Merge pull request #42 from gestaogovbr/clean-libs

clean libs and add requirements files

Author: vitorbellini

Dockerfile

@@ -1,28 +1,6 @@
-FROM apache/airflow:2.7.1-python3.10
+# for dev: docker build -t ghcr.io/gestaogovbr/airflow2-docker:latest-dev --build-arg dev_build=true .
 
-ARG PYTHON_DEPS=" \
-    ctds==1.12.0 \
-    tqdm==4.60.0 \
-    ijson==3.0.4 \
-    pysmb==1.2.6 \
-    xlrd==1.2.0 \
-    pygsheets==2.0.5 \
-    ipdb==0.13.3 \
-    py-trello==0.17.1 \
-    PyPDF2==1.26.0 \
-    frictionless==5.11.1 \
-    great-expectations==0.17.2 \
-    unidecode==1.2.0 \
-    odfpy==1.4.1 \
-    openpyxl==3.0.7 \
-    pytest==6.2.5 \
-    ckanapi==4.6 \
-    sharepy==1.3.0 \
-    Office365-REST-Python-Client==2.3.14 \
-    GeoAlchemy2==0.10.2 \
-    acryl-datahub-airflow-plugin==0.10.4 \
-    geopandas==0.12.2 \
-    "
+FROM apache/airflow:2.7.3-python3.10
 
 USER root
 RUN apt-get update \
@@ -38,7 +16,7 @@ RUN apt-get update \
   && curl https://packages.microsoft.com/keys/microsoft.asc | apt-key add --no-tty - \
   && curl https://packages.microsoft.com/config/debian/10/prod.list > /etc/apt/sources.list.d/mssql-release.list \
   && apt-get update -yqq \
-  && ACCEPT_EULA=Y apt-get install -yqq msodbcsql17 \
+  && ACCEPT_EULA=Y apt-get install -yqq msodbcsql17 mssql-tools \
   && sed -i 's,^\(MinProtocol[ ]*=\).*,\1'TLSv1.0',g' /etc/ssl/openssl.cnf \
   && sed -i 's,^\(CipherString[ ]*=\).*,\1'DEFAULT@SECLEVEL=1',g' /etc/ssl/openssl.cnf \
   && curl -O http://acraiz.icpbrasil.gov.br/credenciadas/CertificadosAC-ICP-Brasil/ACcompactado.zip \
@@ -63,24 +41,33 @@ RUN curl https://ssltools.digicert.com/chainTester/webservice/validatecerts/cert
 
 USER airflow
 
-RUN if [ -n "${PYTHON_DEPS}" ]; \
-      then pip install --no-cache-dir --user ${PYTHON_DEPS}; \
-    fi \
-    && mkdir /opt/airflow/export-data
+WORKDIR /opt/airflow
 
-RUN pip install --no-cache-dir --user \
-    apache-airflow[jdbc,microsoft.mssql,samba,google_auth,odbc,sentry] \
+COPY requirements-uninstall.txt .
+COPY requirements-cdata-dags.txt .
+
+RUN pip uninstall -y -r requirements-uninstall.txt && \
+    pip install --no-cache-dir --user -r requirements-cdata-dags.txt && \
+    pip install --no-cache-dir --user -r \
+    https://raw.githubusercontent.com/gestaogovbr/Ro-dou/main/requirements.txt && \
+    pip install --no-cache-dir --user \
+    apache-airflow[jdbc,microsoft.mssql,samba,odbc,sentry] \
     apache-airflow-providers-docker \
     apache-airflow-providers-common-sql \
-    apache-airflow-providers-telegram
+    apache-airflow-providers-telegram \
+    acryl-datahub-airflow-plugin==0.10.4
 
 ARG dev_build="false"
 RUN \
   if [[ "${dev_build}" == "false" ]] ; \
   then pip install --no-cache-dir --user apache-airflow-providers-fastetl; \
   else \
-  echo ***apache-airflow-providers-fastetl not installed*** ; \
+  echo ***apache-airflow-providers-fastetl not installed***  && \
+  pip install --no-cache-dir --user -r https://raw.githubusercontent.com/gestaogovbr/FastETL/main/requirements.txt ; \
   fi
 
 RUN while [[ "$(curl -s -o /tmp/thawte.pem -w ''%{http_code}'' https://ssltools.digicert.com/chainTester/webservice/validatecerts/certificate?certKey=issuer.intermediate.cert.98&fileName=Thawte%20RSA%20CA%202018&fileExtension=txt)" != "200" ]]; do sleep 1; done
 RUN cat /tmp/thawte.pem >> /home/airflow/.local/lib/python3.10/site-packages/certifi/cacert.pem
+RUN echo 'export PATH="$PATH:/opt/mssql-tools/bin"' >> ~/.bashrc && \
+    source ~/.bashrc
+RUN rm ACcompactado.zip requirements-cdata-dags.txt requirements-uninstall.txt

requirements-cdata-dags.txt

@@ -0,0 +1,19 @@
+# airflow-dags
+ctds==1.12.0
+tqdm==4.60.0
+py-trello==0.17.1
+frictionless==5.11.1
+great-expectations==0.17.2
+openpyxl==3.0.7
+Office365-REST-Python-Client==2.3.14
+geopandas==0.12.2
+pandas==1.5.2,<2
+
+# airflow-dags-delog
+# tqdm==4.60.0
+unidecode==1.2.0
+
+# airflow-dags-detru
+
+# airflow-commons
+# great-expectations==0.17.2

requirements-uninstall.txt

@@ -0,0 +1,129 @@
+# from apache-airflow-providers-amazon
+apache-airflow-providers-amazon
+aiobotocore
+jsonpath-ng
+boto3
+mypy-boto3-appflow
+mypy-boto3-rds
+mypy-boto3-redshift-data
+mypy-boto3-s3
+redshift-connector
+scramp
+sqlalchemy-redshift
+watchtower
+
+# from apache-airflow-providers-elasticsearch
+apache-airflow-providers-elasticsearch
+elasticsearch
+elastic-transport
+
+# from apache-airflow-providers-sendgrid
+apache-airflow-providers-sendgrid
+sendgrid
+starkbank-ecdsa
+
+# from apache-airflow-providers-snowflake
+apache-airflow-providers-snowflake
+snowflake-connector-python
+snowflake-sqlalchemy
+pycryptodomex
+tomlkit
+asn1crypto
+oscrypto
+
+# from apache-airflow-providers-microsoft-azure
+apache-airflow-providers-microsoft-azure
+adal
+azure-batch
+azure-cosmos
+azure-datalake-store
+azure-identity
+azure-keyvault-secrets
+azure-kusto-data
+azure-mgmt-containerinstance
+azure-mgmt-cosmosdb
+azure-mgmt-datafactory
+azure-mgmt-datalake-store
+azure-mgmt-resource
+azure-servicebus
+azure-storage-blob
+azure-storage-common
+azure-storage-file
+azure-storage-file-datalake
+azure-synapse-spark
+azure-common
+azure-core
+msal-extensions
+msrest
+azure-mgmt-datalake-nspkg
+adal
+azure-mgmt-nspkg
+azure-mgmt-containerregistry
+azure-mgmt-core
+azure-mgmt-storage
+azure-nspkg
+azure-storage-file-share
+
+# from apache-airflow-providers-google
+apache-airflow-providers-google
+gcloud-aio-auth
+gcloud-aio-bigquery
+gcloud-aio-storage
+google-ads
+google-api-core
+google-api-python-client
+google-cloud-aiplatform
+google-cloud-automl
+google-cloud-batch
+google-cloud-bigquery-datatransfer
+google-cloud-bigtable
+google-cloud-build
+google-cloud-compute
+google-cloud-container
+google-cloud-datacatalog
+google-cloud-dataflow-client
+google-cloud-dataform
+google-cloud-dataplex
+google-cloud-dataproc
+google-cloud-dataproc-metastore
+google-cloud-dlp
+google-cloud-kms
+google-cloud-language
+google-cloud-logging
+google-cloud-memcache
+google-cloud-monitoring
+google-cloud-orchestration-airflow
+google-cloud-os-login
+google-cloud-pubsub
+google-cloud-redis
+google-cloud-run
+google-cloud-secret-manager
+google-cloud-spanner
+google-cloud-speech
+google-cloud-storage
+google-cloud-storage-transfer
+google-cloud-tasks
+google-cloud-texttospeech
+google-cloud-translate
+google-cloud-videointelligence
+google-cloud-vision
+google-cloud-workflows
+grpcio-gcp
+json-merge-patch
+looker-sdk
+pandas-gbq
+proto-plus
+PyOpenSSL
+sqlalchemy-bigquery
+sqlalchemy-spanner
+google-auth-oauthlib
+google-cloud-appengine-logging
+google-cloud-audit-log
+google-cloud-bigquery
+google-cloud-bigquery-storage
+google-cloud-core
+google-cloud-resource-manager
+google-crc32c
+google-resumable-media
+grpc-google-iam-v1
+pydata-google-auth