MatchUp Web Service:Introduction: Difference between revisions
No edit summary |
No edit summary |
||
Line 39: | Line 39: | ||
a FullName or parsed Last / First names for matchcodes evaluating a name | a FullName or parsed Last / First names for matchcodes evaluating a name | ||
at least 1, but up to 3 Address Lines for matchcodes evaluating an address | at least 1, but up to 3 Address Lines for matchcodes evaluating an address | ||
a Zip code or a City & State pair for matchcodes evaluating an address | a Zip code or a City & State pair for matchcodes evaluating an address. | ||
===SubmitJob=== | ===SubmitJob=== |
Revision as of 19:05, 4 April 2017
MatchUp Navigation | ||||||||
---|---|---|---|---|---|---|---|---|
| ||||||||
| ||||||||
| ||||||||
| ||||||||
Sample Code |
Welcome to the Melissa Data MatchUp Web Service. The MatchUp Web Service is used to match user records according to a user-specified matchcode. This web service is different from most of the other Melissa Data Web Services in that it is batch-oriented. This means that a single request record is not returned with response data about that record independent of other record requests. A user submits a series of records, tells the web service to start processing, and then retrieves the records. The service response for each record will indicate whether it matched any other records in the submitted job.
You can use MatchUp to:
- Indicate if a record is unique or matches any other records in a submitted record set.
- Create a unique group identifier used to group and query all records that matched each other
- Count number of records in each matched group
- Using these output properties not only indicates the status of a single record, but by aggregating them, can be used to clean or evaluate the status of an entire database.
Basic Usage
Order of Operations
Unlike our other single request Web Service Solutions, to have MatchUp Web Service process your database requires a series of requests and responses.
The developer should evaluate the response of the submitted request before proceeding to the next required request.
GetMatchcodeList
Retrieves a list of available matchcodes and their descriptions. Although this request is optional, it is good practice to select a matchcode from the response of available matchcodes to ensure a CreateJob request will not return an error. The GetMatchcodeList response will also return the description of the matchcode and the combination of input Field datatypes necessary to process your database.
CreateJob
Creates a new deduplication job. A response which does not not return an error code indicates that you can start submitting records to your job.
SubmitRecords
Submit a subset of records for processing. This request is called as many times as needed to pass Arrays of records to the service for processing. Each request requires the dataframe property to be uniquely set, so when the output is retrieved, the records can be identified. The number of dataframes submitted should equal the number of dataframes returned.
While GetMatchcodeList will give a list of required Input Fields required for successful matching, MatchUp can still accurately match records if each record submitted contains... a FullName or parsed Last / First names for matchcodes evaluating a name at least 1, but up to 3 Address Lines for matchcodes evaluating an address a Zip code or a City & State pair for matchcodes evaluating an address.
SubmitJob
Submit a job for processing. Using this request lets the service know that you are done adding records to our servers for this job and you are now ready to begin processing.
GetJobStatus
Query a submitted job's status. Although this request is non-essential, it is recommended that this request always be called. Depending on the size of your submitted database, evaluating this request lets you know when it is OK to proceed with retrieving records, or if the job was terminated.
RetrieveRecords
Retrieve processing results. When GetJobStatus returns a Result Code informing you that the job is done and can begin retrieving records. The user must then send a Retrieve Results request for each data frame that was successfully submitted.