Add support for mongoDB (#1207)

* [mongo] build libmongoc

* [tests] setup container for tests

* add platform based genrule conditions

* expose handler python api

* lint fixes

* disable sasl build for linux

* modify build opts for libmongoc

add unistd.h for windows build

add missing include

os based bson build

bson os fix

* expose MongoDBIODataset

* expose MongoDBWriter python api

* modify build options

add missing header

modify ssl build opts

modify ssl opts

disable aws auth opt

ssl secure channel for windows

ssl opts for linux and mac

disable ssl for linux

modify bson opts

openssl based build

disable ssl for mac and linux

enable dnsapi for windows

use mongo-c-driver-1.16.2

enable sasl support

add linkopts for boringssl build

update boringssl version

add openssl deps for centos

add libssl-dev dep

become sudo

update README.md instructions

modify build options

* build separate .so for mongo_ops

* install make for centos builds

* build libtensorflow_io_mongodb.so in windows

* handle scopes for centos 7

* use boringssl for windows

* lint fixes

* disable libcrypto for windows

* patch boringssl dependency

Author: kvignesh1420

.github/workflows/build.yml

@@ -302,6 +302,7 @@ jobs:
           bash -x -e tests/test_pubsub/pubsub_test.sh
           bash -x -e tests/test_prometheus/prometheus_test.sh start
           bash -x -e tests/test_elasticsearch/elasticsearch_test.sh start
+          bash -x -e tests/test_mongodb/mongodb_test.sh start
           bash -x -e tests/test_azure/start_azure.sh
           bash -x -e tests/test_sql/sql_test.sh
           bash -x -e tests/test_gcloud/test_gcs.sh gcs-emulator

.kokorun/io_cpu.sh

@@ -80,6 +80,7 @@ bash -x -e tests/test_prometheus/prometheus_test.sh start
 bash -x -e tests/test_azure/start_azure.sh
 bash -x -e tests/test_sql/sql_test.sh sql
 bash -x -e tests/test_elasticsearch/elasticsearch_test.sh start
+bash -x -e tests/test_mongodb/mongodb_test.sh start
 
 docker run -i --rm -v $PWD:/v -w /v --net=host \
   buildpack-deps:20.04 bash -x -e .github/workflows/build.wheel.sh python${PYTHON_VERSION}

README.md

@@ -282,7 +282,7 @@ the shared libraries on CentOS 8:
 #!/usr/bin/env bash
 
 # Install gcc/g++, git, unzip/which (for bazel), and python3
-sudo yum install -y python3 python3-devel gcc gcc-c++ git unzip which
+sudo yum install -y python3 python3-devel gcc gcc-c++ git unzip which make
 
 # Install Bazel version specified in .bazelversion
 curl -sSOL https://github.com/bazelbuild/bazel/releases/download/$(cat .bazelversion)/bazel-$(cat .bazelversion)-installer-linux-x86_64.sh
@@ -321,7 +321,7 @@ The following will install bazel, devtoolset-9, rh-python36, and build the share
 
 # Install centos-release-scl, then install gcc/g++ (devtoolset), git, and python 3
 sudo yum install -y centos-release-scl
-sudo yum install -y devtoolset-9 git rh-python36
+sudo yum install -y devtoolset-9 git rh-python36 make
 
 # Install Bazel version specified in .bazelversion
 curl -sSOL https://github.com/bazelbuild/bazel/releases/download/$(cat .bazelversion)/bazel-$(cat .bazelversion)-installer-linux-x86_64.sh

WORKSPACE

@@ -281,7 +281,7 @@ http_archive(
     patch_cmds = [
         "rm -f src/win32_config.h",
         # TODO: Remove the fowllowing once librdkafka issue is resolved.
-        """sed -i.bak '\|rd_kafka_log(rk,|,/ exceeded);/ s/^/\/\//' src/rdkafka_cgrp.c""",
+        """sed -i.bak '\\|rd_kafka_log(rk,|,/ exceeded);/ s/^/\\/\\//' src/rdkafka_cgrp.c""",
     ],
     sha256 = "f7fee59fdbf1286ec23ef0b35b2dfb41031c8727c90ced6435b8cf576f23a656",
     strip_prefix = "librdkafka-1.5.0",
@@ -1092,3 +1092,16 @@ http_archive(
         "https://github.com/apache/pulsar/archive/v2.6.1.tar.gz",
     ],
 )
+
+http_archive(
+    name = "libmongoc",
+    build_file = "//third_party:libmongoc.BUILD",
+    patch_cmds = [
+        "sed -i.bak 's/undef MONGOC_LOG_DOMAIN/undef MONGOC_LOG_DOMAIN\\'$'\\n''# define BIO_get_ssl(b,sslp)  BIO_ctrl(b,BIO_C_GET_SSL,0,(char *)(sslp))\\'$'\\n''# define BIO_do_handshake(b)  BIO_ctrl(b,BIO_C_DO_STATE_MACHINE,0,NULL)/g' src/libmongoc/src/mongoc/mongoc-stream-tls-openssl.c",
+    ],
+    sha256 = "0a722180e5b5c86c415b9256d753b2d5552901dc5d95c9f022072c3cd336887e",
+    strip_prefix = "mongo-c-driver-1.16.2",
+    urls = [
+        "https://github.com/mongodb/mongo-c-driver/releases/download/1.16.2/mongo-c-driver-1.16.2.tar.gz",
+    ],
+)

tensorflow_io/core/BUILD

@@ -679,6 +679,22 @@ cc_library(
     alwayslink = 1,
 )
 
+cc_library(
+    name = "mongodb_ops",
+    srcs = [
+        "kernels/mongodb_kernels.cc",
+        "ops/mongodb_ops.cc",
+    ],
+    copts = tf_io_copts(),
+    linkstatic = True,
+    deps = [
+        "@libmongoc",
+        "@local_config_tf//:libtensorflow_framework",
+        "@local_config_tf//:tf_header_lib",
+    ],
+    alwayslink = 1,
+)
+
 cc_binary(
     name = "python/ops/libtensorflow_io.so",
     copts = tf_io_copts(),
@@ -707,6 +723,7 @@ cc_binary(
         "//tensorflow_io/core:sql_ops",
         "//tensorflow_io/core:text_ops",
         "//tensorflow_io/core:ignite_ops",
+        "//tensorflow_io/core:mongodb_ops",
         "@local_config_tf//:libtensorflow_framework",
         "@local_config_tf//:tf_header_lib",
     ] + select({