path.abspath(".."))
 

diff --git a/docs/source/python/geoarrow.rst b/docs/source/python/geoarrow.rst
@@ -2,7 +2,7 @@
 Core API
 ========
 
-.. automodule:: geoarrow
+.. automodule:: geoarrow.c
    :members:
 
 Constants

diff --git a/docs/source/python/pyarrow.rst b/docs/source/python/pyarrow.rst
@@ -2,7 +2,7 @@
 Integration with pyarrow
 ========================
 
-.. automodule:: geoarrow.pyarrow
+.. automodule:: geoarrow.c.pyarrow
 
     Array constructors
     ------------------
@@ -100,8 +100,8 @@ Integration with pyarrow
     .. autoclass:: MultiPolygonType
         :members:
 
-    .. autoclass:: geoarrow.pyarrow._dataset.GeoDataset
+    .. autoclass:: geoarrow.c.pyarrow._dataset.GeoDataset
         :members:
 
-    .. autoclass:: geoarrow.pyarrow._dataset.ParquetRowGroupGeoDataset
+    .. autoclass:: geoarrow.c.pyarrow._dataset.ParquetRowGroupGeoDataset
         :members:
diff --git a/python/.coveragerc → python/geoarrow-c/.coveragerc b/python/.coveragerc → python/geoarrow-c/.coveragerc
diff --git a/python/.gitignore → python/geoarrow-c/.gitignore b/python/.gitignore → python/geoarrow-c/.gitignore
@@ -16,8 +16,8 @@
 # specific language governing permissions and limitations
 # under the License.
 
-src/geoarrow/geoarrow
-src/geoarrow/_lib.cpp
+src/geoarrow/c/geoarrow
+src/geoarrow/c/_lib.cpp
 
 # Byte-compiled / optimized / DLL files
 __pycache__/

diff --git a/python/MANIFEST.in → python/geoarrow-c/MANIFEST.in b/python/MANIFEST.in → python/geoarrow-c/MANIFEST.in
@@ -16,5 +16,6 @@
 # under the License.
 
 exclude bootstrap.py
-include src/geoarrow/geoarrow/*.h
-include src/geoarrow/geoarrow/*.hpp
+include src/geoarrow/c/**/**/*.h
+include src/geoarrow/c/**/*.h
+include src/geoarrow/c/**/*.hpp
diff --git a/python/README.ipynb → python/geoarrow-c/README.ipynb b/python/README.ipynb → python/geoarrow-c/README.ipynb
@@ -14,7 +14,7 @@
     "Python bindings for nanoarrow are not yet available on PyPI. You can install via URL (requires a C++ compiler):\n",
     "\n",
     "```bash\n",
-    "python -m pip install \"https://github.com/geoarrow/geoarrow-cpp/archive/refs/heads/main.zip#egg=geoarrow&subdirectory=python\"\n",
+    "python -m pip install \"https://github.com/geoarrow/geoarrow-c/archive/refs/heads/main.zip#egg=geoarrow-c&subdirectory=python/geoarrow-c\"\n",
     "```\n",
     "\n",
     "If you can import the namespace, you're good to go! The only reasonable interface to geoarrow currently depends on `pyarrow`, which you can import with:"
@@ -26,7 +26,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "import geoarrow.pyarrow as ga"
+    "import geoarrow.c.pyarrow as ga"
    ]
   },
   {
@@ -48,7 +48,7 @@
      "data": {
       "text/plain": [
        "PointArray:PointType(geoarrow.point)[1]\n",
-       "<POINT (0 1)>\n"
+       "<POINT (0 1)>"
       ]
      },
      "execution_count": 2,
@@ -88,7 +88,7 @@
        "PointArray:PointType(geoarrow.point)[3]\n",
        "<POINT (1 3)>\n",
        "<POINT (2 4)>\n",
-       "<POINT (3 5)>\n"
+       "<POINT (3 5)>"
       ]
      },
      "execution_count": 3,
@@ -117,7 +117,7 @@
        "PointArray:PointType(interleaved geoarrow.point)[3]\n",
        "<POINT (1 2)>\n",
        "<POINT (3 4)>\n",
-       "<POINT (5 6)>\n"
+       "<POINT (5 6)>"
       ]
      },
      "execution_count": 4,
@@ -137,7 +137,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Importing `geoarrow.pyarrow` will register the geoarrow extension types with pyarrow such that you can read/write Arrow streams, Arrow files, and Parquet that contains Geoarrow extension types. A number of these files are available from the [geoarrow-data](https://github.com/geoarrow/geoarrow-data) repository."
+    "Importing `geoarrow.c.pyarrow` will register the geoarrow extension types with pyarrow such that you can read/write Arrow streams, Arrow files, and Parquet that contains Geoarrow extension types. A number of these files are available from the [geoarrow-data](https://github.com/geoarrow/geoarrow-data) repository."
    ]
   },
   {
@@ -189,22 +189,6 @@
    "execution_count": 6,
    "metadata": {},
    "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site-packages/geopandas/_compat.py:124: UserWarning: The Shapely GEOS version (3.11.1-CAPI-1.17.1) is incompatible with the GEOS version PyGEOS was compiled with (3.10.1-CAPI-1.16.0). Conversions between both will be slow.\n",
-      "  warnings.warn(\n",
-      "/var/folders/gt/l87wjg8s7312zs9s7c1fgs900000gn/T/ipykernel_81348/2107898165.py:1: DeprecationWarning: Shapely 2.0 is installed, but because PyGEOS is also installed, GeoPandas still uses PyGEOS by default. However, starting with version 0.14, the default will switch to Shapely. To force to use Shapely 2.0 now, you can either uninstall PyGEOS or set the environment variable USE_PYGEOS=0. You can do this before starting the Python process, or in your code before importing geopandas:\n",
-      "\n",
-      "import os\n",
-      "os.environ['USE_PYGEOS'] = '0'\n",
-      "import geopandas\n",
-      "\n",
-      "In the next release, GeoPandas will switch to using Shapely by default, even if PyGEOS is installed. If you only have PyGEOS installed to get speed-ups, this switch should be smooth. However, if you are using PyGEOS directly (calling PyGEOS functions on geometries from GeoPandas), this will then stop working and you are encouraged to migrate from PyGEOS to Shapely 2.0 (https://shapely.readthedocs.io/en/latest/migration_pygeos.html).\n",
-      "  import geopandas\n"
-     ]
-    },
     {
      "data": {
       "text/plain": [
@@ -216,9 +200,9 @@
        "<MULTILINESTRING ((673606.0199999996 5162961.9823, 673606.01999999...>\n",
        "...245 values...\n",
        "<MULTILINESTRING ((681672.6200000001 5078601.5823, 681866.01999999...>\n",
-       "<MULTILINESTRING ((414867.9170000004 5093040.8807, 414793.81699999...>\n",
-       "<MULTILINESTRING ((414867.9170000004 5093040.8807, 414829.71700000...>\n",
-       "<MULTILINESTRING ((414867.9170000004 5093040.8807, 414937.21700000...>\n",
+       "<MULTILINESTRING ((414867.91700000037 5093040.8807, 414793.8169999...>\n",
+       "<MULTILINESTRING ((414867.91700000037 5093040.8807, 414829.7170000...>\n",
+       "<MULTILINESTRING ((414867.91700000037 5093040.8807, 414937.2170000...>\n",
        "<MULTILINESTRING ((648686.0197000001 5099181.984099999, 648866.019...>"
       ]
      },
@@ -381,9 +365,9 @@
        "<MULTILINESTRING ((673606.0199999996 5162961.9823, 673606.01999999...>\n",
        "...245 values...\n",
        "<MULTILINESTRING ((681672.6200000001 5078601.5823, 681866.01999999...>\n",
-       "<MULTILINESTRING ((414867.9170000004 5093040.8807, 414793.81699999...>\n",
-       "<MULTILINESTRING ((414867.9170000004 5093040.8807, 414829.71700000...>\n",
-       "<MULTILINESTRING ((414867.9170000004 5093040.8807, 414937.21700000...>\n",
+       "<MULTILINESTRING ((414867.91700000037 5093040.8807, 414793.8169999...>\n",
+       "<MULTILINESTRING ((414867.91700000037 5093040.8807, 414829.7170000...>\n",
+       "<MULTILINESTRING ((414867.91700000037 5093040.8807, 414937.2170000...>\n",
        "<MULTILINESTRING ((648686.0197000001 5099181.984099999, 648866.019...>"
       ]
      },
@@ -444,7 +428,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.6"
+   "version": "3.11.2"
   },
   "orig_nbformat": 4
  },

diff --git a/python/README.md → python/geoarrow-c/README.md b/python/README.md → python/geoarrow-c/README.md
@@ -1,20 +1,20 @@
 # geoarrow for Python
 
-The geoarrow Python package provides bindings to the geoarrow-c implementation of the [GeoArrow specification](https://github.com/geoarrow/geoarrow). The geoarrow Python bindings provide input/output to/from Arrow-friendly formats (e.g., Parquet, Arrow Stream, Arrow File) and general-purpose coordinate shuffling tools among GeoArrow, WKT, and WKB encodings. 
+The geoarrow Python package provides bindings to the geoarrow-c implementation of the [GeoArrow specification](https://github.com/geoarrow/geoarrow). The geoarrow Python bindings provide input/output to/from Arrow-friendly formats (e.g., Parquet, Arrow Stream, Arrow File) and general-purpose coordinate shuffling tools among GeoArrow, WKT, and WKB encodings.
 
 ## Installation
 
 Python bindings for nanoarrow are not yet available on PyPI. You can install via URL (requires a C++ compiler):
 
 ```bash
-python -m pip install "https://github.com/geoarrow/geoarrow-cpp/archive/refs/heads/main.zip#egg=geoarrow&subdirectory=python"
+python -m pip install "https://github.com/geoarrow/geoarrow-c/archive/refs/heads/main.zip#egg=geoarrow-c&subdirectory=python/geoarrow-c"
 ```
 
 If you can import the namespace, you're good to go! The only reasonable interface to geoarrow currently depends on `pyarrow`, which you can import with:
 
 
 ```python
-import geoarrow.pyarrow as ga
+import geoarrow.c.pyarrow as ga
 ```
 
 ## Examples
@@ -34,7 +34,6 @@ ga.as_geoarrow(["POINT (0 1)"])
 
 
 
-
 This will work with:
 
 - An existing array created by geoarrow
@@ -51,7 +50,7 @@ Alternatively, you can construct GeoArrow arrays directly from a series of buffe
 import numpy as np
 
 ga.point().from_geobuffers(
-    None, 
+    None,
     np.array([1.0, 2.0, 3.0]),
     np.array([3.0, 4.0, 5.0])
 )
@@ -68,7 +67,6 @@ ga.point().from_geobuffers(
 
 
 
-
 ```python
 ga.point().with_coord_type(ga.CoordType.INTERLEAVED).from_geobuffers(
     None,
@@ -86,8 +84,7 @@ ga.point().with_coord_type(ga.CoordType.INTERLEAVED).from_geobuffers(
 
 
 
-
-Importing `geoarrow.pyarrow` will register the geoarrow extension types with pyarrow such that you can read/write Arrow streams, Arrow files, and Parquet that contains Geoarrow extension types. A number of these files are available from the [geoarrow-data](https://github.com/geoarrow/geoarrow-data) repository.
+Importing `geoarrow.c.pyarrow` will register the geoarrow extension types with pyarrow such that you can read/write Arrow streams, Arrow files, and Parquet that contains Geoarrow extension types. A number of these files are available from the [geoarrow-data](https://github.com/geoarrow/geoarrow-data) repository.
 
 
 ```python
@@ -135,18 +132,6 @@ array = ga.as_geoarrow(df.geometry)
 array
 ```
 
-    /Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site-packages/geopandas/_compat.py:124: UserWarning: The Shapely GEOS version (3.11.1-CAPI-1.17.1) is incompatible with the GEOS version PyGEOS was compiled with (3.10.1-CAPI-1.16.0). Conversions between both will be slow.
-      warnings.warn(
-    /var/folders/gt/l87wjg8s7312zs9s7c1fgs900000gn/T/ipykernel_81348/2107898165.py:1: DeprecationWarning: Shapely 2.0 is installed, but because PyGEOS is also installed, GeoPandas still uses PyGEOS by default. However, starting with version 0.14, the default will switch to Shapely. To force to use Shapely 2.0 now, you can either uninstall PyGEOS or set the environment variable USE_PYGEOS=0. You can do this before starting the Python process, or in your code before importing geopandas:
-
-    import os
-    os.environ['USE_PYGEOS'] = '0'
-    import geopandas
-
-    In the next release, GeoPandas will switch to using Shapely by default, even if PyGEOS is installed. If you only have PyGEOS installed to get speed-ups, this switch should be smooth. However, if you are using PyGEOS directly (calling PyGEOS functions on geometries from GeoPandas), this will then stop working and you are encouraged to migrate from PyGEOS to Shapely 2.0 (https://shapely.readthedocs.io/en/latest/migration_pygeos.html).
-      import geopandas
-
-
 
 
 
@@ -158,9 +143,9 @@ array
     <MULTILINESTRING ((673606.0199999996 5162961.9823, 673606.01999999...>
     ...245 values...
     <MULTILINESTRING ((681672.6200000001 5078601.5823, 681866.01999999...>
-    <MULTILINESTRING ((414867.9170000004 5093040.8807, 414793.81699999...>
-    <MULTILINESTRING ((414867.9170000004 5093040.8807, 414829.71700000...>
-    <MULTILINESTRING ((414867.9170000004 5093040.8807, 414937.21700000...>
+    <MULTILINESTRING ((414867.91700000037 5093040.8807, 414793.8169999...>
+    <MULTILINESTRING ((414867.91700000037 5093040.8807, 414829.7170000...>
+    <MULTILINESTRING ((414867.91700000037 5093040.8807, 414937.2170000...>
     <MULTILINESTRING ((648686.0197000001 5099181.984099999, 648866.019...>
 
 
@@ -180,7 +165,7 @@ geopandas.GeoSeries.from_wkb(ga.as_wkb(array))
     2      MULTILINESTRING ((631355.519 5122892.285, 6313...
     3      MULTILINESTRING ((665166.020 5138641.982, 6651...
     4      MULTILINESTRING ((673606.020 5162961.982, 6736...
-                                 ...                        
+                                 ...
     250    MULTILINESTRING ((681672.620 5078601.582, 6818...
     251    MULTILINESTRING ((414867.917 5093040.881, 4147...
     252    MULTILINESTRING ((414867.917 5093040.881, 4148...
@@ -280,9 +265,9 @@ geoarrow_array2
     <MULTILINESTRING ((673606.0199999996 5162961.9823, 673606.01999999...>
     ...245 values...
     <MULTILINESTRING ((681672.6200000001 5078601.5823, 681866.01999999...>
-    <MULTILINESTRING ((414867.9170000004 5093040.8807, 414793.81699999...>
-    <MULTILINESTRING ((414867.9170000004 5093040.8807, 414829.71700000...>
-    <MULTILINESTRING ((414867.9170000004 5093040.8807, 414937.21700000...>
+    <MULTILINESTRING ((414867.91700000037 5093040.8807, 414793.8169999...>
+    <MULTILINESTRING ((414867.91700000037 5093040.8807, 414829.7170000...>
+    <MULTILINESTRING ((414867.91700000037 5093040.8807, 414937.2170000...>
     <MULTILINESTRING ((648686.0197000001 5099181.984099999, 648866.019...>