[beam] branch master updated: sync nltk, orjson for Python image

[em...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

emilyye pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/beam.git


The following commit(s) were added to refs/heads/master by this push:
     new 99b5bfa  sync nltk, orjson for Python image
     new ff70e74  Merge pull request #15326 from emilymye/importsync
99b5bfa is described below

commit 99b5bfa4f3603ff9bd0d4adf374bd4b5adcb436e
Author: Emily Ye <em...@google.com>
AuthorDate: Thu Aug 12 12:01:57 2021 -0700

    sync nltk, orjson for Python image
---
 sdks/python/container/Dockerfile                  | 2 ++
 sdks/python/container/base_image_requirements.txt | 3 ++-
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/sdks/python/container/Dockerfile b/sdks/python/container/Dockerfile
index 359ca2c..f517dc1 100644
--- a/sdks/python/container/Dockerfile
+++ b/sdks/python/container/Dockerfile
@@ -43,6 +43,8 @@ RUN apt-get update && \
 COPY target/base_image_requirements.txt /tmp/base_image_requirements.txt
 RUN \
     pip install -r /tmp/base_image_requirements.txt && \
+    python -c "import nltk; nltk.download('stopwords')" && \
+    rm /root/nltk_data/corpora/stopwords.zip && \
     # Check that the fast implementation of protobuf is used.
     python -c "from google.protobuf.internal import api_implementation; assert api_implementation._default_implementation_type == 'cpp'; print ('Verified fast protobuf used.')" && \
     # Remove pip cache.
diff --git a/sdks/python/container/base_image_requirements.txt b/sdks/python/container/base_image_requirements.txt
index c627b4e..a5ed07d 100644
--- a/sdks/python/container/base_image_requirements.txt
+++ b/sdks/python/container/base_image_requirements.txt
@@ -68,7 +68,7 @@ cachetools == 3.1.1
 dataclasses == 0.8 ; python_version=="3.6"
 guppy3==3.0.10
 mmh3==2.5.1
-orjson==3.5.3
+orjson==3.6.1
 python-dateutil == 2.8.1
 requests == 2.24.0
 freezegun == 0.3.15
@@ -82,6 +82,7 @@ pandas==1.2.4 ; python_version>="3.7"
 protorpc==0.12.0
 python-gflags==3.1.2
 tensorflow==2.5.0
+nltk==3.5.0
 
 # Packages needed for testing.
 tenacity>=5.0.2