RESTful Interface for Reference Data Management(4.3)
Database operations are supported via a RESTful HTTP Interface on the Platform.
Temporary files are created on the Platform host when the user executes queries via Reference Data Management. These temporary files, which are stored in $MZ_HOME/tmp/rowset
, are removed automatically when the user session expires. User sessions expire automatically after six (6) hours.
Restful Operations
The tables below describe the various operations that are available for Reference Data Management.
The service endpoint URI is http(s)://<Platform host>:<port>/api/v1.
Basic authentication is used and you must pass user credentials for each RESTful call.
Create and Get Session
Resource path: /session
HTTP method: GET
This operation returns a session id which should be appended to all subsequent API calls where this parameter is applicable.
Example
$ curl -u \ user:passw http://localhost:9000/api/v1/session
Close Session
Resource path: /session/close?sessionid=<sessionid>
HTTP method: GET
This operation closes a session. This session id is no longer valid for any subsequent API calls.
Example - Closing a session
$ curl -u user:passw \ http://localhost:9000/api/v1/session/close?sessionid=vusncl88sjghv7h8nkb0ohja6t
Get Reference Data Profiles
Resource path: /refdatas
HTTP method: GET
This operation retrieves a list of all Reference Data profiles that are available to the user.
Example - Retrieving a list
$ curl -u \ user:passw http://localhost:9000/api/v1/refdatas
Get Metadata
Resource path: /refdatas/<Reference Data profile>/table/<table name>?sessionid=<session id>
HTTP method: GET
This operation retrieves table metadata that can be used for viewing or to derive parameters for other REST APIs. If the Last Update feature is enabled, the values stored in the most recent Last Update user and timestamp column are retrieved as well. The result is returned synchronously.
Example - Get metadata
$ curl -u user:passw \ http://localhost:9000/api/v1/refdatas/Default.refprofile_star/table/APP.STAR?sessionid=vusncl88sjghv7h8nkb0ohja6t
Get Query
Resource path: /refdatas/<Reference Data profile>/table/<table name>/rowset?sessionid=<session id>
HTTP method: PUT
Body: This is where the executeQuery is included. The executeQuery JSON payload includes these options:
rowsPerPage - the max number of rows (Data Set size) that are allowed in a
rowset
of the retrieved result.selectedColumns - allows for specific columns to be selected.
filterExpression - allows optional definition of query expressions. The query will match all of the specified query expressions.
This operation executes database queries on a specific Reference Data profile and database table. The query is performed asynchronously and control is returned immediately. You can retrieve the result of the query by using /rowset/<rowset number>?sessionid=<session id>
. This operation requires input parameters that are passed in a JSON format as part of the HTTP message body.
Note!
Any ongoing query process running on the same session will be aborted and a new process for the latest query will be executed.
This operation will clear any uncommitted changes saved in the same session.
Example - Get query without query expression in JSON payload body
$ curl -X PUT -u user:passw -H 'content-type: application/json' \ http://localhost:9000/api/v1/refdatas/Default.refprofile_star/table/APP.STAR/rowset?sessionid=vusncl88sjghv7h8nkb0ohja6t \ -d '{"executeQuery": {"rowsPerPage":500,"selectedColumns":["ID","FIRSTNAME","LASTNAME"]}}'
Example - Get query with query expression in JSON payload body
$ curl -X PUT -u user:passw -H 'content-type: application/json' \ http://localhost:9000/api/v1/refdatas/Default.refprofile_star/table/APP.STAR/rowset?sessionid=vusncl88sjghv7h8nkb0ohja6t \ -d '{"executeQuery": {"rowsPerPage":500,"selectedColumns":["ID","FIRSTNAME","LASTNAME"],"filterExpression":[{"col":"ID","expr":9,"op":">="},{"col":"LAST_NAME","expr":"Smith","op":"not like"}]}}'
Example - Get query using a JSON payload file
$ curl -X PUT -T="example.json" -u user:passw -H 'content-type: application/json' \ 'http://localhost:9000/api/v1/refdatas/Default.refprofile_star/table/APP.STAR/rowset?sessionid=vusncl88sjghv7h8nkb0ohja6t
example.json:
{ "executeQuery": { "rowsPerPage": 500, "selectedColumns": ["ID", "FIRSTNAME", "LASTNAME"] } }
Get Data Sets
Resource path: /rowset/<rowset number>?sessionid=<session id>
HTTP method: GET
Data sets can be retrieved once downloaded to the file system of the Platform. This operation returns a data set for the given rowset (sequence number). The total number of available data sets can be queried with the status
operation.
Example - Get data sets
$ curl -u user:passw \ http://localhost:9000/api/v1/rowset/0?sessionid=vusncl88sjghv7h8nkb0ohja6t
Get Status
Resource path: /status?sessionid=<session id>
HTTP method: GET
This operation returns a status message. It can be used to retrieve active processes and to query the number of available rows, data sets, and the status of imports and exports. If the Last Update feature is enabled, the values stored in the most recent Last Update user and timestamp column are retrieved as well.
Example - Get status
$ curl -u user:passw \ http://localhost:9000/api/v1/status?sessionid=vusncl88sjghv7h8nkb0ohja6t
Abort Process
Resource path: /status/abort?sessionid=<session id>
HTTP method: GET
This operation requests the active process to abort.
Note!
To prevent a user from initiating another operation before the first operation initiated is complete, Abort Process
can be used before an operation is complete.
Example - Abort process
$ curl -u user:passw \ http://localhost:9000/api/v1/status/abort?sessionid=vusncl88sjghv7h8nkb0ohja6t
Table Export
Resource path: /refdatas/<Reference Data profile>/table/<table name>/download?sessionid=<sessionid>
HTTP method: POST
Body: This is where the exportParams are included. The exportParams JSON payload includes these options:
opts
textQualifier - designated as double quotes by default.
separator - designated as a comma by default.
extent - designated as
all
by default.all - export all rows in the table.
selected - export rows from the result of
Get Query.
selectedColumns - allows for specific columns to be selected.
Note!
For the extend
option, selected
value is only applicable when Get Query
is applied prior Table Export
.
This operation performs a database table export. Input parameters are passed in a JSON format as part of the HTTP message body.
Example - Table export without options
$ curl -X POST \ 'http://localhost:9000/api/v1/refdatas/ref_data_mgmt_oracle.ref_data_mgmt_pf/table/TABLE.CUSTOMERS/download?sessionid=p3hce86dkb4rmls9peh4e8rps9' \ -u "user:passw" \ -d 'exportParams={}' \ > Export.csv
Example - Table export with options
$ curl -X POST \ 'http://localhost:9000/api/v1/refdatas/ref_data_mgmt_oracle.ref_data_mgmt_pf/table/TABLE.CUSTOMERS/download?sessionid=p3hce86dkb4rmls9peh4e8rps9' \ -u "user:passw" \ -d 'exportParams={"opts":{"textQualifier":"'\''","separator":";","extent":"all"},"selectedColumns":["CITY","CUSTOMER_NAME"]}' > Export.csv
Table Import
Resource path: /refdatas/<Reference Data profile>/table/<table name>/upload?sessionid=<sessionid>
HTTP method: POST
Body: This is where the file and the input parameters are included. The exportParams format includes these options:
textQualifier - designated as double quotes by default.
separator - designated as a comma by default.
opts - designated as
append
by default.append - imported rows are appended to the table.
truncate - the existing data in the table are truncated before the import is executed.
force - designated as
false
by default.
This operation performs a database table import. Input parameters are passed in a JSON format as part of the HTTP message body.
Example - Table import without options
$ curl -i -u "user:passw" \ 'http://localhost:9000/api/v1/refdatas/ref_data_mgmt_oracle.ref_data_mgmt_pf/table/TABLE.CUSTOMERS/upload?sessionid=rssrh20dcd8lc1j505b3bqnstc' \ -F file=@/path/to/import_test.csv
Example - Table import with options
$ curl -i -u "user:passw" \ 'http://localhost:9000/api/v1/refdatas/ref_data_mgmt_oracle.ref_data_mgmt_pf/table/TABLE.CUSTOMERS/upload?sessionid=p3hce86dkb4rmls9peh4e8rps9' \ -F file=@/path/to/import_test.csv \ -F 'opts={"textQualifier":"\'","separator":",","op":"truncate","force":false}'
Save Changes
Resource path: /save?sessionid=<session id>
HTTP method: PUT
Body: This is where the dataEdits are included. The dataEdit JSON payload includes these options:
refProfile - Reference Data Management Profile.
tableName - Database table name.
updates - modification parameters.
action - to specify the type of modification.
insert - insert a new row to the table.
update - edit an existing row in the table.
delete - delete an existing row in the table.
ids - to specify column value pairs of primary key(s).
column - private key column name.
value - private key value for the respective column.
values - to specify column value pairs to be inserted/updated.
column - column name to be inserted/updated.
value - insert/update value for the respective column.
Note!
When inserting a row, specifying a pseudo ids
column value pair allows Cancel Changes
to be applied to that specific insert row modification.
This operation saves data grid edit modification (post insert/update/delete). Changes are saved within the client session. Input parameters are passed in a JSON format as part of the HTTP message body.
Note!
tableName
and refProfile
are mandatory in order to save. A single table can be modified in a single session only.
Note!
The save operation is supported either on Oracle (based on the ROWID pseudo column) or non-Oracle type tables containing a Primary Key constraint. Non-Oracle tables without a Primary Key are not supported for data modifications.
Example - Save changes with a JSON payload body
$ curl -X PUT -u "user:passw" -H 'content-type: application/json \ -d '{"dataEdit":{"refProfile":"Default.refTest","tableName":"MZADMIN.REFRENCE_DATA","updates":[{"action":"delete","ids":[{"column":"ID","value":8}]}]}}' \ http://localhost:9000/api/v1/save?sessionid=vusncl88sjghv7h8nkb0ohja6t
Example - Save changes using a JSON payload file
$ curl -X PUT -T="example.json" -u "user:passw" -H 'content-type: application/json \ http://localhost:9000/api/v1/save?sessionid=vusncl88sjghv7h8nkb0ohja6t
example.json:
{ "dataEdit" : { "refProfile" : "Default.refTest", "tableName" : "MZADMIN.REFRENCE_DATA", "updates" : [ { "action" : "insert", "ids" : [ { "column" : "ROWID", "value" : "ins0" } ], "values" : [ { "column" : "ID", "value" : 645 }, { "column" : "FIRST_NAME", "value" : "Roberts" }, { "column" : "LAST_NAME", "value" : "Polis" } ] }, { "action" : "update", "ids" : [ { "column" : "ID", "value" : "6" } ], "values" : [ { "column" : "LAST_NAME", "value" : "Wick" }, { "column" : "FIRST_NAME", "value" : "John" } ] }, { "action" : "delete", "ids" : [ { "column" : "id", "value" : "8" } ] } ] } }
Commit Changes
Resource path: /save/commit?force=<true|false>&sessionid=<session id>
HTTP method: GET
This operation applies saved edits in the database and commits the work in case of success. You can use force
commit in case of errors. If the Last Update feature is enabled, the user name and modification timestamp values for insert/update modifications are stored in the Last Update columns specified in the Reference Data Management Profile. The Last Update information is used by the Get Status
operation to retrieve the most recent Last Update user and timestamp.
Example - Commit changes
$ curl -u user:passw \ http://localhost:9000/api/v1/commit?force=false&sessionid=vusncl88sjghv7h8nkb0ohja6t
List Changes
Resource path: /save/list?sessionid=<session id>
HTTP method: GET
This operation returns a list of the modifications saved.
Example - List changes
$ curl -u user:passw \ http://localhost:9000/api/v1/save/list?sessionid=vusncl88sjghv7h8nkb0ohja6t
Cancel Changes
Resource path: /save/cancel?sessionid=<session id>
HTTP method: PUT
Body: This is where the dataEdits are included. The dataEdit JSON payload includes these options:
scope - specify the scope of changes to be canceled.
single - allow only specific modifications identified by
ids
to be cancelled.all - allow all the modifications to be canceled.
ids - to specify column value pairs of primary key(s). This is only required when the
scope
issingle
.column - private key column name.
value - private key value for the respective column.
This operation cancels the changes made from being saved. Input parameters are passed in a JSON format as part of the HTTP message body.
Note!
Pseudo Primary Keys for inserted rows can be included in the Save Changes
operation to allow the cancel function for a specific insert row modification.
Note!
SaveSize
in the Cancel Changes
response indicates the number of changes remaining.
Example - Cancel changes
$ curl -u user:passw -H 'content-type: application/json' \ -d '{"dataEdit":{"scope":"single","ids": [{"id":[{"column":"ORDER_NUM","value":10398005}]}] \ http://localhost:9000/api/v1/save/cancel?sessionid=vusncl88sjghv7h8nkb0ohja6t
Show Demo Query
Resource path: /demo/queryRequestParameters
HTTP method: GET
This operation shows an example JSON payload format that applies for a Get Query
operation.
Show Demo Changes
Resource path: /demo/dataEditRequestParameters
HTTP method: GET
This operation shows and example JSON payload format that applies for a Save Changes
operation.