Add support for running as a container.

.github/workflows/push-to-docker-hub.yml

@@ -0,0 +1,15 @@
+name: Build, scan and push to Docker Hub
+
+on:
+  push:
+    tags:
+      - '*'
+    branches:
+      - '*'
+
+jobs:
+  run_docker_build_workflow:
+    uses: ukwa/ukwa-services/.github/workflows/push-to-docker-hub.yml@master
+    secrets:
+      DOCKER_HUB_USERNAME: ${{ secrets.DOCKER_HUB_USERNAME }}
+      DOCKER_HUB_ACCESS_TOKEN: ${{ secrets.DOCKER_HUB_ACCESS_TOKEN }}

Dockerfile

@@ -1,4 +1,4 @@
-FROM python:3-alpine
+FROM python:3
 
 WORKDIR /usr/src/access
 

requirements.txt

@@ -1,18 +1,24 @@
-certifi==2020.12.5
-chardet==4.0.0
-crawlstreams @ git+https://github.com/ukwa/crawl-streams.git@a5c9627272e947141987d233f32fc446d560c70d
-filelock==3.0.12
-idna==2.10
-kafka-python==1.4.7
-kevals @ git+https://github.com/ukwa/kevals.git@e6203ea850c854983f5f1dc72580fcb1d4435a94
-mmh3==2.5.1
-mrjob==0.7.4
-psycopg2-binary==2.8.6
-python-snappy==0.5.4
-PyYAML==5.4.1
-requests==2.25.1
+certifi==2023.7.22
+charset-normalizer==3.3.2
+cramjam==2.7.0
+crawlstreams @ git+https://github.com/ukwa/crawl-streams.git@37d9755f196e13eee11642e5111321057dc5c701
+fastparquet==2023.10.1
+filelock==3.13.1
+fsspec==2023.10.0
+idna==3.4
+kafka-python==2.0.2
+kevals @ git+https://github.com/ukwa/kevals.git@cfbeef82365764c0b829c223859903097a8f421f
+mmh3==4.0.1
+numpy==1.26.2
+packaging==23.2
+pandas==2.1.3
+python-dateutil==2.8.2
+python-snappy==0.6.1
+pytz==2023.3.post1
+requests==2.31.0
 requests-file==1.5.1
-six==1.15.0
-tldextract==3.1.0
+six==1.16.0
+tldextract==5.1.0
+tzdata==2023.3
 urlcanon==0.3.1
-urllib3==1.26.2
+urllib3==2.1.0

setup.py

@@ -21,7 +21,7 @@ def get_version():
     zip_safe = False, # Avoid installing as .egg file, as Luigi can't handle these.
     entry_points={
         'console_scripts': [
-            'crawldb=crawldb.solr.cli:main'
+            'crawldb=crawldb.parquet.cli:main'
         ]
     }
 )