From 145fb1a24fc4b60e9e8a1df041f74a617e71d4da Mon Sep 17 00:00:00 2001
From: Jael Gu <mengjia.gu@zilliz.com>
Date: Mon, 20 Feb 2023 19:32:53 +0800
Subject: [PATCH] Add files

Signed-off-by: Jael Gu <mengjia.gu@zilliz.com>
---
 README.md   | 37 ++++++++++++++++++++++++++++++++++++-
 __init__.py |  4 ++++
 es_index.py | 53 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 93 insertions(+), 1 deletion(-)
 create mode 100644 __init__.py
 create mode 100644 es_index.py

diff --git a/README.md b/README.md
index 90ee163..0780e78 100644
--- a/README.md
+++ b/README.md
@@ -1,2 +1,37 @@
-# index-client
+# ElasticSearch Index
+
+## Description
+
+The index operator index the given documents in ElasticSearch to get ready for retrieval.
+It accepts a single document in dictionary or a list of documents (dictionaries) as input.
+For each document, the index automatically generates a unique id.
+To use this operator, you need to [set up ElasticSearch](https://www.elastic.co/guide/en/elasticsearch/reference/current/setup.html) in advance.
+
+## Code Example
+
+Insert an example document into ElasticSearch with address of localhost:9200 and index of 'test_index'.
+
+```python
+from datetime import datetime
+from towhee.dc2 import pipe, ops, DataCollection
+
+
+example_doc = {
+    'title': 'Test Title',
+    'author': 'Towhee',
+    'content': 'This is an example.',
+    'timestamp': datetime.now()
+    }
+
+es_insert = (
+    pipe.input('doc')
+        .map('doc', 'res', ops.elasticsearch.index_client(
+            host='localhost', port=9200, index_name='test_index'
+        ))
+        .output('doc', 'res')
+)
+
+res = es_insert(example_doc) # OR: es_insert([example_doc])
+DataCollection(res).show() # Optional: display output data
+```
 
diff --git a/__init__.py b/__init__.py
new file mode 100644
index 0000000..79ba822
--- /dev/null
+++ b/__init__.py
@@ -0,0 +1,4 @@
+from .es_index import ESIndex
+
+def index_client(*args, **kwargs):
+    return ESIndex(*args, **kwargs)
\ No newline at end of file
diff --git a/es_index.py b/es_index.py
new file mode 100644
index 0000000..83fb15e
--- /dev/null
+++ b/es_index.py
@@ -0,0 +1,53 @@
+import logging
+from typing import Union, List
+
+from elasticsearch import Elasticsearch
+import elasticsearch.helpers # type: ignore
+
+from towhee.operator import PyOperator, SharedType # type: ignore
+
+
+logger = logging.getLogger()
+
+
+class ESIndex(PyOperator):
+    """
+    Use bulk to insert docs into ElasticSearch index, using auto id generated.
+
+    Args:
+        host (`str`): host to connect ElasticSearch client
+        port (`int`): port to connect ElasticSearch client
+        index_name (`str`): index name to index input docs
+        user (`str=None`): user name to connect ElasticSearch client, defaults to None
+        password (`str=None`): user password to connect ElasticSearch client, defaults to None
+        ca_certs (`str=None`): path to CA certificate, defaults to None
+    """
+    def __init__(self, host: str, port: int, index_name: str, user: str = None, password: str = None, ca_certs: str = None):
+        super().__init__()
+        self.index_name = index_name
+        try:
+            self.client = Elasticsearch(
+                f'https://{host}:{port}',
+                ca_certs=ca_certs,
+                basic_auth=(user, password))
+            logger.info('Successfully connected to ElasticSearch client.')
+        except Exception as e:
+            logger.error('Failed to connect ElasticSearch client:\n', e)
+            raise e
+
+    def __call__(self, doc: Union[dict, List[dict]]):
+        if isinstance(doc, dict):
+            docs = [doc]
+        for x in docs:
+            assert isinstance(x, dict)
+        
+        actions = [
+            {
+                '_op_type': 'index',
+                '_index': self.index_name,
+                '_source': docs[i] 
+            }
+            for i in range(len(docs))
+        ]
+        res = elasticsearch.helpers.bulk(self.client, actions)
+        return res