firehose.d.ts 98 KB
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510
import {Request} from '../lib/request';
import {Response} from '../lib/response';
import {AWSError} from '../lib/error';
import {Service} from '../lib/service';
import {ServiceConfigurationOptions} from '../lib/service';
import {ConfigBase as Config} from '../lib/config';
interface Blob {}
declare class Firehose extends Service {
  /**
   * Constructs a service object. This object has one method for each API operation.
   */
  constructor(options?: Firehose.Types.ClientConfiguration)
  config: Config & Firehose.Types.ClientConfiguration;
  /**
   * Creates a Kinesis Data Firehose delivery stream. By default, you can create up to 50 delivery streams per AWS Region. This is an asynchronous operation that immediately returns. The initial status of the delivery stream is CREATING. After the delivery stream is created, its status is ACTIVE and it now accepts data. Attempts to send data to a delivery stream that is not in the ACTIVE state cause an exception. To check the state of a delivery stream, use DescribeDeliveryStream. A Kinesis Data Firehose delivery stream can be configured to receive records directly from providers using PutRecord or PutRecordBatch, or it can be configured to use an existing Kinesis stream as its source. To specify a Kinesis data stream as input, set the DeliveryStreamType parameter to KinesisStreamAsSource, and provide the Kinesis stream Amazon Resource Name (ARN) and role ARN in the KinesisStreamSourceConfiguration parameter. A delivery stream is configured with a single destination: Amazon S3, Amazon ES, Amazon Redshift, or Splunk. You must specify only one of the following destination configuration parameters: ExtendedS3DestinationConfiguration, S3DestinationConfiguration, ElasticsearchDestinationConfiguration, RedshiftDestinationConfiguration, or SplunkDestinationConfiguration. When you specify S3DestinationConfiguration, you can also provide the following optional values: BufferingHints, EncryptionConfiguration, and CompressionFormat. By default, if no BufferingHints value is provided, Kinesis Data Firehose buffers data up to 5 MB or for 5 minutes, whichever condition is satisfied first. BufferingHints is a hint, so there are some cases where the service cannot adhere to these conditions strictly. For example, record boundaries might be such that the size is a little over or under the configured buffering size. By default, no encryption is performed. We strongly recommend that you enable encryption to ensure secure data storage in Amazon S3. A few notes about Amazon Redshift as a destination:   An Amazon Redshift destination requires an S3 bucket as intermediate location. Kinesis Data Firehose first delivers data to Amazon S3 and then uses COPY syntax to load data into an Amazon Redshift table. This is specified in the RedshiftDestinationConfiguration.S3Configuration parameter.   The compression formats SNAPPY or ZIP cannot be specified in RedshiftDestinationConfiguration.S3Configuration because the Amazon Redshift COPY operation that reads from the S3 bucket doesn't support these compression formats.   We strongly recommend that you use the user name and password you provide exclusively with Kinesis Data Firehose, and that the permissions for the account are restricted for Amazon Redshift INSERT permissions.   Kinesis Data Firehose assumes the IAM role that is configured as part of the destination. The role should allow the Kinesis Data Firehose principal to assume the role, and the role should have permissions that allow the service to deliver the data. For more information, see Grant Kinesis Data Firehose Access to an Amazon S3 Destination in the Amazon Kinesis Data Firehose Developer Guide.
   */
  createDeliveryStream(params: Firehose.Types.CreateDeliveryStreamInput, callback?: (err: AWSError, data: Firehose.Types.CreateDeliveryStreamOutput) => void): Request<Firehose.Types.CreateDeliveryStreamOutput, AWSError>;
  /**
   * Creates a Kinesis Data Firehose delivery stream. By default, you can create up to 50 delivery streams per AWS Region. This is an asynchronous operation that immediately returns. The initial status of the delivery stream is CREATING. After the delivery stream is created, its status is ACTIVE and it now accepts data. Attempts to send data to a delivery stream that is not in the ACTIVE state cause an exception. To check the state of a delivery stream, use DescribeDeliveryStream. A Kinesis Data Firehose delivery stream can be configured to receive records directly from providers using PutRecord or PutRecordBatch, or it can be configured to use an existing Kinesis stream as its source. To specify a Kinesis data stream as input, set the DeliveryStreamType parameter to KinesisStreamAsSource, and provide the Kinesis stream Amazon Resource Name (ARN) and role ARN in the KinesisStreamSourceConfiguration parameter. A delivery stream is configured with a single destination: Amazon S3, Amazon ES, Amazon Redshift, or Splunk. You must specify only one of the following destination configuration parameters: ExtendedS3DestinationConfiguration, S3DestinationConfiguration, ElasticsearchDestinationConfiguration, RedshiftDestinationConfiguration, or SplunkDestinationConfiguration. When you specify S3DestinationConfiguration, you can also provide the following optional values: BufferingHints, EncryptionConfiguration, and CompressionFormat. By default, if no BufferingHints value is provided, Kinesis Data Firehose buffers data up to 5 MB or for 5 minutes, whichever condition is satisfied first. BufferingHints is a hint, so there are some cases where the service cannot adhere to these conditions strictly. For example, record boundaries might be such that the size is a little over or under the configured buffering size. By default, no encryption is performed. We strongly recommend that you enable encryption to ensure secure data storage in Amazon S3. A few notes about Amazon Redshift as a destination:   An Amazon Redshift destination requires an S3 bucket as intermediate location. Kinesis Data Firehose first delivers data to Amazon S3 and then uses COPY syntax to load data into an Amazon Redshift table. This is specified in the RedshiftDestinationConfiguration.S3Configuration parameter.   The compression formats SNAPPY or ZIP cannot be specified in RedshiftDestinationConfiguration.S3Configuration because the Amazon Redshift COPY operation that reads from the S3 bucket doesn't support these compression formats.   We strongly recommend that you use the user name and password you provide exclusively with Kinesis Data Firehose, and that the permissions for the account are restricted for Amazon Redshift INSERT permissions.   Kinesis Data Firehose assumes the IAM role that is configured as part of the destination. The role should allow the Kinesis Data Firehose principal to assume the role, and the role should have permissions that allow the service to deliver the data. For more information, see Grant Kinesis Data Firehose Access to an Amazon S3 Destination in the Amazon Kinesis Data Firehose Developer Guide.
   */
  createDeliveryStream(callback?: (err: AWSError, data: Firehose.Types.CreateDeliveryStreamOutput) => void): Request<Firehose.Types.CreateDeliveryStreamOutput, AWSError>;
  /**
   * Deletes a delivery stream and its data. You can delete a delivery stream only if it is in ACTIVE or DELETING state, and not in the CREATING state. While the deletion request is in process, the delivery stream is in the DELETING state. To check the state of a delivery stream, use DescribeDeliveryStream. While the delivery stream is DELETING state, the service might continue to accept the records, but it doesn't make any guarantees with respect to delivering the data. Therefore, as a best practice, you should first stop any applications that are sending records before deleting a delivery stream.
   */
  deleteDeliveryStream(params: Firehose.Types.DeleteDeliveryStreamInput, callback?: (err: AWSError, data: Firehose.Types.DeleteDeliveryStreamOutput) => void): Request<Firehose.Types.DeleteDeliveryStreamOutput, AWSError>;
  /**
   * Deletes a delivery stream and its data. You can delete a delivery stream only if it is in ACTIVE or DELETING state, and not in the CREATING state. While the deletion request is in process, the delivery stream is in the DELETING state. To check the state of a delivery stream, use DescribeDeliveryStream. While the delivery stream is DELETING state, the service might continue to accept the records, but it doesn't make any guarantees with respect to delivering the data. Therefore, as a best practice, you should first stop any applications that are sending records before deleting a delivery stream.
   */
  deleteDeliveryStream(callback?: (err: AWSError, data: Firehose.Types.DeleteDeliveryStreamOutput) => void): Request<Firehose.Types.DeleteDeliveryStreamOutput, AWSError>;
  /**
   * Describes the specified delivery stream and gets the status. For example, after your delivery stream is created, call DescribeDeliveryStream to see whether the delivery stream is ACTIVE and therefore ready for data to be sent to it.
   */
  describeDeliveryStream(params: Firehose.Types.DescribeDeliveryStreamInput, callback?: (err: AWSError, data: Firehose.Types.DescribeDeliveryStreamOutput) => void): Request<Firehose.Types.DescribeDeliveryStreamOutput, AWSError>;
  /**
   * Describes the specified delivery stream and gets the status. For example, after your delivery stream is created, call DescribeDeliveryStream to see whether the delivery stream is ACTIVE and therefore ready for data to be sent to it.
   */
  describeDeliveryStream(callback?: (err: AWSError, data: Firehose.Types.DescribeDeliveryStreamOutput) => void): Request<Firehose.Types.DescribeDeliveryStreamOutput, AWSError>;
  /**
   * Lists your delivery streams in alphabetical order of their names. The number of delivery streams might be too large to return using a single call to ListDeliveryStreams. You can limit the number of delivery streams returned, using the Limit parameter. To determine whether there are more delivery streams to list, check the value of HasMoreDeliveryStreams in the output. If there are more delivery streams to list, you can request them by calling this operation again and setting the ExclusiveStartDeliveryStreamName parameter to the name of the last delivery stream returned in the last call.
   */
  listDeliveryStreams(params: Firehose.Types.ListDeliveryStreamsInput, callback?: (err: AWSError, data: Firehose.Types.ListDeliveryStreamsOutput) => void): Request<Firehose.Types.ListDeliveryStreamsOutput, AWSError>;
  /**
   * Lists your delivery streams in alphabetical order of their names. The number of delivery streams might be too large to return using a single call to ListDeliveryStreams. You can limit the number of delivery streams returned, using the Limit parameter. To determine whether there are more delivery streams to list, check the value of HasMoreDeliveryStreams in the output. If there are more delivery streams to list, you can request them by calling this operation again and setting the ExclusiveStartDeliveryStreamName parameter to the name of the last delivery stream returned in the last call.
   */
  listDeliveryStreams(callback?: (err: AWSError, data: Firehose.Types.ListDeliveryStreamsOutput) => void): Request<Firehose.Types.ListDeliveryStreamsOutput, AWSError>;
  /**
   * Lists the tags for the specified delivery stream. This operation has a limit of five transactions per second per account. 
   */
  listTagsForDeliveryStream(params: Firehose.Types.ListTagsForDeliveryStreamInput, callback?: (err: AWSError, data: Firehose.Types.ListTagsForDeliveryStreamOutput) => void): Request<Firehose.Types.ListTagsForDeliveryStreamOutput, AWSError>;
  /**
   * Lists the tags for the specified delivery stream. This operation has a limit of five transactions per second per account. 
   */
  listTagsForDeliveryStream(callback?: (err: AWSError, data: Firehose.Types.ListTagsForDeliveryStreamOutput) => void): Request<Firehose.Types.ListTagsForDeliveryStreamOutput, AWSError>;
  /**
   * Writes a single data record into an Amazon Kinesis Data Firehose delivery stream. To write multiple data records into a delivery stream, use PutRecordBatch. Applications using these operations are referred to as producers. By default, each delivery stream can take in up to 2,000 transactions per second, 5,000 records per second, or 5 MB per second. If you use PutRecord and PutRecordBatch, the limits are an aggregate across these two operations for each delivery stream. For more information about limits and how to request an increase, see Amazon Kinesis Data Firehose Limits.  You must specify the name of the delivery stream and the data record when using PutRecord. The data record consists of a data blob that can be up to 1,000 KB in size, and any kind of data. For example, it can be a segment from a log file, geographic location data, website clickstream data, and so on. Kinesis Data Firehose buffers records before delivering them to the destination. To disambiguate the data blobs at the destination, a common solution is to use delimiters in the data, such as a newline (\n) or some other character unique within the data. This allows the consumer application to parse individual data items when reading the data from the destination. The PutRecord operation returns a RecordId, which is a unique string assigned to each record. Producer applications can use this ID for purposes such as auditability and investigation. If the PutRecord operation throws a ServiceUnavailableException, back off and retry. If the exception persists, it is possible that the throughput limits have been exceeded for the delivery stream.  Data records sent to Kinesis Data Firehose are stored for 24 hours from the time they are added to a delivery stream as it tries to send the records to the destination. If the destination is unreachable for more than 24 hours, the data is no longer available.  Don't concatenate two or more base64 strings to form the data fields of your records. Instead, concatenate the raw data, then perform base64 encoding. 
   */
  putRecord(params: Firehose.Types.PutRecordInput, callback?: (err: AWSError, data: Firehose.Types.PutRecordOutput) => void): Request<Firehose.Types.PutRecordOutput, AWSError>;
  /**
   * Writes a single data record into an Amazon Kinesis Data Firehose delivery stream. To write multiple data records into a delivery stream, use PutRecordBatch. Applications using these operations are referred to as producers. By default, each delivery stream can take in up to 2,000 transactions per second, 5,000 records per second, or 5 MB per second. If you use PutRecord and PutRecordBatch, the limits are an aggregate across these two operations for each delivery stream. For more information about limits and how to request an increase, see Amazon Kinesis Data Firehose Limits.  You must specify the name of the delivery stream and the data record when using PutRecord. The data record consists of a data blob that can be up to 1,000 KB in size, and any kind of data. For example, it can be a segment from a log file, geographic location data, website clickstream data, and so on. Kinesis Data Firehose buffers records before delivering them to the destination. To disambiguate the data blobs at the destination, a common solution is to use delimiters in the data, such as a newline (\n) or some other character unique within the data. This allows the consumer application to parse individual data items when reading the data from the destination. The PutRecord operation returns a RecordId, which is a unique string assigned to each record. Producer applications can use this ID for purposes such as auditability and investigation. If the PutRecord operation throws a ServiceUnavailableException, back off and retry. If the exception persists, it is possible that the throughput limits have been exceeded for the delivery stream.  Data records sent to Kinesis Data Firehose are stored for 24 hours from the time they are added to a delivery stream as it tries to send the records to the destination. If the destination is unreachable for more than 24 hours, the data is no longer available.  Don't concatenate two or more base64 strings to form the data fields of your records. Instead, concatenate the raw data, then perform base64 encoding. 
   */
  putRecord(callback?: (err: AWSError, data: Firehose.Types.PutRecordOutput) => void): Request<Firehose.Types.PutRecordOutput, AWSError>;
  /**
   * Writes multiple data records into a delivery stream in a single call, which can achieve higher throughput per producer than when writing single records. To write single data records into a delivery stream, use PutRecord. Applications using these operations are referred to as producers. By default, each delivery stream can take in up to 2,000 transactions per second, 5,000 records per second, or 5 MB per second. If you use PutRecord and PutRecordBatch, the limits are an aggregate across these two operations for each delivery stream. For more information about limits, see Amazon Kinesis Data Firehose Limits. Each PutRecordBatch request supports up to 500 records. Each record in the request can be as large as 1,000 KB (before 64-bit encoding), up to a limit of 4 MB for the entire request. These limits cannot be changed. You must specify the name of the delivery stream and the data record when using PutRecord. The data record consists of a data blob that can be up to 1,000 KB in size, and any kind of data. For example, it could be a segment from a log file, geographic location data, website clickstream data, and so on. Kinesis Data Firehose buffers records before delivering them to the destination. To disambiguate the data blobs at the destination, a common solution is to use delimiters in the data, such as a newline (\n) or some other character unique within the data. This allows the consumer application to parse individual data items when reading the data from the destination. The PutRecordBatch response includes a count of failed records, FailedPutCount, and an array of responses, RequestResponses. Even if the PutRecordBatch call succeeds, the value of FailedPutCount may be greater than 0, indicating that there are records for which the operation didn't succeed. Each entry in the RequestResponses array provides additional information about the processed record. It directly correlates with a record in the request array using the same ordering, from the top to the bottom. The response array always includes the same number of records as the request array. RequestResponses includes both successfully and unsuccessfully processed records. Kinesis Data Firehose tries to process all records in each PutRecordBatch request. A single record failure does not stop the processing of subsequent records.  A successfully processed record includes a RecordId value, which is unique for the record. An unsuccessfully processed record includes ErrorCode and ErrorMessage values. ErrorCode reflects the type of error, and is one of the following values: ServiceUnavailableException or InternalFailure. ErrorMessage provides more detailed information about the error. If there is an internal server error or a timeout, the write might have completed or it might have failed. If FailedPutCount is greater than 0, retry the request, resending only those records that might have failed processing. This minimizes the possible duplicate records and also reduces the total bytes sent (and corresponding charges). We recommend that you handle any duplicates at the destination. If PutRecordBatch throws ServiceUnavailableException, back off and retry. If the exception persists, it is possible that the throughput limits have been exceeded for the delivery stream. Data records sent to Kinesis Data Firehose are stored for 24 hours from the time they are added to a delivery stream as it attempts to send the records to the destination. If the destination is unreachable for more than 24 hours, the data is no longer available.  Don't concatenate two or more base64 strings to form the data fields of your records. Instead, concatenate the raw data, then perform base64 encoding. 
   */
  putRecordBatch(params: Firehose.Types.PutRecordBatchInput, callback?: (err: AWSError, data: Firehose.Types.PutRecordBatchOutput) => void): Request<Firehose.Types.PutRecordBatchOutput, AWSError>;
  /**
   * Writes multiple data records into a delivery stream in a single call, which can achieve higher throughput per producer than when writing single records. To write single data records into a delivery stream, use PutRecord. Applications using these operations are referred to as producers. By default, each delivery stream can take in up to 2,000 transactions per second, 5,000 records per second, or 5 MB per second. If you use PutRecord and PutRecordBatch, the limits are an aggregate across these two operations for each delivery stream. For more information about limits, see Amazon Kinesis Data Firehose Limits. Each PutRecordBatch request supports up to 500 records. Each record in the request can be as large as 1,000 KB (before 64-bit encoding), up to a limit of 4 MB for the entire request. These limits cannot be changed. You must specify the name of the delivery stream and the data record when using PutRecord. The data record consists of a data blob that can be up to 1,000 KB in size, and any kind of data. For example, it could be a segment from a log file, geographic location data, website clickstream data, and so on. Kinesis Data Firehose buffers records before delivering them to the destination. To disambiguate the data blobs at the destination, a common solution is to use delimiters in the data, such as a newline (\n) or some other character unique within the data. This allows the consumer application to parse individual data items when reading the data from the destination. The PutRecordBatch response includes a count of failed records, FailedPutCount, and an array of responses, RequestResponses. Even if the PutRecordBatch call succeeds, the value of FailedPutCount may be greater than 0, indicating that there are records for which the operation didn't succeed. Each entry in the RequestResponses array provides additional information about the processed record. It directly correlates with a record in the request array using the same ordering, from the top to the bottom. The response array always includes the same number of records as the request array. RequestResponses includes both successfully and unsuccessfully processed records. Kinesis Data Firehose tries to process all records in each PutRecordBatch request. A single record failure does not stop the processing of subsequent records.  A successfully processed record includes a RecordId value, which is unique for the record. An unsuccessfully processed record includes ErrorCode and ErrorMessage values. ErrorCode reflects the type of error, and is one of the following values: ServiceUnavailableException or InternalFailure. ErrorMessage provides more detailed information about the error. If there is an internal server error or a timeout, the write might have completed or it might have failed. If FailedPutCount is greater than 0, retry the request, resending only those records that might have failed processing. This minimizes the possible duplicate records and also reduces the total bytes sent (and corresponding charges). We recommend that you handle any duplicates at the destination. If PutRecordBatch throws ServiceUnavailableException, back off and retry. If the exception persists, it is possible that the throughput limits have been exceeded for the delivery stream. Data records sent to Kinesis Data Firehose are stored for 24 hours from the time they are added to a delivery stream as it attempts to send the records to the destination. If the destination is unreachable for more than 24 hours, the data is no longer available.  Don't concatenate two or more base64 strings to form the data fields of your records. Instead, concatenate the raw data, then perform base64 encoding. 
   */
  putRecordBatch(callback?: (err: AWSError, data: Firehose.Types.PutRecordBatchOutput) => void): Request<Firehose.Types.PutRecordBatchOutput, AWSError>;
  /**
   * Enables server-side encryption (SSE) for the delivery stream.  This operation is asynchronous. It returns immediately. When you invoke it, Kinesis Data Firehose first sets the status of the stream to ENABLING, and then to ENABLED. You can continue to read and write data to your stream while its status is ENABLING, but the data is not encrypted. It can take up to 5 seconds after the encryption status changes to ENABLED before all records written to the delivery stream are encrypted. To find out whether a record or a batch of records was encrypted, check the response elements PutRecordOutput$Encrypted and PutRecordBatchOutput$Encrypted, respectively. To check the encryption state of a delivery stream, use DescribeDeliveryStream. You can only enable SSE for a delivery stream that uses DirectPut as its source.  The StartDeliveryStreamEncryption and StopDeliveryStreamEncryption operations have a combined limit of 25 calls per delivery stream per 24 hours. For example, you reach the limit if you call StartDeliveryStreamEncryption 13 times and StopDeliveryStreamEncryption 12 times for the same delivery stream in a 24-hour period.
   */
  startDeliveryStreamEncryption(params: Firehose.Types.StartDeliveryStreamEncryptionInput, callback?: (err: AWSError, data: Firehose.Types.StartDeliveryStreamEncryptionOutput) => void): Request<Firehose.Types.StartDeliveryStreamEncryptionOutput, AWSError>;
  /**
   * Enables server-side encryption (SSE) for the delivery stream.  This operation is asynchronous. It returns immediately. When you invoke it, Kinesis Data Firehose first sets the status of the stream to ENABLING, and then to ENABLED. You can continue to read and write data to your stream while its status is ENABLING, but the data is not encrypted. It can take up to 5 seconds after the encryption status changes to ENABLED before all records written to the delivery stream are encrypted. To find out whether a record or a batch of records was encrypted, check the response elements PutRecordOutput$Encrypted and PutRecordBatchOutput$Encrypted, respectively. To check the encryption state of a delivery stream, use DescribeDeliveryStream. You can only enable SSE for a delivery stream that uses DirectPut as its source.  The StartDeliveryStreamEncryption and StopDeliveryStreamEncryption operations have a combined limit of 25 calls per delivery stream per 24 hours. For example, you reach the limit if you call StartDeliveryStreamEncryption 13 times and StopDeliveryStreamEncryption 12 times for the same delivery stream in a 24-hour period.
   */
  startDeliveryStreamEncryption(callback?: (err: AWSError, data: Firehose.Types.StartDeliveryStreamEncryptionOutput) => void): Request<Firehose.Types.StartDeliveryStreamEncryptionOutput, AWSError>;
  /**
   * Disables server-side encryption (SSE) for the delivery stream.  This operation is asynchronous. It returns immediately. When you invoke it, Kinesis Data Firehose first sets the status of the stream to DISABLING, and then to DISABLED. You can continue to read and write data to your stream while its status is DISABLING. It can take up to 5 seconds after the encryption status changes to DISABLED before all records written to the delivery stream are no longer subject to encryption. To find out whether a record or a batch of records was encrypted, check the response elements PutRecordOutput$Encrypted and PutRecordBatchOutput$Encrypted, respectively. To check the encryption state of a delivery stream, use DescribeDeliveryStream.  The StartDeliveryStreamEncryption and StopDeliveryStreamEncryption operations have a combined limit of 25 calls per delivery stream per 24 hours. For example, you reach the limit if you call StartDeliveryStreamEncryption 13 times and StopDeliveryStreamEncryption 12 times for the same delivery stream in a 24-hour period.
   */
  stopDeliveryStreamEncryption(params: Firehose.Types.StopDeliveryStreamEncryptionInput, callback?: (err: AWSError, data: Firehose.Types.StopDeliveryStreamEncryptionOutput) => void): Request<Firehose.Types.StopDeliveryStreamEncryptionOutput, AWSError>;
  /**
   * Disables server-side encryption (SSE) for the delivery stream.  This operation is asynchronous. It returns immediately. When you invoke it, Kinesis Data Firehose first sets the status of the stream to DISABLING, and then to DISABLED. You can continue to read and write data to your stream while its status is DISABLING. It can take up to 5 seconds after the encryption status changes to DISABLED before all records written to the delivery stream are no longer subject to encryption. To find out whether a record or a batch of records was encrypted, check the response elements PutRecordOutput$Encrypted and PutRecordBatchOutput$Encrypted, respectively. To check the encryption state of a delivery stream, use DescribeDeliveryStream.  The StartDeliveryStreamEncryption and StopDeliveryStreamEncryption operations have a combined limit of 25 calls per delivery stream per 24 hours. For example, you reach the limit if you call StartDeliveryStreamEncryption 13 times and StopDeliveryStreamEncryption 12 times for the same delivery stream in a 24-hour period.
   */
  stopDeliveryStreamEncryption(callback?: (err: AWSError, data: Firehose.Types.StopDeliveryStreamEncryptionOutput) => void): Request<Firehose.Types.StopDeliveryStreamEncryptionOutput, AWSError>;
  /**
   * Adds or updates tags for the specified delivery stream. A tag is a key-value pair that you can define and assign to AWS resources. If you specify a tag that already exists, the tag value is replaced with the value that you specify in the request. Tags are metadata. For example, you can add friendly names and descriptions or other types of information that can help you distinguish the delivery stream. For more information about tags, see Using Cost Allocation Tags in the AWS Billing and Cost Management User Guide.  Each delivery stream can have up to 50 tags.  This operation has a limit of five transactions per second per account. 
   */
  tagDeliveryStream(params: Firehose.Types.TagDeliveryStreamInput, callback?: (err: AWSError, data: Firehose.Types.TagDeliveryStreamOutput) => void): Request<Firehose.Types.TagDeliveryStreamOutput, AWSError>;
  /**
   * Adds or updates tags for the specified delivery stream. A tag is a key-value pair that you can define and assign to AWS resources. If you specify a tag that already exists, the tag value is replaced with the value that you specify in the request. Tags are metadata. For example, you can add friendly names and descriptions or other types of information that can help you distinguish the delivery stream. For more information about tags, see Using Cost Allocation Tags in the AWS Billing and Cost Management User Guide.  Each delivery stream can have up to 50 tags.  This operation has a limit of five transactions per second per account. 
   */
  tagDeliveryStream(callback?: (err: AWSError, data: Firehose.Types.TagDeliveryStreamOutput) => void): Request<Firehose.Types.TagDeliveryStreamOutput, AWSError>;
  /**
   * Removes tags from the specified delivery stream. Removed tags are deleted, and you can't recover them after this operation successfully completes. If you specify a tag that doesn't exist, the operation ignores it. This operation has a limit of five transactions per second per account. 
   */
  untagDeliveryStream(params: Firehose.Types.UntagDeliveryStreamInput, callback?: (err: AWSError, data: Firehose.Types.UntagDeliveryStreamOutput) => void): Request<Firehose.Types.UntagDeliveryStreamOutput, AWSError>;
  /**
   * Removes tags from the specified delivery stream. Removed tags are deleted, and you can't recover them after this operation successfully completes. If you specify a tag that doesn't exist, the operation ignores it. This operation has a limit of five transactions per second per account. 
   */
  untagDeliveryStream(callback?: (err: AWSError, data: Firehose.Types.UntagDeliveryStreamOutput) => void): Request<Firehose.Types.UntagDeliveryStreamOutput, AWSError>;
  /**
   * Updates the specified destination of the specified delivery stream. Use this operation to change the destination type (for example, to replace the Amazon S3 destination with Amazon Redshift) or change the parameters associated with a destination (for example, to change the bucket name of the Amazon S3 destination). The update might not occur immediately. The target delivery stream remains active while the configurations are updated, so data writes to the delivery stream can continue during this process. The updated configurations are usually effective within a few minutes. Switching between Amazon ES and other services is not supported. For an Amazon ES destination, you can only update to another Amazon ES destination. If the destination type is the same, Kinesis Data Firehose merges the configuration parameters specified with the destination configuration that already exists on the delivery stream. If any of the parameters are not specified in the call, the existing values are retained. For example, in the Amazon S3 destination, if EncryptionConfiguration is not specified, then the existing EncryptionConfiguration is maintained on the destination. If the destination type is not the same, for example, changing the destination from Amazon S3 to Amazon Redshift, Kinesis Data Firehose does not merge any parameters. In this case, all parameters must be specified. Kinesis Data Firehose uses CurrentDeliveryStreamVersionId to avoid race conditions and conflicting merges. This is a required field, and the service updates the configuration only if the existing configuration has a version ID that matches. After the update is applied successfully, the version ID is updated, and can be retrieved using DescribeDeliveryStream. Use the new version ID to set CurrentDeliveryStreamVersionId in the next call.
   */
  updateDestination(params: Firehose.Types.UpdateDestinationInput, callback?: (err: AWSError, data: Firehose.Types.UpdateDestinationOutput) => void): Request<Firehose.Types.UpdateDestinationOutput, AWSError>;
  /**
   * Updates the specified destination of the specified delivery stream. Use this operation to change the destination type (for example, to replace the Amazon S3 destination with Amazon Redshift) or change the parameters associated with a destination (for example, to change the bucket name of the Amazon S3 destination). The update might not occur immediately. The target delivery stream remains active while the configurations are updated, so data writes to the delivery stream can continue during this process. The updated configurations are usually effective within a few minutes. Switching between Amazon ES and other services is not supported. For an Amazon ES destination, you can only update to another Amazon ES destination. If the destination type is the same, Kinesis Data Firehose merges the configuration parameters specified with the destination configuration that already exists on the delivery stream. If any of the parameters are not specified in the call, the existing values are retained. For example, in the Amazon S3 destination, if EncryptionConfiguration is not specified, then the existing EncryptionConfiguration is maintained on the destination. If the destination type is not the same, for example, changing the destination from Amazon S3 to Amazon Redshift, Kinesis Data Firehose does not merge any parameters. In this case, all parameters must be specified. Kinesis Data Firehose uses CurrentDeliveryStreamVersionId to avoid race conditions and conflicting merges. This is a required field, and the service updates the configuration only if the existing configuration has a version ID that matches. After the update is applied successfully, the version ID is updated, and can be retrieved using DescribeDeliveryStream. Use the new version ID to set CurrentDeliveryStreamVersionId in the next call.
   */
  updateDestination(callback?: (err: AWSError, data: Firehose.Types.UpdateDestinationOutput) => void): Request<Firehose.Types.UpdateDestinationOutput, AWSError>;
}
declare namespace Firehose {
  export type AWSKMSKeyARN = string;
  export type BlockSizeBytes = number;
  export type BooleanObject = boolean;
  export type BucketARN = string;
  export interface BufferingHints {
    /**
     * Buffer incoming data to the specified size, in MiBs, before delivering it to the destination. The default value is 5. This parameter is optional but if you specify a value for it, you must also specify a value for IntervalInSeconds, and vice versa. We recommend setting this parameter to a value greater than the amount of data you typically ingest into the delivery stream in 10 seconds. For example, if you typically ingest data at 1 MiB/sec, the value should be 10 MiB or higher.
     */
    SizeInMBs?: SizeInMBs;
    /**
     * Buffer incoming data for the specified period of time, in seconds, before delivering it to the destination. The default value is 300. This parameter is optional but if you specify a value for it, you must also specify a value for SizeInMBs, and vice versa.
     */
    IntervalInSeconds?: IntervalInSeconds;
  }
  export interface CloudWatchLoggingOptions {
    /**
     * Enables or disables CloudWatch logging.
     */
    Enabled?: BooleanObject;
    /**
     * The CloudWatch group name for logging. This value is required if CloudWatch logging is enabled.
     */
    LogGroupName?: LogGroupName;
    /**
     * The CloudWatch log stream name for logging. This value is required if CloudWatch logging is enabled.
     */
    LogStreamName?: LogStreamName;
  }
  export type ClusterJDBCURL = string;
  export type ColumnToJsonKeyMappings = {[key: string]: NonEmptyString};
  export type CompressionFormat = "UNCOMPRESSED"|"GZIP"|"ZIP"|"Snappy"|string;
  export interface CopyCommand {
    /**
     * The name of the target table. The table must already exist in the database.
     */
    DataTableName: DataTableName;
    /**
     * A comma-separated list of column names.
     */
    DataTableColumns?: DataTableColumns;
    /**
     * Optional parameters to use with the Amazon Redshift COPY command. For more information, see the "Optional Parameters" section of Amazon Redshift COPY command. Some possible examples that would apply to Kinesis Data Firehose are as follows:  delimiter '\t' lzop; - fields are delimited with "\t" (TAB character) and compressed using lzop.  delimiter '|' - fields are delimited with "|" (this is the default delimiter).  delimiter '|' escape - the delimiter should be escaped.  fixedwidth 'venueid:3,venuename:25,venuecity:12,venuestate:2,venueseats:6' - fields are fixed width in the source, with each width specified after every column in the table.  JSON 's3://mybucket/jsonpaths.txt' - data is in JSON format, and the path specified is the format of the data. For more examples, see Amazon Redshift COPY command examples.
     */
    CopyOptions?: CopyOptions;
  }
  export type CopyOptions = string;
  export interface CreateDeliveryStreamInput {
    /**
     * The name of the delivery stream. This name must be unique per AWS account in the same AWS Region. If the delivery streams are in different accounts or different Regions, you can have multiple delivery streams with the same name.
     */
    DeliveryStreamName: DeliveryStreamName;
    /**
     * The delivery stream type. This parameter can be one of the following values:    DirectPut: Provider applications access the delivery stream directly.    KinesisStreamAsSource: The delivery stream uses a Kinesis data stream as a source.  
     */
    DeliveryStreamType?: DeliveryStreamType;
    /**
     * When a Kinesis data stream is used as the source for the delivery stream, a KinesisStreamSourceConfiguration containing the Kinesis data stream Amazon Resource Name (ARN) and the role ARN for the source stream.
     */
    KinesisStreamSourceConfiguration?: KinesisStreamSourceConfiguration;
    /**
     * [Deprecated] The destination in Amazon S3. You can specify only one destination.
     */
    S3DestinationConfiguration?: S3DestinationConfiguration;
    /**
     * The destination in Amazon S3. You can specify only one destination.
     */
    ExtendedS3DestinationConfiguration?: ExtendedS3DestinationConfiguration;
    /**
     * The destination in Amazon Redshift. You can specify only one destination.
     */
    RedshiftDestinationConfiguration?: RedshiftDestinationConfiguration;
    /**
     * The destination in Amazon ES. You can specify only one destination.
     */
    ElasticsearchDestinationConfiguration?: ElasticsearchDestinationConfiguration;
    /**
     * The destination in Splunk. You can specify only one destination.
     */
    SplunkDestinationConfiguration?: SplunkDestinationConfiguration;
    /**
     * A set of tags to assign to the delivery stream. A tag is a key-value pair that you can define and assign to AWS resources. Tags are metadata. For example, you can add friendly names and descriptions or other types of information that can help you distinguish the delivery stream. For more information about tags, see Using Cost Allocation Tags in the AWS Billing and Cost Management User Guide. You can specify up to 50 tags when creating a delivery stream.
     */
    Tags?: TagDeliveryStreamInputTagList;
  }
  export interface CreateDeliveryStreamOutput {
    /**
     * The ARN of the delivery stream.
     */
    DeliveryStreamARN?: DeliveryStreamARN;
  }
  export type Data = Buffer|Uint8Array|Blob|string;
  export interface DataFormatConversionConfiguration {
    /**
     * Specifies the AWS Glue Data Catalog table that contains the column information.
     */
    SchemaConfiguration?: SchemaConfiguration;
    /**
     * Specifies the deserializer that you want Kinesis Data Firehose to use to convert the format of your data from JSON.
     */
    InputFormatConfiguration?: InputFormatConfiguration;
    /**
     * Specifies the serializer that you want Kinesis Data Firehose to use to convert the format of your data to the Parquet or ORC format.
     */
    OutputFormatConfiguration?: OutputFormatConfiguration;
    /**
     * Defaults to true. Set it to false if you want to disable format conversion while preserving the configuration details.
     */
    Enabled?: BooleanObject;
  }
  export type DataTableColumns = string;
  export type DataTableName = string;
  export interface DeleteDeliveryStreamInput {
    /**
     * The name of the delivery stream.
     */
    DeliveryStreamName: DeliveryStreamName;
  }
  export interface DeleteDeliveryStreamOutput {
  }
  export type DeliveryStartTimestamp = Date;
  export type DeliveryStreamARN = string;
  export interface DeliveryStreamDescription {
    /**
     * The name of the delivery stream.
     */
    DeliveryStreamName: DeliveryStreamName;
    /**
     * The Amazon Resource Name (ARN) of the delivery stream. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    DeliveryStreamARN: DeliveryStreamARN;
    /**
     * The status of the delivery stream.
     */
    DeliveryStreamStatus: DeliveryStreamStatus;
    /**
     * Indicates the server-side encryption (SSE) status for the delivery stream.
     */
    DeliveryStreamEncryptionConfiguration?: DeliveryStreamEncryptionConfiguration;
    /**
     * The delivery stream type. This can be one of the following values:    DirectPut: Provider applications access the delivery stream directly.    KinesisStreamAsSource: The delivery stream uses a Kinesis data stream as a source.  
     */
    DeliveryStreamType: DeliveryStreamType;
    /**
     * Each time the destination is updated for a delivery stream, the version ID is changed, and the current version ID is required when updating the destination. This is so that the service knows it is applying the changes to the correct version of the delivery stream.
     */
    VersionId: DeliveryStreamVersionId;
    /**
     * The date and time that the delivery stream was created.
     */
    CreateTimestamp?: Timestamp;
    /**
     * The date and time that the delivery stream was last updated.
     */
    LastUpdateTimestamp?: Timestamp;
    /**
     * If the DeliveryStreamType parameter is KinesisStreamAsSource, a SourceDescription object describing the source Kinesis data stream.
     */
    Source?: SourceDescription;
    /**
     * The destinations.
     */
    Destinations: DestinationDescriptionList;
    /**
     * Indicates whether there are more destinations available to list.
     */
    HasMoreDestinations: BooleanObject;
  }
  export interface DeliveryStreamEncryptionConfiguration {
    /**
     * For a full description of the different values of this status, see StartDeliveryStreamEncryption and StopDeliveryStreamEncryption.
     */
    Status?: DeliveryStreamEncryptionStatus;
  }
  export type DeliveryStreamEncryptionStatus = "ENABLED"|"ENABLING"|"DISABLED"|"DISABLING"|string;
  export type DeliveryStreamName = string;
  export type DeliveryStreamNameList = DeliveryStreamName[];
  export type DeliveryStreamStatus = "CREATING"|"DELETING"|"ACTIVE"|string;
  export type DeliveryStreamType = "DirectPut"|"KinesisStreamAsSource"|string;
  export type DeliveryStreamVersionId = string;
  export interface DescribeDeliveryStreamInput {
    /**
     * The name of the delivery stream.
     */
    DeliveryStreamName: DeliveryStreamName;
    /**
     * The limit on the number of destinations to return. You can have one destination per delivery stream.
     */
    Limit?: DescribeDeliveryStreamInputLimit;
    /**
     * The ID of the destination to start returning the destination information. Kinesis Data Firehose supports one destination per delivery stream.
     */
    ExclusiveStartDestinationId?: DestinationId;
  }
  export type DescribeDeliveryStreamInputLimit = number;
  export interface DescribeDeliveryStreamOutput {
    /**
     * Information about the delivery stream.
     */
    DeliveryStreamDescription: DeliveryStreamDescription;
  }
  export interface Deserializer {
    /**
     * The OpenX SerDe. Used by Kinesis Data Firehose for deserializing data, which means converting it from the JSON format in preparation for serializing it to the Parquet or ORC format. This is one of two deserializers you can choose, depending on which one offers the functionality you need. The other option is the native Hive / HCatalog JsonSerDe.
     */
    OpenXJsonSerDe?: OpenXJsonSerDe;
    /**
     * The native Hive / HCatalog JsonSerDe. Used by Kinesis Data Firehose for deserializing data, which means converting it from the JSON format in preparation for serializing it to the Parquet or ORC format. This is one of two deserializers you can choose, depending on which one offers the functionality you need. The other option is the OpenX SerDe.
     */
    HiveJsonSerDe?: HiveJsonSerDe;
  }
  export interface DestinationDescription {
    /**
     * The ID of the destination.
     */
    DestinationId: DestinationId;
    /**
     * [Deprecated] The destination in Amazon S3.
     */
    S3DestinationDescription?: S3DestinationDescription;
    /**
     * The destination in Amazon S3.
     */
    ExtendedS3DestinationDescription?: ExtendedS3DestinationDescription;
    /**
     * The destination in Amazon Redshift.
     */
    RedshiftDestinationDescription?: RedshiftDestinationDescription;
    /**
     * The destination in Amazon ES.
     */
    ElasticsearchDestinationDescription?: ElasticsearchDestinationDescription;
    /**
     * The destination in Splunk.
     */
    SplunkDestinationDescription?: SplunkDestinationDescription;
  }
  export type DestinationDescriptionList = DestinationDescription[];
  export type DestinationId = string;
  export interface ElasticsearchBufferingHints {
    /**
     * Buffer incoming data for the specified period of time, in seconds, before delivering it to the destination. The default value is 300 (5 minutes).
     */
    IntervalInSeconds?: ElasticsearchBufferingIntervalInSeconds;
    /**
     * Buffer incoming data to the specified size, in MBs, before delivering it to the destination. The default value is 5. We recommend setting this parameter to a value greater than the amount of data you typically ingest into the delivery stream in 10 seconds. For example, if you typically ingest data at 1 MB/sec, the value should be 10 MB or higher.
     */
    SizeInMBs?: ElasticsearchBufferingSizeInMBs;
  }
  export type ElasticsearchBufferingIntervalInSeconds = number;
  export type ElasticsearchBufferingSizeInMBs = number;
  export type ElasticsearchClusterEndpoint = string;
  export interface ElasticsearchDestinationConfiguration {
    /**
     * The Amazon Resource Name (ARN) of the IAM role to be assumed by Kinesis Data Firehose for calling the Amazon ES Configuration API and for indexing documents. For more information, see Grant Kinesis Data Firehose Access to an Amazon S3 Destination and Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN: RoleARN;
    /**
     * The ARN of the Amazon ES domain. The IAM role must have permissions for DescribeElasticsearchDomain, DescribeElasticsearchDomains, and DescribeElasticsearchDomainConfig after assuming the role specified in RoleARN. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces. Specify either ClusterEndpoint or DomainARN.
     */
    DomainARN?: ElasticsearchDomainARN;
    /**
     * The endpoint to use when communicating with the cluster. Specify either this ClusterEndpoint or the DomainARN field.
     */
    ClusterEndpoint?: ElasticsearchClusterEndpoint;
    /**
     * The Elasticsearch index name.
     */
    IndexName: ElasticsearchIndexName;
    /**
     * The Elasticsearch type name. For Elasticsearch 6.x, there can be only one type per index. If you try to specify a new type for an existing index that already has another type, Kinesis Data Firehose returns an error during run time. For Elasticsearch 7.x, don't specify a TypeName.
     */
    TypeName?: ElasticsearchTypeName;
    /**
     * The Elasticsearch index rotation period. Index rotation appends a timestamp to the IndexName to facilitate the expiration of old data. For more information, see Index Rotation for the Amazon ES Destination. The default value is OneDay.
     */
    IndexRotationPeriod?: ElasticsearchIndexRotationPeriod;
    /**
     * The buffering options. If no value is specified, the default values for ElasticsearchBufferingHints are used.
     */
    BufferingHints?: ElasticsearchBufferingHints;
    /**
     * The retry behavior in case Kinesis Data Firehose is unable to deliver documents to Amazon ES. The default value is 300 (5 minutes).
     */
    RetryOptions?: ElasticsearchRetryOptions;
    /**
     * Defines how documents should be delivered to Amazon S3. When it is set to FailedDocumentsOnly, Kinesis Data Firehose writes any documents that could not be indexed to the configured Amazon S3 destination, with elasticsearch-failed/ appended to the key prefix. When set to AllDocuments, Kinesis Data Firehose delivers all incoming records to Amazon S3, and also writes failed documents with elasticsearch-failed/ appended to the prefix. For more information, see Amazon S3 Backup for the Amazon ES Destination. Default value is FailedDocumentsOnly.
     */
    S3BackupMode?: ElasticsearchS3BackupMode;
    /**
     * The configuration for the backup Amazon S3 location.
     */
    S3Configuration: S3DestinationConfiguration;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The Amazon CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export interface ElasticsearchDestinationDescription {
    /**
     * The Amazon Resource Name (ARN) of the AWS credentials. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN?: RoleARN;
    /**
     * The ARN of the Amazon ES domain. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces. Kinesis Data Firehose uses either ClusterEndpoint or DomainARN to send data to Amazon ES.
     */
    DomainARN?: ElasticsearchDomainARN;
    /**
     * The endpoint to use when communicating with the cluster. Kinesis Data Firehose uses either this ClusterEndpoint or the DomainARN field to send data to Amazon ES.
     */
    ClusterEndpoint?: ElasticsearchClusterEndpoint;
    /**
     * The Elasticsearch index name.
     */
    IndexName?: ElasticsearchIndexName;
    /**
     * The Elasticsearch type name. This applies to Elasticsearch 6.x and lower versions. For Elasticsearch 7.x, there's no value for TypeName.
     */
    TypeName?: ElasticsearchTypeName;
    /**
     * The Elasticsearch index rotation period
     */
    IndexRotationPeriod?: ElasticsearchIndexRotationPeriod;
    /**
     * The buffering options.
     */
    BufferingHints?: ElasticsearchBufferingHints;
    /**
     * The Amazon ES retry options.
     */
    RetryOptions?: ElasticsearchRetryOptions;
    /**
     * The Amazon S3 backup mode.
     */
    S3BackupMode?: ElasticsearchS3BackupMode;
    /**
     * The Amazon S3 destination.
     */
    S3DestinationDescription?: S3DestinationDescription;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The Amazon CloudWatch logging options.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export interface ElasticsearchDestinationUpdate {
    /**
     * The Amazon Resource Name (ARN) of the IAM role to be assumed by Kinesis Data Firehose for calling the Amazon ES Configuration API and for indexing documents. For more information, see Grant Kinesis Data Firehose Access to an Amazon S3 Destination and Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN?: RoleARN;
    /**
     * The ARN of the Amazon ES domain. The IAM role must have permissions for DescribeElasticsearchDomain, DescribeElasticsearchDomains, and DescribeElasticsearchDomainConfig after assuming the IAM role specified in RoleARN. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces. Specify either ClusterEndpoint or DomainARN.
     */
    DomainARN?: ElasticsearchDomainARN;
    /**
     * The endpoint to use when communicating with the cluster. Specify either this ClusterEndpoint or the DomainARN field.
     */
    ClusterEndpoint?: ElasticsearchClusterEndpoint;
    /**
     * The Elasticsearch index name.
     */
    IndexName?: ElasticsearchIndexName;
    /**
     * The Elasticsearch type name. For Elasticsearch 6.x, there can be only one type per index. If you try to specify a new type for an existing index that already has another type, Kinesis Data Firehose returns an error during runtime. If you upgrade Elasticsearch from 6.x to 7.x and don’t update your delivery stream, Kinesis Data Firehose still delivers data to Elasticsearch with the old index name and type name. If you want to update your delivery stream with a new index name, provide an empty string for TypeName. 
     */
    TypeName?: ElasticsearchTypeName;
    /**
     * The Elasticsearch index rotation period. Index rotation appends a timestamp to IndexName to facilitate the expiration of old data. For more information, see Index Rotation for the Amazon ES Destination. Default value is OneDay.
     */
    IndexRotationPeriod?: ElasticsearchIndexRotationPeriod;
    /**
     * The buffering options. If no value is specified, ElasticsearchBufferingHints object default values are used. 
     */
    BufferingHints?: ElasticsearchBufferingHints;
    /**
     * The retry behavior in case Kinesis Data Firehose is unable to deliver documents to Amazon ES. The default value is 300 (5 minutes).
     */
    RetryOptions?: ElasticsearchRetryOptions;
    /**
     * The Amazon S3 destination.
     */
    S3Update?: S3DestinationUpdate;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export type ElasticsearchDomainARN = string;
  export type ElasticsearchIndexName = string;
  export type ElasticsearchIndexRotationPeriod = "NoRotation"|"OneHour"|"OneDay"|"OneWeek"|"OneMonth"|string;
  export type ElasticsearchRetryDurationInSeconds = number;
  export interface ElasticsearchRetryOptions {
    /**
     * After an initial failure to deliver to Amazon ES, the total amount of time during which Kinesis Data Firehose retries delivery (including the first attempt). After this time has elapsed, the failed documents are written to Amazon S3. Default value is 300 seconds (5 minutes). A value of 0 (zero) results in no retries.
     */
    DurationInSeconds?: ElasticsearchRetryDurationInSeconds;
  }
  export type ElasticsearchS3BackupMode = "FailedDocumentsOnly"|"AllDocuments"|string;
  export type ElasticsearchTypeName = string;
  export interface EncryptionConfiguration {
    /**
     * Specifically override existing encryption information to ensure that no encryption is used.
     */
    NoEncryptionConfig?: NoEncryptionConfig;
    /**
     * The encryption key.
     */
    KMSEncryptionConfig?: KMSEncryptionConfig;
  }
  export type ErrorCode = string;
  export type ErrorMessage = string;
  export type ErrorOutputPrefix = string;
  export interface ExtendedS3DestinationConfiguration {
    /**
     * The Amazon Resource Name (ARN) of the AWS credentials. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN: RoleARN;
    /**
     * The ARN of the S3 bucket. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    BucketARN: BucketARN;
    /**
     * The "YYYY/MM/DD/HH" time format prefix is automatically used for delivered Amazon S3 files. You can also specify a custom prefix, as described in Custom Prefixes for Amazon S3 Objects.
     */
    Prefix?: Prefix;
    /**
     * A prefix that Kinesis Data Firehose evaluates and adds to failed records before writing them to S3. This prefix appears immediately following the bucket name. For information about how to specify this prefix, see Custom Prefixes for Amazon S3 Objects.
     */
    ErrorOutputPrefix?: ErrorOutputPrefix;
    /**
     * The buffering option.
     */
    BufferingHints?: BufferingHints;
    /**
     * The compression format. If no value is specified, the default is UNCOMPRESSED.
     */
    CompressionFormat?: CompressionFormat;
    /**
     * The encryption configuration. If no value is specified, the default is no encryption.
     */
    EncryptionConfiguration?: EncryptionConfiguration;
    /**
     * The Amazon CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The Amazon S3 backup mode.
     */
    S3BackupMode?: S3BackupMode;
    /**
     * The configuration for backup in Amazon S3.
     */
    S3BackupConfiguration?: S3DestinationConfiguration;
    /**
     * The serializer, deserializer, and schema for converting data from the JSON format to the Parquet or ORC format before writing it to Amazon S3.
     */
    DataFormatConversionConfiguration?: DataFormatConversionConfiguration;
  }
  export interface ExtendedS3DestinationDescription {
    /**
     * The Amazon Resource Name (ARN) of the AWS credentials. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN: RoleARN;
    /**
     * The ARN of the S3 bucket. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    BucketARN: BucketARN;
    /**
     * The "YYYY/MM/DD/HH" time format prefix is automatically used for delivered Amazon S3 files. You can also specify a custom prefix, as described in Custom Prefixes for Amazon S3 Objects.
     */
    Prefix?: Prefix;
    /**
     * A prefix that Kinesis Data Firehose evaluates and adds to failed records before writing them to S3. This prefix appears immediately following the bucket name. For information about how to specify this prefix, see Custom Prefixes for Amazon S3 Objects.
     */
    ErrorOutputPrefix?: ErrorOutputPrefix;
    /**
     * The buffering option.
     */
    BufferingHints: BufferingHints;
    /**
     * The compression format. If no value is specified, the default is UNCOMPRESSED.
     */
    CompressionFormat: CompressionFormat;
    /**
     * The encryption configuration. If no value is specified, the default is no encryption.
     */
    EncryptionConfiguration: EncryptionConfiguration;
    /**
     * The Amazon CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The Amazon S3 backup mode.
     */
    S3BackupMode?: S3BackupMode;
    /**
     * The configuration for backup in Amazon S3.
     */
    S3BackupDescription?: S3DestinationDescription;
    /**
     * The serializer, deserializer, and schema for converting data from the JSON format to the Parquet or ORC format before writing it to Amazon S3.
     */
    DataFormatConversionConfiguration?: DataFormatConversionConfiguration;
  }
  export interface ExtendedS3DestinationUpdate {
    /**
     * The Amazon Resource Name (ARN) of the AWS credentials. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN?: RoleARN;
    /**
     * The ARN of the S3 bucket. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    BucketARN?: BucketARN;
    /**
     * The "YYYY/MM/DD/HH" time format prefix is automatically used for delivered Amazon S3 files. You can also specify a custom prefix, as described in Custom Prefixes for Amazon S3 Objects.
     */
    Prefix?: Prefix;
    /**
     * A prefix that Kinesis Data Firehose evaluates and adds to failed records before writing them to S3. This prefix appears immediately following the bucket name. For information about how to specify this prefix, see Custom Prefixes for Amazon S3 Objects.
     */
    ErrorOutputPrefix?: ErrorOutputPrefix;
    /**
     * The buffering option.
     */
    BufferingHints?: BufferingHints;
    /**
     * The compression format. If no value is specified, the default is UNCOMPRESSED. 
     */
    CompressionFormat?: CompressionFormat;
    /**
     * The encryption configuration. If no value is specified, the default is no encryption.
     */
    EncryptionConfiguration?: EncryptionConfiguration;
    /**
     * The Amazon CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * Enables or disables Amazon S3 backup mode.
     */
    S3BackupMode?: S3BackupMode;
    /**
     * The Amazon S3 destination for backup.
     */
    S3BackupUpdate?: S3DestinationUpdate;
    /**
     * The serializer, deserializer, and schema for converting data from the JSON format to the Parquet or ORC format before writing it to Amazon S3.
     */
    DataFormatConversionConfiguration?: DataFormatConversionConfiguration;
  }
  export type HECAcknowledgmentTimeoutInSeconds = number;
  export type HECEndpoint = string;
  export type HECEndpointType = "Raw"|"Event"|string;
  export type HECToken = string;
  export interface HiveJsonSerDe {
    /**
     * Indicates how you want Kinesis Data Firehose to parse the date and timestamps that may be present in your input data JSON. To specify these format strings, follow the pattern syntax of JodaTime's DateTimeFormat format strings. For more information, see Class DateTimeFormat. You can also use the special value millis to parse timestamps in epoch milliseconds. If you don't specify a format, Kinesis Data Firehose uses java.sql.Timestamp::valueOf by default.
     */
    TimestampFormats?: ListOfNonEmptyStrings;
  }
  export interface InputFormatConfiguration {
    /**
     * Specifies which deserializer to use. You can choose either the Apache Hive JSON SerDe or the OpenX JSON SerDe. If both are non-null, the server rejects the request.
     */
    Deserializer?: Deserializer;
  }
  export type IntervalInSeconds = number;
  export interface KMSEncryptionConfig {
    /**
     * The Amazon Resource Name (ARN) of the encryption key. Must belong to the same AWS Region as the destination Amazon S3 bucket. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    AWSKMSKeyARN: AWSKMSKeyARN;
  }
  export type KinesisStreamARN = string;
  export interface KinesisStreamSourceConfiguration {
    /**
     * The ARN of the source Kinesis data stream. For more information, see Amazon Kinesis Data Streams ARN Format.
     */
    KinesisStreamARN: KinesisStreamARN;
    /**
     * The ARN of the role that provides access to the source Kinesis data stream. For more information, see AWS Identity and Access Management (IAM) ARN Format.
     */
    RoleARN: RoleARN;
  }
  export interface KinesisStreamSourceDescription {
    /**
     * The Amazon Resource Name (ARN) of the source Kinesis data stream. For more information, see Amazon Kinesis Data Streams ARN Format.
     */
    KinesisStreamARN?: KinesisStreamARN;
    /**
     * The ARN of the role used by the source Kinesis data stream. For more information, see AWS Identity and Access Management (IAM) ARN Format.
     */
    RoleARN?: RoleARN;
    /**
     * Kinesis Data Firehose starts retrieving records from the Kinesis data stream starting with this timestamp.
     */
    DeliveryStartTimestamp?: DeliveryStartTimestamp;
  }
  export interface ListDeliveryStreamsInput {
    /**
     * The maximum number of delivery streams to list. The default value is 10.
     */
    Limit?: ListDeliveryStreamsInputLimit;
    /**
     * The delivery stream type. This can be one of the following values:    DirectPut: Provider applications access the delivery stream directly.    KinesisStreamAsSource: The delivery stream uses a Kinesis data stream as a source.   This parameter is optional. If this parameter is omitted, delivery streams of all types are returned.
     */
    DeliveryStreamType?: DeliveryStreamType;
    /**
     * The list of delivery streams returned by this call to ListDeliveryStreams will start with the delivery stream whose name comes alphabetically immediately after the name you specify in ExclusiveStartDeliveryStreamName.
     */
    ExclusiveStartDeliveryStreamName?: DeliveryStreamName;
  }
  export type ListDeliveryStreamsInputLimit = number;
  export interface ListDeliveryStreamsOutput {
    /**
     * The names of the delivery streams.
     */
    DeliveryStreamNames: DeliveryStreamNameList;
    /**
     * Indicates whether there are more delivery streams available to list.
     */
    HasMoreDeliveryStreams: BooleanObject;
  }
  export type ListOfNonEmptyStrings = NonEmptyString[];
  export type ListOfNonEmptyStringsWithoutWhitespace = NonEmptyStringWithoutWhitespace[];
  export interface ListTagsForDeliveryStreamInput {
    /**
     * The name of the delivery stream whose tags you want to list.
     */
    DeliveryStreamName: DeliveryStreamName;
    /**
     * The key to use as the starting point for the list of tags. If you set this parameter, ListTagsForDeliveryStream gets all tags that occur after ExclusiveStartTagKey.
     */
    ExclusiveStartTagKey?: TagKey;
    /**
     * The number of tags to return. If this number is less than the total number of tags associated with the delivery stream, HasMoreTags is set to true in the response. To list additional tags, set ExclusiveStartTagKey to the last key in the response. 
     */
    Limit?: ListTagsForDeliveryStreamInputLimit;
  }
  export type ListTagsForDeliveryStreamInputLimit = number;
  export interface ListTagsForDeliveryStreamOutput {
    /**
     * A list of tags associated with DeliveryStreamName, starting with the first tag after ExclusiveStartTagKey and up to the specified Limit.
     */
    Tags: ListTagsForDeliveryStreamOutputTagList;
    /**
     * If this is true in the response, more tags are available. To list the remaining tags, set ExclusiveStartTagKey to the key of the last tag returned and call ListTagsForDeliveryStream again.
     */
    HasMoreTags: BooleanObject;
  }
  export type ListTagsForDeliveryStreamOutputTagList = Tag[];
  export type LogGroupName = string;
  export type LogStreamName = string;
  export type NoEncryptionConfig = "NoEncryption"|string;
  export type NonEmptyString = string;
  export type NonEmptyStringWithoutWhitespace = string;
  export type NonNegativeIntegerObject = number;
  export interface OpenXJsonSerDe {
    /**
     * When set to true, specifies that the names of the keys include dots and that you want Kinesis Data Firehose to replace them with underscores. This is useful because Apache Hive does not allow dots in column names. For example, if the JSON contains a key whose name is "a.b", you can define the column name to be "a_b" when using this option. The default is false.
     */
    ConvertDotsInJsonKeysToUnderscores?: BooleanObject;
    /**
     * When set to true, which is the default, Kinesis Data Firehose converts JSON keys to lowercase before deserializing them.
     */
    CaseInsensitive?: BooleanObject;
    /**
     * Maps column names to JSON keys that aren't identical to the column names. This is useful when the JSON contains keys that are Hive keywords. For example, timestamp is a Hive keyword. If you have a JSON key named timestamp, set this parameter to {"ts": "timestamp"} to map this key to a column named ts.
     */
    ColumnToJsonKeyMappings?: ColumnToJsonKeyMappings;
  }
  export type OrcCompression = "NONE"|"ZLIB"|"SNAPPY"|string;
  export type OrcFormatVersion = "V0_11"|"V0_12"|string;
  export type OrcRowIndexStride = number;
  export interface OrcSerDe {
    /**
     * The number of bytes in each stripe. The default is 64 MiB and the minimum is 8 MiB.
     */
    StripeSizeBytes?: OrcStripeSizeBytes;
    /**
     * The Hadoop Distributed File System (HDFS) block size. This is useful if you intend to copy the data from Amazon S3 to HDFS before querying. The default is 256 MiB and the minimum is 64 MiB. Kinesis Data Firehose uses this value for padding calculations.
     */
    BlockSizeBytes?: BlockSizeBytes;
    /**
     * The number of rows between index entries. The default is 10,000 and the minimum is 1,000.
     */
    RowIndexStride?: OrcRowIndexStride;
    /**
     * Set this to true to indicate that you want stripes to be padded to the HDFS block boundaries. This is useful if you intend to copy the data from Amazon S3 to HDFS before querying. The default is false.
     */
    EnablePadding?: BooleanObject;
    /**
     * A number between 0 and 1 that defines the tolerance for block padding as a decimal fraction of stripe size. The default value is 0.05, which means 5 percent of stripe size. For the default values of 64 MiB ORC stripes and 256 MiB HDFS blocks, the default block padding tolerance of 5 percent reserves a maximum of 3.2 MiB for padding within the 256 MiB block. In such a case, if the available size within the block is more than 3.2 MiB, a new, smaller stripe is inserted to fit within that space. This ensures that no stripe crosses block boundaries and causes remote reads within a node-local task. Kinesis Data Firehose ignores this parameter when OrcSerDe$EnablePadding is false.
     */
    PaddingTolerance?: Proportion;
    /**
     * The compression code to use over data blocks. The default is SNAPPY.
     */
    Compression?: OrcCompression;
    /**
     * The column names for which you want Kinesis Data Firehose to create bloom filters. The default is null.
     */
    BloomFilterColumns?: ListOfNonEmptyStringsWithoutWhitespace;
    /**
     * The Bloom filter false positive probability (FPP). The lower the FPP, the bigger the Bloom filter. The default value is 0.05, the minimum is 0, and the maximum is 1.
     */
    BloomFilterFalsePositiveProbability?: Proportion;
    /**
     * Represents the fraction of the total number of non-null rows. To turn off dictionary encoding, set this fraction to a number that is less than the number of distinct keys in a dictionary. To always use dictionary encoding, set this threshold to 1.
     */
    DictionaryKeyThreshold?: Proportion;
    /**
     * The version of the file to write. The possible values are V0_11 and V0_12. The default is V0_12.
     */
    FormatVersion?: OrcFormatVersion;
  }
  export type OrcStripeSizeBytes = number;
  export interface OutputFormatConfiguration {
    /**
     * Specifies which serializer to use. You can choose either the ORC SerDe or the Parquet SerDe. If both are non-null, the server rejects the request.
     */
    Serializer?: Serializer;
  }
  export type ParquetCompression = "UNCOMPRESSED"|"GZIP"|"SNAPPY"|string;
  export type ParquetPageSizeBytes = number;
  export interface ParquetSerDe {
    /**
     * The Hadoop Distributed File System (HDFS) block size. This is useful if you intend to copy the data from Amazon S3 to HDFS before querying. The default is 256 MiB and the minimum is 64 MiB. Kinesis Data Firehose uses this value for padding calculations.
     */
    BlockSizeBytes?: BlockSizeBytes;
    /**
     * The Parquet page size. Column chunks are divided into pages. A page is conceptually an indivisible unit (in terms of compression and encoding). The minimum value is 64 KiB and the default is 1 MiB.
     */
    PageSizeBytes?: ParquetPageSizeBytes;
    /**
     * The compression code to use over data blocks. The possible values are UNCOMPRESSED, SNAPPY, and GZIP, with the default being SNAPPY. Use SNAPPY for higher decompression speed. Use GZIP if the compression ration is more important than speed.
     */
    Compression?: ParquetCompression;
    /**
     * Indicates whether to enable dictionary compression.
     */
    EnableDictionaryCompression?: BooleanObject;
    /**
     * The maximum amount of padding to apply. This is useful if you intend to copy the data from Amazon S3 to HDFS before querying. The default is 0.
     */
    MaxPaddingBytes?: NonNegativeIntegerObject;
    /**
     * Indicates the version of row format to output. The possible values are V1 and V2. The default is V1.
     */
    WriterVersion?: ParquetWriterVersion;
  }
  export type ParquetWriterVersion = "V1"|"V2"|string;
  export type Password = string;
  export type Prefix = string;
  export interface ProcessingConfiguration {
    /**
     * Enables or disables data processing.
     */
    Enabled?: BooleanObject;
    /**
     * The data processors.
     */
    Processors?: ProcessorList;
  }
  export interface Processor {
    /**
     * The type of processor.
     */
    Type: ProcessorType;
    /**
     * The processor parameters.
     */
    Parameters?: ProcessorParameterList;
  }
  export type ProcessorList = Processor[];
  export interface ProcessorParameter {
    /**
     * The name of the parameter.
     */
    ParameterName: ProcessorParameterName;
    /**
     * The parameter value.
     */
    ParameterValue: ProcessorParameterValue;
  }
  export type ProcessorParameterList = ProcessorParameter[];
  export type ProcessorParameterName = "LambdaArn"|"NumberOfRetries"|"RoleArn"|"BufferSizeInMBs"|"BufferIntervalInSeconds"|string;
  export type ProcessorParameterValue = string;
  export type ProcessorType = "Lambda"|string;
  export type Proportion = number;
  export interface PutRecordBatchInput {
    /**
     * The name of the delivery stream.
     */
    DeliveryStreamName: DeliveryStreamName;
    /**
     * One or more records.
     */
    Records: PutRecordBatchRequestEntryList;
  }
  export interface PutRecordBatchOutput {
    /**
     * The number of records that might have failed processing. This number might be greater than 0 even if the PutRecordBatch call succeeds. Check FailedPutCount to determine whether there are records that you need to resend.
     */
    FailedPutCount: NonNegativeIntegerObject;
    /**
     * Indicates whether server-side encryption (SSE) was enabled during this operation.
     */
    Encrypted?: BooleanObject;
    /**
     * The results array. For each record, the index of the response element is the same as the index used in the request array.
     */
    RequestResponses: PutRecordBatchResponseEntryList;
  }
  export type PutRecordBatchRequestEntryList = Record[];
  export interface PutRecordBatchResponseEntry {
    /**
     * The ID of the record.
     */
    RecordId?: PutResponseRecordId;
    /**
     * The error code for an individual record result.
     */
    ErrorCode?: ErrorCode;
    /**
     * The error message for an individual record result.
     */
    ErrorMessage?: ErrorMessage;
  }
  export type PutRecordBatchResponseEntryList = PutRecordBatchResponseEntry[];
  export interface PutRecordInput {
    /**
     * The name of the delivery stream.
     */
    DeliveryStreamName: DeliveryStreamName;
    /**
     * The record.
     */
    Record: Record;
  }
  export interface PutRecordOutput {
    /**
     * The ID of the record.
     */
    RecordId: PutResponseRecordId;
    /**
     * Indicates whether server-side encryption (SSE) was enabled during this operation.
     */
    Encrypted?: BooleanObject;
  }
  export type PutResponseRecordId = string;
  export interface Record {
    /**
     * The data blob, which is base64-encoded when the blob is serialized. The maximum size of the data blob, before base64-encoding, is 1,000 KiB.
     */
    Data: Data;
  }
  export interface RedshiftDestinationConfiguration {
    /**
     * The Amazon Resource Name (ARN) of the AWS credentials. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN: RoleARN;
    /**
     * The database connection string.
     */
    ClusterJDBCURL: ClusterJDBCURL;
    /**
     * The COPY command.
     */
    CopyCommand: CopyCommand;
    /**
     * The name of the user.
     */
    Username: Username;
    /**
     * The user password.
     */
    Password: Password;
    /**
     * The retry behavior in case Kinesis Data Firehose is unable to deliver documents to Amazon Redshift. Default value is 3600 (60 minutes).
     */
    RetryOptions?: RedshiftRetryOptions;
    /**
     * The configuration for the intermediate Amazon S3 location from which Amazon Redshift obtains data. Restrictions are described in the topic for CreateDeliveryStream. The compression formats SNAPPY or ZIP cannot be specified in RedshiftDestinationConfiguration.S3Configuration because the Amazon Redshift COPY operation that reads from the S3 bucket doesn't support these compression formats.
     */
    S3Configuration: S3DestinationConfiguration;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The Amazon S3 backup mode.
     */
    S3BackupMode?: RedshiftS3BackupMode;
    /**
     * The configuration for backup in Amazon S3.
     */
    S3BackupConfiguration?: S3DestinationConfiguration;
    /**
     * The CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export interface RedshiftDestinationDescription {
    /**
     * The Amazon Resource Name (ARN) of the AWS credentials. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN: RoleARN;
    /**
     * The database connection string.
     */
    ClusterJDBCURL: ClusterJDBCURL;
    /**
     * The COPY command.
     */
    CopyCommand: CopyCommand;
    /**
     * The name of the user.
     */
    Username: Username;
    /**
     * The retry behavior in case Kinesis Data Firehose is unable to deliver documents to Amazon Redshift. Default value is 3600 (60 minutes).
     */
    RetryOptions?: RedshiftRetryOptions;
    /**
     * The Amazon S3 destination.
     */
    S3DestinationDescription: S3DestinationDescription;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The Amazon S3 backup mode.
     */
    S3BackupMode?: RedshiftS3BackupMode;
    /**
     * The configuration for backup in Amazon S3.
     */
    S3BackupDescription?: S3DestinationDescription;
    /**
     * The Amazon CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export interface RedshiftDestinationUpdate {
    /**
     * The Amazon Resource Name (ARN) of the AWS credentials. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN?: RoleARN;
    /**
     * The database connection string.
     */
    ClusterJDBCURL?: ClusterJDBCURL;
    /**
     * The COPY command.
     */
    CopyCommand?: CopyCommand;
    /**
     * The name of the user.
     */
    Username?: Username;
    /**
     * The user password.
     */
    Password?: Password;
    /**
     * The retry behavior in case Kinesis Data Firehose is unable to deliver documents to Amazon Redshift. Default value is 3600 (60 minutes).
     */
    RetryOptions?: RedshiftRetryOptions;
    /**
     * The Amazon S3 destination. The compression formats SNAPPY or ZIP cannot be specified in RedshiftDestinationUpdate.S3Update because the Amazon Redshift COPY operation that reads from the S3 bucket doesn't support these compression formats.
     */
    S3Update?: S3DestinationUpdate;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The Amazon S3 backup mode.
     */
    S3BackupMode?: RedshiftS3BackupMode;
    /**
     * The Amazon S3 destination for backup.
     */
    S3BackupUpdate?: S3DestinationUpdate;
    /**
     * The Amazon CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export type RedshiftRetryDurationInSeconds = number;
  export interface RedshiftRetryOptions {
    /**
     * The length of time during which Kinesis Data Firehose retries delivery after a failure, starting from the initial request and including the first attempt. The default value is 3600 seconds (60 minutes). Kinesis Data Firehose does not retry if the value of DurationInSeconds is 0 (zero) or if the first delivery attempt takes longer than the current value.
     */
    DurationInSeconds?: RedshiftRetryDurationInSeconds;
  }
  export type RedshiftS3BackupMode = "Disabled"|"Enabled"|string;
  export type RoleARN = string;
  export type S3BackupMode = "Disabled"|"Enabled"|string;
  export interface S3DestinationConfiguration {
    /**
     * The Amazon Resource Name (ARN) of the AWS credentials. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN: RoleARN;
    /**
     * The ARN of the S3 bucket. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    BucketARN: BucketARN;
    /**
     * The "YYYY/MM/DD/HH" time format prefix is automatically used for delivered Amazon S3 files. You can also specify a custom prefix, as described in Custom Prefixes for Amazon S3 Objects.
     */
    Prefix?: Prefix;
    /**
     * A prefix that Kinesis Data Firehose evaluates and adds to failed records before writing them to S3. This prefix appears immediately following the bucket name. For information about how to specify this prefix, see Custom Prefixes for Amazon S3 Objects.
     */
    ErrorOutputPrefix?: ErrorOutputPrefix;
    /**
     * The buffering option. If no value is specified, BufferingHints object default values are used.
     */
    BufferingHints?: BufferingHints;
    /**
     * The compression format. If no value is specified, the default is UNCOMPRESSED. The compression formats SNAPPY or ZIP cannot be specified for Amazon Redshift destinations because they are not supported by the Amazon Redshift COPY operation that reads from the S3 bucket.
     */
    CompressionFormat?: CompressionFormat;
    /**
     * The encryption configuration. If no value is specified, the default is no encryption.
     */
    EncryptionConfiguration?: EncryptionConfiguration;
    /**
     * The CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export interface S3DestinationDescription {
    /**
     * The Amazon Resource Name (ARN) of the AWS credentials. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN: RoleARN;
    /**
     * The ARN of the S3 bucket. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    BucketARN: BucketARN;
    /**
     * The "YYYY/MM/DD/HH" time format prefix is automatically used for delivered Amazon S3 files. You can also specify a custom prefix, as described in Custom Prefixes for Amazon S3 Objects.
     */
    Prefix?: Prefix;
    /**
     * A prefix that Kinesis Data Firehose evaluates and adds to failed records before writing them to S3. This prefix appears immediately following the bucket name. For information about how to specify this prefix, see Custom Prefixes for Amazon S3 Objects.
     */
    ErrorOutputPrefix?: ErrorOutputPrefix;
    /**
     * The buffering option. If no value is specified, BufferingHints object default values are used.
     */
    BufferingHints: BufferingHints;
    /**
     * The compression format. If no value is specified, the default is UNCOMPRESSED.
     */
    CompressionFormat: CompressionFormat;
    /**
     * The encryption configuration. If no value is specified, the default is no encryption.
     */
    EncryptionConfiguration: EncryptionConfiguration;
    /**
     * The Amazon CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export interface S3DestinationUpdate {
    /**
     * The Amazon Resource Name (ARN) of the AWS credentials. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    RoleARN?: RoleARN;
    /**
     * The ARN of the S3 bucket. For more information, see Amazon Resource Names (ARNs) and AWS Service Namespaces.
     */
    BucketARN?: BucketARN;
    /**
     * The "YYYY/MM/DD/HH" time format prefix is automatically used for delivered Amazon S3 files. You can also specify a custom prefix, as described in Custom Prefixes for Amazon S3 Objects.
     */
    Prefix?: Prefix;
    /**
     * A prefix that Kinesis Data Firehose evaluates and adds to failed records before writing them to S3. This prefix appears immediately following the bucket name. For information about how to specify this prefix, see Custom Prefixes for Amazon S3 Objects.
     */
    ErrorOutputPrefix?: ErrorOutputPrefix;
    /**
     * The buffering option. If no value is specified, BufferingHints object default values are used.
     */
    BufferingHints?: BufferingHints;
    /**
     * The compression format. If no value is specified, the default is UNCOMPRESSED. The compression formats SNAPPY or ZIP cannot be specified for Amazon Redshift destinations because they are not supported by the Amazon Redshift COPY operation that reads from the S3 bucket.
     */
    CompressionFormat?: CompressionFormat;
    /**
     * The encryption configuration. If no value is specified, the default is no encryption.
     */
    EncryptionConfiguration?: EncryptionConfiguration;
    /**
     * The CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export interface SchemaConfiguration {
    /**
     * The role that Kinesis Data Firehose can use to access AWS Glue. This role must be in the same account you use for Kinesis Data Firehose. Cross-account roles aren't allowed.
     */
    RoleARN?: NonEmptyStringWithoutWhitespace;
    /**
     * The ID of the AWS Glue Data Catalog. If you don't supply this, the AWS account ID is used by default.
     */
    CatalogId?: NonEmptyStringWithoutWhitespace;
    /**
     * Specifies the name of the AWS Glue database that contains the schema for the output data.
     */
    DatabaseName?: NonEmptyStringWithoutWhitespace;
    /**
     * Specifies the AWS Glue table that contains the column information that constitutes your data schema.
     */
    TableName?: NonEmptyStringWithoutWhitespace;
    /**
     * If you don't specify an AWS Region, the default is the current Region.
     */
    Region?: NonEmptyStringWithoutWhitespace;
    /**
     * Specifies the table version for the output data schema. If you don't specify this version ID, or if you set it to LATEST, Kinesis Data Firehose uses the most recent version. This means that any updates to the table are automatically picked up.
     */
    VersionId?: NonEmptyStringWithoutWhitespace;
  }
  export interface Serializer {
    /**
     * A serializer to use for converting data to the Parquet format before storing it in Amazon S3. For more information, see Apache Parquet.
     */
    ParquetSerDe?: ParquetSerDe;
    /**
     * A serializer to use for converting data to the ORC format before storing it in Amazon S3. For more information, see Apache ORC.
     */
    OrcSerDe?: OrcSerDe;
  }
  export type SizeInMBs = number;
  export interface SourceDescription {
    /**
     * The KinesisStreamSourceDescription value for the source Kinesis data stream.
     */
    KinesisStreamSourceDescription?: KinesisStreamSourceDescription;
  }
  export interface SplunkDestinationConfiguration {
    /**
     * The HTTP Event Collector (HEC) endpoint to which Kinesis Data Firehose sends your data.
     */
    HECEndpoint: HECEndpoint;
    /**
     * This type can be either "Raw" or "Event."
     */
    HECEndpointType: HECEndpointType;
    /**
     * This is a GUID that you obtain from your Splunk cluster when you create a new HEC endpoint.
     */
    HECToken: HECToken;
    /**
     * The amount of time that Kinesis Data Firehose waits to receive an acknowledgment from Splunk after it sends it data. At the end of the timeout period, Kinesis Data Firehose either tries to send the data again or considers it an error, based on your retry settings.
     */
    HECAcknowledgmentTimeoutInSeconds?: HECAcknowledgmentTimeoutInSeconds;
    /**
     * The retry behavior in case Kinesis Data Firehose is unable to deliver data to Splunk, or if it doesn't receive an acknowledgment of receipt from Splunk.
     */
    RetryOptions?: SplunkRetryOptions;
    /**
     * Defines how documents should be delivered to Amazon S3. When set to FailedDocumentsOnly, Kinesis Data Firehose writes any data that could not be indexed to the configured Amazon S3 destination. When set to AllDocuments, Kinesis Data Firehose delivers all incoming records to Amazon S3, and also writes failed documents to Amazon S3. Default value is FailedDocumentsOnly. 
     */
    S3BackupMode?: SplunkS3BackupMode;
    /**
     * The configuration for the backup Amazon S3 location.
     */
    S3Configuration: S3DestinationConfiguration;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The Amazon CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export interface SplunkDestinationDescription {
    /**
     * The HTTP Event Collector (HEC) endpoint to which Kinesis Data Firehose sends your data.
     */
    HECEndpoint?: HECEndpoint;
    /**
     * This type can be either "Raw" or "Event."
     */
    HECEndpointType?: HECEndpointType;
    /**
     * A GUID you obtain from your Splunk cluster when you create a new HEC endpoint.
     */
    HECToken?: HECToken;
    /**
     * The amount of time that Kinesis Data Firehose waits to receive an acknowledgment from Splunk after it sends it data. At the end of the timeout period, Kinesis Data Firehose either tries to send the data again or considers it an error, based on your retry settings.
     */
    HECAcknowledgmentTimeoutInSeconds?: HECAcknowledgmentTimeoutInSeconds;
    /**
     * The retry behavior in case Kinesis Data Firehose is unable to deliver data to Splunk or if it doesn't receive an acknowledgment of receipt from Splunk.
     */
    RetryOptions?: SplunkRetryOptions;
    /**
     * Defines how documents should be delivered to Amazon S3. When set to FailedDocumentsOnly, Kinesis Data Firehose writes any data that could not be indexed to the configured Amazon S3 destination. When set to AllDocuments, Kinesis Data Firehose delivers all incoming records to Amazon S3, and also writes failed documents to Amazon S3. Default value is FailedDocumentsOnly. 
     */
    S3BackupMode?: SplunkS3BackupMode;
    /**
     * The Amazon S3 destination.&gt;
     */
    S3DestinationDescription?: S3DestinationDescription;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The Amazon CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export interface SplunkDestinationUpdate {
    /**
     * The HTTP Event Collector (HEC) endpoint to which Kinesis Data Firehose sends your data.
     */
    HECEndpoint?: HECEndpoint;
    /**
     * This type can be either "Raw" or "Event."
     */
    HECEndpointType?: HECEndpointType;
    /**
     * A GUID that you obtain from your Splunk cluster when you create a new HEC endpoint.
     */
    HECToken?: HECToken;
    /**
     * The amount of time that Kinesis Data Firehose waits to receive an acknowledgment from Splunk after it sends data. At the end of the timeout period, Kinesis Data Firehose either tries to send the data again or considers it an error, based on your retry settings.
     */
    HECAcknowledgmentTimeoutInSeconds?: HECAcknowledgmentTimeoutInSeconds;
    /**
     * The retry behavior in case Kinesis Data Firehose is unable to deliver data to Splunk or if it doesn't receive an acknowledgment of receipt from Splunk.
     */
    RetryOptions?: SplunkRetryOptions;
    /**
     * Defines how documents should be delivered to Amazon S3. When set to FailedDocumentsOnly, Kinesis Data Firehose writes any data that could not be indexed to the configured Amazon S3 destination. When set to AllDocuments, Kinesis Data Firehose delivers all incoming records to Amazon S3, and also writes failed documents to Amazon S3. Default value is FailedDocumentsOnly. 
     */
    S3BackupMode?: SplunkS3BackupMode;
    /**
     * Your update to the configuration of the backup Amazon S3 location.
     */
    S3Update?: S3DestinationUpdate;
    /**
     * The data processing configuration.
     */
    ProcessingConfiguration?: ProcessingConfiguration;
    /**
     * The Amazon CloudWatch logging options for your delivery stream.
     */
    CloudWatchLoggingOptions?: CloudWatchLoggingOptions;
  }
  export type SplunkRetryDurationInSeconds = number;
  export interface SplunkRetryOptions {
    /**
     * The total amount of time that Kinesis Data Firehose spends on retries. This duration starts after the initial attempt to send data to Splunk fails. It doesn't include the periods during which Kinesis Data Firehose waits for acknowledgment from Splunk after each attempt.
     */
    DurationInSeconds?: SplunkRetryDurationInSeconds;
  }
  export type SplunkS3BackupMode = "FailedEventsOnly"|"AllEvents"|string;
  export interface StartDeliveryStreamEncryptionInput {
    /**
     * The name of the delivery stream for which you want to enable server-side encryption (SSE).
     */
    DeliveryStreamName: DeliveryStreamName;
  }
  export interface StartDeliveryStreamEncryptionOutput {
  }
  export interface StopDeliveryStreamEncryptionInput {
    /**
     * The name of the delivery stream for which you want to disable server-side encryption (SSE).
     */
    DeliveryStreamName: DeliveryStreamName;
  }
  export interface StopDeliveryStreamEncryptionOutput {
  }
  export interface Tag {
    /**
     * A unique identifier for the tag. Maximum length: 128 characters. Valid characters: Unicode letters, digits, white space, _ . / = + - % @
     */
    Key: TagKey;
    /**
     * An optional string, which you can use to describe or define the tag. Maximum length: 256 characters. Valid characters: Unicode letters, digits, white space, _ . / = + - % @
     */
    Value?: TagValue;
  }
  export interface TagDeliveryStreamInput {
    /**
     * The name of the delivery stream to which you want to add the tags.
     */
    DeliveryStreamName: DeliveryStreamName;
    /**
     * A set of key-value pairs to use to create the tags.
     */
    Tags: TagDeliveryStreamInputTagList;
  }
  export type TagDeliveryStreamInputTagList = Tag[];
  export interface TagDeliveryStreamOutput {
  }
  export type TagKey = string;
  export type TagKeyList = TagKey[];
  export type TagValue = string;
  export type Timestamp = Date;
  export interface UntagDeliveryStreamInput {
    /**
     * The name of the delivery stream.
     */
    DeliveryStreamName: DeliveryStreamName;
    /**
     * A list of tag keys. Each corresponding tag is removed from the delivery stream.
     */
    TagKeys: TagKeyList;
  }
  export interface UntagDeliveryStreamOutput {
  }
  export interface UpdateDestinationInput {
    /**
     * The name of the delivery stream.
     */
    DeliveryStreamName: DeliveryStreamName;
    /**
     * Obtain this value from the VersionId result of DeliveryStreamDescription. This value is required, and helps the service perform conditional operations. For example, if there is an interleaving update and this value is null, then the update destination fails. After the update is successful, the VersionId value is updated. The service then performs a merge of the old configuration with the new configuration.
     */
    CurrentDeliveryStreamVersionId: DeliveryStreamVersionId;
    /**
     * The ID of the destination.
     */
    DestinationId: DestinationId;
    /**
     * [Deprecated] Describes an update for a destination in Amazon S3.
     */
    S3DestinationUpdate?: S3DestinationUpdate;
    /**
     * Describes an update for a destination in Amazon S3.
     */
    ExtendedS3DestinationUpdate?: ExtendedS3DestinationUpdate;
    /**
     * Describes an update for a destination in Amazon Redshift.
     */
    RedshiftDestinationUpdate?: RedshiftDestinationUpdate;
    /**
     * Describes an update for a destination in Amazon ES.
     */
    ElasticsearchDestinationUpdate?: ElasticsearchDestinationUpdate;
    /**
     * Describes an update for a destination in Splunk.
     */
    SplunkDestinationUpdate?: SplunkDestinationUpdate;
  }
  export interface UpdateDestinationOutput {
  }
  export type Username = string;
  /**
   * A string in YYYY-MM-DD format that represents the latest possible API version that can be used in this service. Specify 'latest' to use the latest possible version.
   */
  export type apiVersion = "2015-08-04"|"latest"|string;
  export interface ClientApiVersions {
    /**
     * A string in YYYY-MM-DD format that represents the latest possible API version that can be used in this service. Specify 'latest' to use the latest possible version.
     */
    apiVersion?: apiVersion;
  }
  export type ClientConfiguration = ServiceConfigurationOptions & ClientApiVersions;
  /**
   * Contains interfaces for use with the Firehose client.
   */
  export import Types = Firehose;
}
export = Firehose;