GH-233: add lang_detect pipeline (#736)

.github/workflows/ci-pipe-langdetect.yml

@@ -0,0 +1,35 @@
+on:
+  pull_request:
+    branches: [ main ]
+    types:
+      - labeled
+      - synchronize
+
+jobs:
+  ci-langdetect:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Get PR Labels
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          PR_LABELS=$(gh pr view https://github.com/${{ github.repository }}/pull/${{ github.event.pull_request.number }} --json labels --jq '[.labels[].name]')
+          echo "$PR_LABELS"
+          echo "PR_LABELS=$PR_LABELS" >> $GITHUB_ENV
+      - name: Checkout
+        if: contains(env.PR_LABELS, 'ci-langdetect')
+        uses: actions/checkout@v2
+      - name: Set up Python 3.7
+        if: contains(env.PR_LABELS, 'ci-langdetect')
+        uses: actions/setup-python@v2
+        with:
+          python-version: 3.7
+      - name: Install dependencies
+        if: contains(env.PR_LABELS, 'ci-langdetect')
+        run: |
+          python -m pip install --upgrade pip
+          pip install pytest tox
+      - name: Run Test
+        if: contains(env.PR_LABELS, 'ci-langdetect')
+        run: |
+          tox -e langdetect

README.md

@@ -304,13 +304,13 @@ Lang Detect API. Thanks to awesome work from [FastText](https://fasttext.cc/docs
 Install extend dependencies and models
 
     ```bash
-    $ pip install underthesea[lang-detect]
+    $ pip install underthesea[langdetect]
     ```
 
 Usage examples in script
 
     ```python
-    >>> from underthesea.pipeline.lang_detect import lang_detect
+    >>> from underthesea import lang_detect
 
     >>> lang_detect("Cựu binh Mỹ trả nhật ký nhẹ lòng khi thấy cuộc sống hòa bình tại Việt Nam")
     vi

setup.py

@@ -43,8 +43,8 @@
     'prompt': [
         'openai'
     ],
-    'lang-detect': [
-        'fasttext   '
+    'langdetect': [
+        'fasttext'
     ]
 }
 setup(

tox.ini

@@ -1,5 +1,5 @@
 [tox]
-envlist = traditional,deep,prompt
+envlist = traditional,deep,prompt,langdetect
 
 [testenv:traditional]
 basepython = python3.7
@@ -62,8 +62,23 @@ setenv =
     OPENAI_API_KEY = {env:OPENAI_API_KEY:}
 
 commands =
-    ; Modules with deep learning
+    ; Modules with prompt
     pip install -e .[prompt]
 
-    ; ner module
-    python -m unittest tests.pipeline.classification.test_prompt
+    ; prompt module
+    python -m unittest tests.pipeline.classification.test_prompt
+
+[testenv:langdetect]
+basepython = python3.7
+deps=
+    pip>=20.3
+
+setenv =
+    PYTHONPATH = {toxinidir}:{toxinidir}/underthesea
+
+commands =
+    ; install dependencies
+    pip install -e .[langdetect]
+
+    ; lang_detect module
+    python -m unittest tests.pipeline.lang_detect.test_lang_detect

underthesea/cli.py

@@ -96,6 +96,7 @@ def info():
     print("            classify : OK")
     print("           sentiment : OK")
     print("    dependency_parse : OK")
+    print("         lang_detect : OK")
     print("           resources : OK")