Partager via

Digital Platform API - Batch Segment Service

  • Article

The Batch Segment Service allows you to upload audiences to the Xandr platform through a batch/bulk upload framework. This data can be used in conjunction with data from buyers or sellers for the purposes of campaign targeting or yield management. All data sent through the Batch Segment Service is appended to the existing segment data already in our system.

Features include:

  • Ability to upload compressed files
  • Error checking of segment data
  • Configurable input file format
  • Confirmation of a successful upload
  • Feedback on the overall processing status
  • Association of segments to users regardless of their location
  • A high maximum data volume

For optimal results, it is strongly recommended to implement the best practices in Batch Segment Service Best Practices.

Note

The Batch Segment Service requires configuration prior to use. Consult Batch Segment Service - Configuration to learn how to configure it for your seat.

Important

Gzip is the only file compression method supported by this service.

Add a segment file for processing

Adding your segment file to the system is a three-step process. First, request a job identification number and upload URL. Second, upload your file to the assigned upload URL. Third, check the job's processing status. Note that for now, files are limited to at most 1800 segments on any individual line. If you have more than 1800 segments for one user, you must break that line into multiple lines.

Step 1. Request an upload URL and job ID

Each segment data file that is uploaded must be associated with a particular job ID. This ID is used to create the upload URL and to track the file's processing status. The first step is to send an empty POST request to the service.

Note

This service works for both api.appnexus.com and api.adnxs.com. It is available for both bidder and console logins.

$ curl -b cookies -X POST "https://api.appnexus.com/batch-segment?member_id=456"

{
 "response": {
  "id": 123,
  "status": "OK",
  "batch_segment_upload_job": {
   "job_id": "JFY8l6iMOFAFJIWCMPcy39MCt3Yleo1337618549",
   "id": 123,
   "member_id": 456,
   "last_modified": "2012-05-21 16:42:29",
   "upload_url": "https://01.data-api.prod.adnxs.net/segment-upload/JFY8l6iMOFAFJIWCMPcy39MCt3Yleo1337618549"
  },
  "start_element": 0,
  "count": 1,
  "num_elements": 100,
  "dbg_info": {
  ...
  }
 }
}

Step 2. Post the file to the upload URL

The file upload URL is given in the JSON response to Step 1 in the field upload_url. You must POST your segment file to this URL for processing. You'll receive a JSON object that tells you if the upload succeeded. Do not hardcode the upload URL in your application; make sure to dynamically grab it from the upload_url field.

Note

  • You must begin your upload to the given Upload URL within five minutes, and only one URL is valid at any given time. If you wait longer than five minutes to start your upload, you must request a new URL.
  • We recommend that you don't exceed one upload per minute. If you have more than 200 jobs waiting for processing at any given time, you will be prohibited from uploading additional jobs.
  • Your segment file should not be larger than 0.5GB.

Warning

In order for the file to upload correctly, you must specify the MIME type in the HTTP header as "Content-Type: application/octet-stream". Do not use "Content-Type: application/x-www-form-urlencode" (-d or --data flags in curl). Using an incorrect MIME type will prevent the file from being processed by the API Batch Segment Service.

Your file must conform to the Latin1 character set.

$ curl -v -H 'Content-Type:application/octet-stream' -b cookies -X POST --data-binary @segment_file "https://01.data-api.prod.adnxs.net/segment-upload/JFY8l6iMOFAFJIWCMPcy39MCt3Yleo1337618549"

* About to connect() to 01.data-api.prod.adnxs.net port 80
*   Trying 64.210.62.71... connected
* Connected to 01.data-api.prod.adnxs.net (64.210.62.71) port 80
> POST /segment-upload/FkmOY7oL2Qy2aCE7NrhE1BHVoJA0wi1337697712 HTTP/1.1
> User-Agent: curl/7.15.5 (x86_64-redhat-linux-gnu) libcurl/7.15.5 OpenSSL/0.9.8b zlib/1.2.3 libidn/0.6.5
> Host: 01.data-api.prod.adnxs.net
> Accept: */*
> Content-type:application/octet-stream
> Content-Length: 116
>
> 7652266028043224430;5848:0,5849:1440,5850:14404013681496264948522;5013:0,5014:15508802117132500293405;5851:0,5847:-1HTTP/1.1 200 OK
< Date: Tue, 22 May 2012 14:48:02 GMT
< Content-Type: application/json
< Transfer-Encoding: chunked
< Server: Jetty(7.6.2.v20120308)
* Connection #0 to host 01.data-api.prod.adnxs.net left intact
* Closing connection #0
{"response":{"segment_upload":{"job_id":"FkmOY7oL2Qy2aCE7NrhE1BHVoJA0wi1337697712"},"status":"OK"}}

Example of SSL upload URL

curl -b cookie -c cookie -X POST -s -d '' "https://api.appnexus.com/batch-segment?member_id=958"

"batch_segment_upload_job": {
      "id": 14841671,
      "job_id": "qGQhSZ1qvd2hSsJ9svTz32qgxq7z5b1439315424",
      "member_id": 958,
      "last_modified": "2015-08-11 17:50:24",
      "upload_url": "https://data-api-gslb.adnxs.net/segment-upload/qGQhSZ1qvd2hSsJ9svTz32qgxq7z5b1439315424"
    }

Step 3. Check the job status

Finally, check the processing status by sending a GET request. The JSON response contains information such as how long the file took to process and the number of errors, if any. Note that you should wait until phase="completed" before looking at the results fields, such as num_valid. For more detailed information, see JSON Fields below.

Note

  • Per Xandr SLA, allow up to 24 hours for the file to process.
  • If you are a data provider using the Impbus API, note that the batch_segment_upload_job field will be an array with a single object inside of it, for example: {"batch_segment_upload_job":[{"phase":"completed" }]}
$ curl -b cookies "https://api.appnexus.com/batch-segment?member_id=456&job_id=JFY8l6iMOFAFJIWCMPcy39MCt3Yleo1337618549"

{
 "response": {
  "start_element": 0,
  "count": 1,
  "batch_segment_upload_job": {
   "phase": "completed",
   "start_time": "2012-05-21 20:04:35",
   "uploaded_time": "2012-05-21 20:04:41",
   "validated_time": "2012-05-21 20:07:24",
   "completed_time": "2012-05-21 20:08:24",
   "error_code": null,
   "time_to_process": "2.69",
   "percent_complete": 100,
   "num_valid": 200000,
   "num_valid_user":100000
   "num_invalid_format": 0,
   "num_invalid_user": 0,
   "num_invalid_segment": 0,
   "num_unauth_segment": 1,
   "num_past_expiration": 0,
   "num_inactive_segment": 0,
   "num_other_error": 0,
   "error_log_lines": " \n\nnum_unauth_segment-4013681496264948522;5013:0,5014:1550",
   "segment_log_lines": "\n5010:100000\n5011:50000\n5012:50000",
   "id": 88,
   "job_id": "4tGhzv2PojNGQpq0MYaoaOa70cuF061337630675",
   "member_id": 456,
   "created_on": "2012-05-21 20:04:35",
   "last_modified": "2012-05-21 20:08:24"
  },
  "dbg_info": {
  ...
  },
  "status": "OK",
  "num_elements": 100
 }
}

Possible upload errors

Attempting to upload file larger than 0.5GB

{"response":{"status":"ERROR","error_code":"FILESIZE_LIMIT_EXCEEDED","errors":["Member exceeds maximum byte size allowed for a file"]}}

Error code in batch segment upload job

""batch_segment_upload_job": {
      "phase": "error",
      "start_time": "2015-08-13 18:40:32",
      "uploaded_time": null,
      "validated_time": null,
      "completed_time": null,
      "error_code": "uploading-error",
      "time_to_process": "0.00",
      "percent_complete": 0,
      "num_valid": 0,
      "num_invalid_format": 0,
      "num_valid_user": 0,
      "num_invalid_user": 0,
      "num_invalid_segment": 0,
      "num_invalid_timestamp": 0,
      "num_unauth_segment": 0,
      "num_past_expiration": 0,
      "num_inactive_segment": 0,
      "num_other_error": 0,
      "error_log_lines": null,
      "segment_log_lines": null,
      "id": 11661553,
      "job_id": "Pm3oCUf5CSVKIOt4mAqOzdt6K3qInj1431542432",
      "member_id": 958,
      "created_on": "2015-05-13 18:40:32",
      "last_modified": "2015-05-13 18:40:33"
    }

The following are errors that may happen when:

  • You've canceled the upload.
  • The upload phase exceeds 90 minutes.
  • You've reached one of its four upload limits (daily bytes, hourly bytes, or hourly lines limit).

Attempting to exceed daily byte upload limit

 {"response":{"status":"ERROR","error_code":"RATE_LIMIT_EXCEEDED","errors":["Member exceeds maximum allowed bytes per day"]}}

Attempting to exceed hourly byte upload limit

{"response":{"status":"ERROR","error_code":"RATE_LIMIT_EXCEEDED","errors":["Member exceeds maximum allowed number of lines per hour"]}}

Attempting to exceed daily lines upload limit

{"response":{"status":"ERROR","error_code":"RATE_LIMIT_EXCEEDED","errors":["Member exceeds maximum allowed number of lines per day"]}}

Attempting to exceed hourly lines upload limit

{"response":{"status":"ERROR","error_code":"RATE_LIMIT_EXCEEDED","errors":["Member exceeds maximum allowed lines per hour"]}}

Exceeding maximum time to upload

{"response":{"status":"ERROR","error_code":"RATE_LIMIT_EXCEEDED","errors":["Maximum upload time exceeded"]}}

Possible processing errors

Invalid format

If the value of the num_invalid_format field is greater than "0", check the values in the error_log_lines field.

In the example below, the num_invalid_format field displays a value of "1", with details provided in the error_log_lines field.

In the error_log_lines field:

  • num_invalid_format indicates there was a problem parsing a line in the uploaded file.
  • "failed with an illegal number of fields" indicates that the number of fields in a segment_fields block did not match what was defined in the batch-segment config. For more information, see Batch Segment Service - Configuration.

In this case, the config expects three fields to be defined in the block: SEG_ID, VALUE, and EXPIRATION, but the parser only found two fields - SEG_ID and VALUE, thus showing an error.

num_invalid_format and error_log_lines example

"batch_segment_upload_job": {
phase": "completed",
"error_code": null,
"time_to_process": "0.01",
"percent_complete": 100,
"num_valid": 0,
"num_invalid_format": 1,
"num_valid_user": 0,
"num_invalid_user": 0,
"num_invalid_segment": 0,
"num_invalid_timestamp": 0,
"num_unauth_segment": 0,
"num_past_expiration": 0,
"num_inactive_segment": 0,
"num_other_error": 0,
"error_log_lines": "num_invalid_format-WINDOWSADID-USER-ID;SEG_ID:VALUE~9 failed with an illegal number of fields",
"segment_log_lines": null,
"start_time": "2015-08-13 18:40:32",
"uploaded_time": "2015-08-13 18:42:32",
"validated_time": "2015-08-13 18:42:32",
"completed_time": "2015-08-13 18:42:33",
"id": 123412341234,
"job_id": "Pm3oCUf5CSVKIOt4mAqOzdt6K3qInj1431542432",
"member_id": 958,
"created_on": "2015-08-13 18:40:32",
"last_modified": "2015-08-13 18:42:33"
}

View your file upload history

To see metadata about all of your segment file uploads within the last 30 days, make a GET call to the service with your member_id specified in the query string. The JSON response will include an array of batch_segment_upload_job objects. For more information about the specific fields of the batch_segment_upload_job object, see JSON Fields below.

Note

File upload history is available for the last 30 days only.

m$ curl -b cookies 'https://api.appnexus.com/batch-segment?member_id=456'

{
   "response" : {
      "batch_segment_upload_job" : [
         {
           "phase": "completed",
            "start_time": "2012-05-22 16:48:55",
            "uploaded_time": "2012-05-22 16:48:56",
            "validated_time": "2012-05-22 16:49:01",
            "completed_time": "2012-05-22 16:49:01",
            "error_code": null,
            "time_to_process": "0.04",
            "percent_complete": 100,
            "num_valid": 0,
            "num_invalid_format": 0,
            "num_invalid_user": 2,
            "num_invalid_segment": 0,
            "num_unauth_segment": 1,
            "num_past_expiration": 0,
            "num_inactive_segment": 0,
            "num_other_error": 0,
            "error_log_lines": " \n\nnum_unauth_segment-4013681496264948522;5013:0,5014:1550\nnum_invalid_user-7652266028043224430;5848:0,5849:1440,5850:1440\nnum_invalid_user-8802117132500293405;5851:0,5847:-1",
            "id": 98,
            "job_id": "T1v98eIOlCZndeLGSXD0nrs57L8ES11337705335",
            "member_id": 456,
            "created_on": "2012-05-22 16:48:55",
            "last_modified": "2012-05-22 16:49:01"
         },
     ...
    }
  }
}

Note

Our API limits responses to 100 objects via pagination. You can view additional objects by appending one of these to the API call:

  • &start_element=101
  • &sort=last_modified.desc

You can read more about pagination in Throttling, Pagination, and Filtering.

REST API

HTTP Method Endpoint Description
GET https://api.appnexus.com/batch-segment/meta To find out which fields you can filter and sort by, make a GET call.

JSON fields

Field Type Description
batch_segment_upload_job object The object whose fields contain metadata describing the upload and processing job. If you are using the Impbus API, this will be an array containing a single object. For details, see the Batch Segment Upload Job section below.
id int This is the ID of the batch_segment_upload_job object associated with this request.
Default: An automatically generated number.
status string The status of the API call; successful calls return "OK".

Batch segment upload job

When you request the status of your processing job, the system returns a batch_segment_upload_job object (if you are a data provider, this will be an array containing a single object). Depending on which request you're making to the service, it will contain some or all of the following metadata. For more information about the required sequence of requests, see the Add a Segment File for Processing section below.

Note

Most metadata will only be present when phase = "completed".

Fields Type Description
completed_time date The time at which file processing was completed.
created_on date The creation date of this object.
error_code int If phase='error', this error code describes the type of error encountered. Note that an error code will only be shown here if there was an error with the uploading, validating, or processing of the file itself (i.e., it does not include invalid format or invalid segment errors). Common errors are caused by unreadable files and exceeding defined object limits.
Returns null if no errors were found.
error_log_lines string A string containing newline-separated lines. Each line lists a validation error or the reason for an error while uploading your file. You can choose how many lines (by default, 200) appear in this field.
id int The unique identifier of this object.
job_id string A string of alphanumeric characters that uniquely identifies the processing job associated with this file.
last_modified date The most recent modification date of this object (usually via POST).
member_id int Your member ID.
num_inactive_segment int The number of inactive segments in the file. Deduplicated.
num_invalid_format int The number of uploaded lines containing formatting errors (this depends upon your particular file format configuration). Duplicate lines will also be considered an invalid format.
num_invalid_segment int The number of invalid segments in the file. Deduplicated.
num_invalid_timestamp int The number of invalid timestamps in the file.
num_invalid_user int This is a count of unique input lines that have an invalid or nonexistent user.
num_other_error int This is a placeholder value that is not currently in use.
num_past_expiration int The number of expired segments in the file. Deduplicated.
num_unauth_segment int The number of segments in the file that you are unauthorized to access. Deduplicated.
num_valid int The number of valid lines in the uploaded file. Each user/segment combination is considered one line.
num_valid_user int This is a count of unique input lines that have a valid user ID.
percent_complete int The percentage of the processing that has been completed, given the current phase at the time of the request.
phase enum The current processing status. Returns one of the following values:
- "error"
- "starting"
- "uploading"
- "validating"
- "processing"
- "completed"
segment_log_lines string A string containing newline-separated lines. Each line lists a segment and how many users were successfully added to it. This field defaults to 200 lines.
start_time date The time at which file upload began.
time_to_process decimal Duration of segment file processing, measured in minutes.
upload_url string The URL where you'll upload your segment data file.
uploaded_time date The time at which the file associated with this job ID was uploaded.
validated_time date The time at which file validation was completed.

Note

Targeting may, in some cases, encompass related IDs linked to the same user. In such instances, the identifier(s) in exposure logs represent the same user as the uploaded ID.

Batch Segment Service - File Format