How do I create an index - Simple Log Service - Alibaba Cloud Documentation Center

Call the CreateIndex API to create an index for a specified logstore.

Prerequisites

Simple Log Service is activated.
Simple Log Service SDK for Python is initialized.

Parameter description

def create_index(self, project_name, logstore_name, index_detail):

Request parameters

Parameter	Type	Required	Description
project_name	String	Yes	The name of the project. The project in Simple Log Service is used to isolate the resources of different users and control access to specific resources. For more information, see Manage a project.
logstore_name	String	Yes	The name of the logstore. The logstore in Simple Log Service is used to collect, store, and query logs. For more information, see Manage a logstore.
index_detail	IndexConfig	Yes	Index configuration information.

Description of the IndexConfig parameter:

Name	Type	Required	Description
line_config	IndexLineConfig	Yes	IndexLineConfig class, used to configure full-text index.
key_config_list	Dict[str, IndexLineConfig]	Yes	IndexKeyConfig class, used to configure field index and is of type dictionary. The keys of the dictionary are field names (strings), and the values are the corresponding index configurations (IndexLineConfig).
log_reduce	bool	No	Specifies whether to enable the LogReduce feature. After being enabled, either the whitelist or blacklist takes effect, but not both. True: Enable LogReduce. False (default): Do not enable LogReduce.

Description of the IndexLineConfig parameter:

Name	Type	Required	Description
token_list	List	Yes	The list of delimiters. Words are divided according to the given list of delimiters. For example, when the specified list of delimiters contains `,` and `:` , the log content `hello:aliyun,sls` will be split into three words: `hello`, `aliyun`, and `sls`. You can search for keywords when querying.
case_sensitive	bool	No	Specifies whether to enable case sensitivity. True: Case sensitive. False (default): Not case sensitive.
chinese	bool	No	Specifies whether to enable the Chinese word segmentation feature. After being enabled, the read and write capabilities of the shard will decrease to some extent. True: Enable False (default): Do not enable

Description of the IndexKeyConfig parameter:

Name	Type	Required	Description
index_type	String	No	The type of the index. The default value is `text`. Optional values are `text`, `long`, `double`, and `json`. `text`: Text field. You can configure the `token_list`, `case_sensitive`, `chinese`, `doc_value`, and `alias` parameters. `long`: Integer field. You can configure the `doc_value` and `alias` parameters. `double`: Floating-point number field. You can configure the `doc_value` and `alias` parameters. `json`: JSON text field. You can only configure the `json_key_config` parameter.
token_list	List	Yes	The list of delimiters. Words are divided according to the given list of delimiters. For example, when the specified list of delimiters contains `,` and `:` , the log content `hello:aliyun,sls` will be split into three words: `hello`, `aliyun`, and `sls`. You can search for keywords when querying.
case_sensitive	bool	No	Specifies whether to enable case sensitivity. True: Case sensitive. False (default): Not case sensitive.
doc_value	bool	No	Specifies whether to enable the statistical analysis feature. After it is enabled, the field can be used in analytic statements.
alias	String	No	The alias of the index key.
json_key_config	IndexJsonKeyConfig	No	IndexJsonKeyConfig class. When index_type is `json`, you need to configure this parameter.
chinese	bool	No	Specifies whether to enable the Chinese word segmentation feature. After being enabled, the read and write capabilities of the shard will decrease to some extent. True: Enable False (default): Do not enable

Description of the IndexJsonKeyConfig parameter:

Name	Type	Required	Description
index_all	bool	No	If all string values in JSON keys need to be indexed, set it to True.
max_depth	int	No	When index_all is True, only JSON values with `depth <= max_depth` will be indexed. The default value is `-1`, which indicates no depth limit.
alias	String	No	The alias of the index key.
json_keys	Dict[str, JsonKeyConfigValue]	No	The nested subfield index configuration in JSON fields. You can add subfields by calling the add_key method of the IndexJsonKeyConfig instance. For example, the log field is: `jsonKey:{ "level_1_key_1": 123, "level_1_key_2": { "level_2_key_1": "hello" } }` You can configure the `index_type` of the field `jsonKey` as `json`, and call the `add_key("level_1_key_1", "long")` and `add_key("level_1_key_2.level_2_key_1", "text")` on the `json_key_config` parameter to add field indexes for nested subfields.

Response parameters

For information about the response parameters, see CreateIndex.

Sample code

from aliyun.log import LogClient, IndexConfig, IndexKeyConfig, IndexLineConfig, IndexJsonKeyConfig
import os

# In this example, the AccessKey ID and AccessKey secret are obtained from environment variables.
access_key_id = os.environ.get('ALIBABA_CLOUD_ACCESS_KEY_ID', '')
access_key_secret = os.environ.get('ALIBABA_CLOUD_ACCESS_KEY_SECRET', '')
# Specify a Simple Log Service endpoint.
endpoint = "ap-southeast-1.log.aliyuncs.com"

# Create an instance of the LogClient class.
client = LogClient(endpoint, access_key_id, access_key_secret)

project_name = "project-1"
logstore_name = "logstore-1"


def main():
    ttl = 1
    token_list = [",", "\t", "\n", ";"]

    # Full-text index configuration.
    line_config = IndexLineConfig(token_list, case_sensitive=False, include_keys=None, exclude_keys=None, chinese=None)
    # Field index configuration.
    index_config = IndexKeyConfig(token_list, case_sensitive=False, index_type='text', doc_value=False, alias=None,
                                  json_key_config=None, chinese=None)
    # Construct index_detail configuration information.
    key_config_list = {"key_1": index_config, "key_2": index_config}
    index_detail = IndexConfig(ttl, line_config, key_config_list, all_keys_config=None, log_reduce=None)

    # Create an index for the specified logstore.
    res = client.create_index(project_name, logstore_name, index_detail)
    res.log_print()


if __name__ == '__main__':
    main()

Sample response

CreateIndexResponse:
headers: {'Server': 'AliyunSLS', 'Content-Length': '0', 'Connection': 'keep-alive', 'Access-Control-Allow-Origin': '*', 'Date': 'Sat, 12 Oct 2024 06:27:20 GMT', 'x-log-time': '1728714440', 'x-log-requestid': '670A16C88F14B9A316BC3ADE'}

Process finished with exit code 0

References

The following APIs are available to manage indexes:
For more sample code, see Alibaba Cloud Simple Log Service SDK for Python on GitHub.