feat(clean): add updated version of rapidfuzz and python-crfsuite

add submodule of Levenshtein and python-crfsuite try to compile Levenshtein with setup.py failed poetry build successfully connect poetry and cmake build delete the python-Levenshtein package add cython compiler fix(clean) : add precompiling code for python-crfsuite Polish the build.py Connect the clean util function with the compiled file Add clean_build pipeline file Fix pip install whl Update clean_build.yml Update clean_build.yml Update clean_build Update clean_build.yml feat(clean) : update build.py feat(clean) : update build.py feat(clean) : update build.py feat(clean) : update build.py feat(clean) : update build.py feat(clean) : update build.py feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml feat(clean) : update clean_build.yml to release.yml add all workflows
sfu-db · Jul 7, 2022 · 59f3506 · 59f3506
1 parent 7b2ce3f
commit 59f3506
Show file tree

Hide file tree

Showing 10 changed files with 1,488 additions and 882 deletions.
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
@@ -7,7 +7,7 @@ on:
       - release
   pull_request:
     branches:
-      - develop
+       - develop
 
 jobs:
   build:
@@ -20,6 +20,8 @@ jobs:
     runs-on: ${{ matrix.os }}
     steps:
       - uses: actions/checkout@v2
+        with:
+          submodules: recursive
 
       - uses: actions/setup-python@v2
         with:
@@ -38,6 +40,9 @@ jobs:
           poetry install
           poetry config --list
 
+      - name: Build binary dependencies
+        run: poetry build
+
       - name: Print tool versions
         run: |
           poetry run pylint --version 

diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml
@@ -6,43 +6,45 @@ on:
       - release
 
 jobs:
-  # docs-build:
-  #   runs-on: ubuntu-latest
-  #   steps:
-  #     - uses: actions/checkout@v2
-
-  #     - name: Install dependencies
-  #       run: |
-  #         pip install poetry
-  #         curl -L https://github.com/jgm/pandoc/releases/download/2.11.2/pandoc-2.11.2-1-amd64.deb -o /tmp/pandoc.deb && sudo dpkg -i /tmp/pandoc.deb
-
-  #     - name: Cache venv
-  #       uses: actions/cache@v2
-  #       with:
-  #         path: ~/.cache/pypoetry/virtualenvs
-  #         key: ${{ runner.os }}-build-${{ matrix.python }}-${{ secrets.CACHE_VERSION }}-${{ hashFiles('poetry.lock') }}
-
-  #     - name: Install dependencies
-  #       run: |
-  #         pip install poetry
-  #         poetry install
-
-  #     - name: Build docs
-  #       run: poetry run sphinx-build -M html docs/source docs/build
-
-  #     - name: Archive docs
-  #       uses: actions/upload-artifact@v2
-  #       with:
-  #         name: docs
-  #         path: docs/build/html
-
-  build:
-    runs-on: ubuntu-latest
-    # needs: docs-build
+#   docs-build:
+#     runs-on: ubuntu-latest
+#     steps:
+#       - uses: actions/checkout@v2
+#       - name: Install dependencies
+#         run: |
+#           pip install poetry
+#           curl -L https://github.com/jgm/pandoc/releases/download/2.11.2/pandoc-2.11.2-1-amd64.deb -o /tmp/pandoc.deb && sudo dpkg -i /tmp/pandoc.deb
+
+#       - name: Cache venv
+#         uses: actions/cache@v2
+#         with:
+#           path: ~/.cache/pypoetry/virtualenvs
+#           key: ${{ runner.os }}-build-${{ matrix.python }}-${{ secrets.CACHE_VERSION }}-${{ hashFiles('poetry.lock') }}
+
+#       - name: Install dependencies
+#         run: |
+#           pip install poetry
+#           poetry install
+
+#       - name: Build docs
+#         run: poetry run sphinx-build -M html docs/source docs/build
+
+#       - name: Archive docs
+#         uses: actions/upload-artifact@v2
+#         with:
+#           name: docs
+#           path: docs/build/html
+  build-wheels-linux-macos:
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+          os: [ubuntu-latest, macos-latest]
+          python: ["3.8", "3.9"]
     steps:
       - uses: actions/checkout@v2
         with:
-          fetch-depth: "0"
+          submodules: recursive
 
       - uses: actions/setup-python@v1
         with:
@@ -61,15 +63,148 @@ jobs:
           poetry install
           poetry config --list
 
-      - name: Print tool versions
+      - name: Build wheels
+        run: poetry build
+
+      - name: Upload wheels
+        uses: actions/upload-artifact@v2
+        with:
+          name: "${{ matrix.os }}-${{ matrix.python}}"
+          path: dist/*.whl
+
+  build-wheels-windows:
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+          os: [windows-latest]
+          python: ["3.8", "3.9"]
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          submodules: recursive
+
+      - uses: actions/setup-python@v1
+        with:
+          python-version: ${{ matrix.python }}
+
+      - name: Cache venv
+        uses: actions/cache@v2
+        with:
+          path: ~/.cache/pypoetry/virtualenvs
+          key: ${{ runner.os }}-build-${{ matrix.python }}-${{ secrets.CACHE_VERSION }}-${{ hashFiles('poetry.lock') }}
+
+      - name: Install dependencies
         run: |
-          poetry run pylint --version 
-          poetry run pytest --version 
-          poetry run black --version
-          poetry run pyright --version
+          echo "Cache Version ${{ secrets.CACHE_VERSION }}"
+          pip install poetry toml-cli
+          poetry install
+          poetry config --list
 
       - name: Build wheels
         run: poetry build
+
+      - name: Upload wheels
+        uses: actions/upload-artifact@v2
+        with:
+          name: "${{ matrix.os }}-${{ matrix.python}}"
+          path: dist/*.whl
+
+  build-sdist:
+    name: Build source distribution
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        python: ["3.9"]
+        os: [ubuntu-latest]
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          submodules: recursive
+
+      - uses: actions/setup-python@v1
+        with:
+          python-version: '3.9'
+
+      - name: Cache venv
+        uses: actions/cache@v2
+        with:
+          path: ~/.cache/pypoetry/virtualenvs
+          key: ${{ runner.os }}-build-${{ matrix.python }}-${{ secrets.CACHE_VERSION }}-${{ hashFiles('poetry.lock') }}
+
+      - name: Install dependencies
+        run: |
+          echo "Cache Version ${{ secrets.CACHE_VERSION }}"
+          pip install poetry toml-cli
+          poetry install
+          poetry config --list
+
+      - name: Build wheels
+        run: poetry build
+
+      - name: Upload sdist
+        uses: actions/upload-artifact@v2
+        with:
+          path: dist/*.tar.gz
+
+  verify:
+    runs-on: ${{ matrix.os }}
+    needs: [build-wheels-linux-macos, build-wheels-windows, build-sdist]
+    strategy:
+      matrix:
+        python: ["3.8", "3.9"]
+        os: [macos-latest, ubuntu-latest, windows-latest]
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          submodules: recursive
+
+      - uses: actions/setup-python@v1
+        with:
+          python-version: ${{ matrix.python }}
+
+      - uses: actions/download-artifact@v3
+        with:
+          name: "${{ matrix.os }}-${{ matrix.python}}"
+
+      - run: |
+          pip install *.whl
+          python -c "import dataprep"
+
+  upload:
+    runs-on: ${{ matrix.os }}
+    # needs: docs-build
+    needs: [verify]
+    strategy:
+      matrix:
+        python: ["3.9"]
+        os: [ubuntu-latest]
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          submodules: recursive
+
+      - uses: actions/setup-python@v1
+        with:
+          python-version: ${{ matrix.python }}
+
+      - name: Cache venv
+        uses: actions/cache@v2
+        with:
+          path: ~/.cache/pypoetry/virtualenvs
+          key: ${{ runner.os }}-build-${{ matrix.python }}-${{ secrets.CACHE_VERSION }}-${{ hashFiles('poetry.lock') }}
+
+      - name: Install dependencies
+        run: |
+          echo "Cache Version ${{ secrets.CACHE_VERSION }}"
+          pip install poetry toml-cli
+          poetry install
+          poetry config --list
+          
+      - name: Download all artifacts
+        uses: actions/download-artifact@v3
+        with:
+          path: dist
 
       - name: Parse version from pyproject.toml
         run: echo "DATAPREP_VERSION=`toml get --toml-path pyproject.toml tool.poetry.version`" >> $GITHUB_ENV
@@ -79,7 +214,7 @@ jobs:
 
       - uses: ncipollo/release-action@v1
         with:
-          artifacts: "dist/*.whl,dist/*.tar.gz"
+          artifacts: "dist/**/*.whl,dist/**/*.tar.gz"
           bodyFile: "RELEASE.md"
           token: ${{ secrets.GITHUB_TOKEN }}
           draft: true
@@ -88,25 +223,25 @@ jobs:
 
       - name: Upload wheels
         run: poetry publish --username __token__ --password ${{ secrets.PYPI_TOKEN }}
-
-  # docs-deploy:
-  #   runs-on: ubuntu-latest
-  #   needs: build
-  #   if: ${{ github.event.ref == 'refs/heads/release' }}
-  #   steps:
-  #     - uses: actions/checkout@v2
-
-  #     - name: Download docs
-  #       uses: actions/download-artifact@v2
-  #       with:
-  #         name: docs
-  #         path: docs/build/html
-
-  #     - run: echo 'docs.dataprep.ai' > docs/build/html/CNAME
-
-  #     - name: Deploy 🚀
-  #       uses: JamesIves/[email protected]
-  #       with:
-  #         branch: gh-pages # The branch the action should deploy to.
-  #         folder: docs/build/html # The folder the action should deploy.
-  #         clean-exclude: dev
+    
+#  docs-deploy:
+#    runs-on: ubuntu-latest
+#    needs: build
+#    if: ${{ github.event.ref == 'refs/heads/release' }}
+#    steps:
+#      - uses: actions/checkout@v2
+
+#      - name: Download docs
+#        uses: actions/download-artifact@v2
+#        with:
+#          name: docs
+#          path: docs/build/html
+
+      # - run: echo 'docs.dataprep.ai' > docs/build/html/CNAME
+
+      # - name: Deploy 🚀
+      #   uses: JamesIves/[email protected]
+      #   with:
+      #     branch: gh-pages # The branch the action should deploy to.
+      #     folder: docs/build/html # The folder the action should deploy.
+      #     clean-exclude: dev
diff --git a/.gitmodules b/.gitmodules
@@ -0,0 +1 @@
+
diff --git a/dataprep/clean/__init__.py b/dataprep/clean/__init__.py
@@ -2,6 +2,7 @@
 dataprep.clean
 ==============
 """
+
 from .clean_lat_long import clean_lat_long, validate_lat_long
 
 from .clean_email import clean_email, validate_email

diff --git a/dataprep/clean/address_utils.py b/dataprep/clean/address_utils.py
@@ -7,14 +7,12 @@
 from builtins import str
 from typing import Any, Dict, List, Optional, Tuple
 from collections import OrderedDict
-
-import os
 import string
 import re
+import os
 import warnings
 import pycrfsuite
 
-
 TAG_MAPPING = {
     "OccupancyType": "apartment",
     "OccupancyIdentifier": "apartment",

diff --git a/dataprep/clean/clean_duplication_utils.py b/dataprep/clean/clean_duplication_utils.py
@@ -11,16 +11,16 @@
 from itertools import permutations
 from os import path
 from tempfile import mkdtemp
-
 import pandas as pd
 import dask.dataframe as dd
 import dask
 from IPython.display import Javascript, display
-from Levenshtein import distance
 from metaphone import doublemetaphone
+from rapidfuzz.distance.Levenshtein import distance as LevenshteinDistance
 
 from .utils import to_dask
 
+
 DECODE_FUNC = """
     function b64DecodeUnicode(str) {
         // Going backwards: from bytestream, to percent-encoding, to original string.
@@ -141,7 +141,7 @@ def _get_nearest_neighbour_clusters(
                     continue
 
                 cluster_map[center].add(center)
-                dist = distance(center, val)
+                dist = LevenshteinDistance(center, val)
                 if dist <= radius or radius < 0:
                     cluster_map[center].add(val)
 

diff --git a/dataprep/clean/gui/clean_gui.py b/dataprep/clean/gui/clean_gui.py
@@ -229,7 +229,7 @@
     "clean_json",
     "clean_address",
     "clean_df",
-    # "clean_duplication",
+    "clean_duplication",
     "clean_currency",
     "clean_au_abn",
     "clean_au_acn",

diff --git a/dataprep/clean/utils.py b/dataprep/clean/utils.py
@@ -1,13 +1,12 @@
 """Common functions"""
-from typing import Dict, Union, Any
 import http.client
 import json
+from math import ceil
+from typing import Any, Dict, Union
 
 import dask.dataframe as dd
 import numpy as np
 import pandas as pd
-from math import ceil
-
 
 NULL_VALUES = {
     np.nan,