Merge branch 'main' into lorenzo-dev

.github/workflows/act-event.json

@@ -0,0 +1,4 @@
+{
+    "Comment": "Configuration for running-act event flag (see https://github.com/nektos/act#skipping-jobs)",
+    "act": true
+}

.github/workflows/act.yml

@@ -12,7 +12,10 @@ on: [push, pull_request]
 jobs:
 
   build-and-test-runner:
-    name: Run Tests
+    name: Build and Run Tests via Runner
+    # skip during local actions testing (see https://github.com/nektos/act#skipping-jobs)
+    if: ${{ !github.event.act }}        
+
     runs-on: ${{ matrix.os }}
     strategy:
       # Note: The matrix is only for the GitHub runner VM. Separate support is needed
@@ -47,9 +50,14 @@ jobs:
 
       - name: Install Dependencies
         run: |
-          pip install -r requirements.txt
+          pip install --verbose --requirement requirements.txt
+          ## TEST: pip install --verbose --requirement non-binary-requirements.txt
+
+      - name: Download data
+        run: |
+          python -m nltk.downloader -d "$HOME/nltk_data" punkt averaged_perceptron_tagger stopwords
 
-      - name: Run Python Tests Directly
+      - name: Run Python Tests via Runner
         run: |
           PYTHONPATH="$(pwd):$PYTHONPATH" ./tools/run_tests.bash
 
@@ -61,6 +69,8 @@ jobs:
 
   build-and-test-docker:
     name: Build and Run Tests via Docker
+    # include unless skip-docker set (see skip-docker.json)
+    if: ${{ !github.event.skip-docker }}
     runs-on: [ubuntu-20.04]
 
     ## TEST:

.github/workflows/debug.yml

@@ -19,9 +19,11 @@ on: [push, pull_request]
 jobs:
 
   build-and-test-runner:
-    name: Build and Run Tests
-    runs-on: ${{ matrix.os }}
+    name: Build and Run Tests via Runner
+    # skip during local actions testing (see https://github.com/nektos/act#skipping-jobs)
+    if: ${{ !github.event.act }}        
 
+    runs-on: ${{ matrix.os }}
     strategy:
       # Note: The matrix is only for the GitHub runner VM. Separate support is needed
       # by the docker container.
@@ -57,9 +59,14 @@ jobs:
 
       - name: Install Dependencies
         run: |
-          pip install -r requirements.txt
+          pip install --verbose --requirement requirements.txt
+          ## TEST: pip install --verbose --requirement non-binary-requirements.txt
+
+      - name: Download data
+        run: |
+          python -m nltk.downloader -d "$HOME/nltk_data" punkt averaged_perceptron_tagger stopwords
 
-      - name: Run Python Tests Directly
+      - name: Run Python Tests via Runner
         run: |
           PYTHONPATH="$(pwd):$PYTHONPATH" ./tools/run_tests.bash
 
@@ -71,6 +78,8 @@ jobs:
 
   build-and-test-docker:
     name: Build and Run Tests via Docker
+    # include unless skip-docker set (see skip-docker.json)
+    if: ${{ !github.event.skip-docker }}
     runs-on: [ubuntu-20.04]
 
     ## TEST:

.github/workflows/github-event.json

@@ -0,0 +1,4 @@
+{
+    "Comment": "Configuration for not-running-act event flag (see https://github.com/nektos/act#skipping-jobs)",
+    "act": false
+}

.github/workflows/github.yml

@@ -22,7 +22,7 @@ on: [push, pull_request]
 jobs:
 
   build-and-test-runner:
-    name: Build and Run Tests via Workflow Runner VM
+    name: Build and Run Tests via Runner VM
     runs-on: ${{ matrix.os }}
 
     strategy:
@@ -54,10 +54,12 @@ jobs:
 
       - name: Install Dependencies
         run: |
-          pip install -r requirements.txt
+          pip install --verbose --requirement requirements.txt
+          ## TEST: pip install --verbose --requirement non-binary-requirements.txt
 
       - name: Download data
         run: |
+          ## TODO3: put post-install support in mezcla to minimize redundancy
           python -m nltk.downloader -d "$HOME/nltk_data" punkt averaged_perceptron_tagger stopwords
 
       - name: Run Python Tests under Runner

.github/workflows/skip-docker.json

@@ -0,0 +1,4 @@
+{
+    "Comment": "Configuration for skip-docker event flag (see https://github.com/nektos/act#skipping-jobs)",
+    "skip-docker": true
+}

.github/workflows/use-docker.json

@@ -0,0 +1,4 @@
+{
+    "Comment": "Configuration for use-docker event flag (see https://github.com/nektos/act#skipping-jobs)",
+    "skip-docker": false
+}

Dockerfile

@@ -6,6 +6,9 @@
 # - Mostly based initially on
 #     https://stackoverflow.com/a/70866416 [How to install python specific version on docker?]
 # - For Docker docs, see https://docs.docker.com/get-started.
+# - The avoid linux continuation characters (i.e., \<newline), <<END-style heredocs are used: See
+#      https://www.docker.com/blog/introduction-to-heredocs-in-dockerfiles
+#   This allows for commenting out code (e.g., inline comment within heredoc group).
 #
 # Usage:
 # 1. Build the image:
@@ -98,12 +101,16 @@ RUN if [ "$PYTHON_VERSION" == "" ]; then
 COPY ./requirements.txt $REQUIREMENTS
 
 # Install the package requirements
-# This is normally handled via workflow, but is needed if docker used standalone.
-# NOTE: The workflow only handles requirements for the runner VM, not the docker container.
-## TEST: RUN if [ "$PYTHON_VERSION" != "" ]; then                                                \
-RUN if [ "$(which nltk)" == "" ]; then                                                  \
-        python -m pip install --verbose --no-cache-dir --requirement $REQUIREMENTS;     \
-    fi
+# NOTE: The workflow only handles requirements for the runner VM, not the docker container;
+# Also, the results aren't cached to save space in the image.
+RUN <<END_RUN
+  if [ "$(which nltk)" == "" ]; then
+       python -m pip install --verbose --no-cache-dir --requirement $REQUIREMENTS;
+       ## TODO?
+       ## # note: makes a second pass for failed installations, doing non-binary
+       ## python -m pip install --verbose --no-cache-dir --ignore-installed --no-binary --requirement $REQUIREMENTS;
+  fi
+END_RUN
 ## TODO3: add option for optional requirements (likewise, for all via '#full#")
 ##   RUN python -m pip install --verbose $(perl -pe 's/^#opt#\s*//g;' $REQUIREMENTS | grep -v '^#')
 
@@ -115,9 +122,17 @@ RUN python -m nltk.downloader -d /usr/local/share/nltk_data punkt averaged_perce
 
 # Install required tools and libraries (TODO: why lsb-release?)
 # Note: cleans the apt-get cache
-RUN apt-get update -y && apt-get install -y lsb-release && apt-get clean all
+RUN apt-get update -y && apt-get install --yes lsb-release && apt-get clean all
 # note: rcs needed for merge (TODO: place in required-packages.txt)
-RUN apt-get install rcs
+RUN apt-get install --yes enchant-2 rcs
+
+# Show disk usage when debugging
+RUN <<END_RUN
+  if [ "$DEBUG_LEVEL" -ge 5 ]; then
+      echo "Top directories by disk usage:";
+      du --block-size=1K / 2>&1 | sort -rn | head -20;
+  fi
+END_RUN
 
 # Run the test, normally pytest over mezcla/tests
 # Note: the status code (i.e., $?) determines whether docker run succeeds (e.g., OK if 0)

mezcla/__init__.py

@@ -18,7 +18,7 @@
 Tom O'Hara
 Feb 2022
 """
-version = "1.3.9.6"
+version = "1.3.9.7"
 __VERSION__ = version
 __version__ = __VERSION__