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
//! Kafka Client - A mid-level abstraction for a kafka cluster //! allowing building higher level constructs. //! //! The entry point into this module is `KafkaClient` obtained by a //! call to `KafkaClient::new()`. use std::collections::hash_map::HashMap; use std::io::Cursor; use std::iter::Iterator; use std::mem; use openssl::ssl::SslContext; // pub re-export pub use compression::Compression; pub use utils::PartitionOffset; pub use utils::TopicPartitionOffset; use codecs::{ToByte, FromByte}; use connection::KafkaConnection; use error::{Result, Error, KafkaCode}; use protocol::{self, ResponseParser}; pub mod metadata; mod state; // ~ re-export (only) certain types from the protocol::fetch module as // 'client::fetch'. pub mod fetch { //! A representation of fetched messages from Kafka. pub use protocol::fetch::{Data, Message, Partition, Response, Topic}; } const CLIENTID: &'static str = "kafka-rust"; const DEFAULT_SO_TIMEOUT_SECS: i32 = 120; // socket read, write timeout seconds /// The default value for `KafkaClient::set_compression(..)` pub const DEFAULT_COMPRESSION: Compression = Compression::NONE; /// The default value for `KafkaClient::set_fetch_max_wait_time(..)` pub const DEFAULT_FETCH_MAX_WAIT_TIME: i32 = 100; // milliseconds /// The default value for `KafkaClient::set_fetch_min_bytes(..)` pub const DEFAULT_FETCH_MIN_BYTES: i32 = 4096; /// The default value for `KafkaClient::set_fetch_max_bytes(..)` pub const DEFAULT_FETCH_MAX_BYTES_PER_PARTITION: i32 = 32 * 1024; /// The default value for `KafkaClient::set_fetch_crc_validation(..)` pub const DEFAULT_FETCH_CRC_VALIDATION: bool = true; /// Client struct keeping track of brokers and topic metadata. /// /// Implements methods described by the [Kafka Protocol](https://cwiki.apache.org/confluence/display/KAFKA/A+Guide+To+The+Kafka+Protocol). /// /// You will have to load metadata before making any other request. #[derive(Debug)] pub struct KafkaClient { // ~ this kafka client configuration config: ClientConfig, // ~ a pool of re-usable connections to kafka brokers conn_pool: ConnectionPool, // ~ the current state of this client state: state::ClientState, } #[derive(Debug)] struct ClientConfig { client_id: String, hosts: Vec<String>, // ~ compression to use when sending messages compression: Compression, // ~ these are the defaults when fetching messages for details // refer to the kafka wire protocol fetch_max_wait_time: i32, fetch_min_bytes: i32, fetch_max_bytes_per_partition: i32, fetch_crc_validation: bool, } #[derive(Debug)] struct ConnectionPool { conns: HashMap<String, KafkaConnection>, timeout: i32, security_config: Option<SecurityConfig> } impl ConnectionPool { fn new(timeout: i32, security: Option<SecurityConfig>) -> ConnectionPool { ConnectionPool { conns: HashMap::new(), timeout: timeout, security_config: security, } } fn get_conn<'a>(&'a mut self, host: &str) -> Result<&'a mut KafkaConnection> { if let Some(conn) = self.conns.get_mut(host) { // ~ decouple the lifetimes to make borrowck happy; this // is actually safe since we're immediatelly returning // this, so the follow up code is not affected here (this // method is no longer recursive). return Ok(unsafe { mem::transmute(conn) }); } self.conns.insert(host.to_owned(), try!(KafkaConnection::new( host, self.timeout, self.security_config.as_ref().map(|c| &c.0)))); Ok(self.conns.get_mut(host).unwrap()) } } // -------------------------------------------------------------------- /// Possible values when querying a topic's offset. /// See `KafkaClient::fetch_offsets`. #[derive(Debug, Copy, Clone)] pub enum FetchOffset { /// Receive the earliest available offset. Earliest, /// Receive the latest offset. Latest, /// Used to ask for all messages before a certain time (ms); unix /// timestamp in milliseconds. See also /// https://cwiki.apache.org/confluence/display/KAFKA/Writing+a+Driver+for+Kafka#WritingaDriverforKafka-Offsets ByTime(i64), } impl FetchOffset { fn to_kafka_value(&self) -> i64 { match *self { FetchOffset::Earliest => -2, FetchOffset::Latest => -1, FetchOffset::ByTime(n) => n, } } } // -------------------------------------------------------------------- /// Data point identifying a topic partition to fetch a group's offset /// for. See `KafkaClient::fetch_group_offsets`. #[derive(Debug)] pub struct FetchGroupOffset<'a> { /// The topic to fetch the group offset for pub topic: &'a str, /// The partition to fetch the group offset for pub partition: i32, } impl<'a> FetchGroupOffset<'a> { #[inline] pub fn new(topic: &'a str, partition: i32) -> Self { FetchGroupOffset { topic: topic, partition: partition } } } impl<'a> AsRef<FetchGroupOffset<'a>> for FetchGroupOffset<'a> { fn as_ref(&self) -> &Self { self } } // -------------------------------------------------------------------- /// Data point identifying a particular topic partition offset to be /// commited. /// See `KafkaClient::commit_offsets`. #[derive(Debug)] pub struct CommitOffset<'a> { /// The offset to be committed pub offset: i64, /// The topic to commit the offset for pub topic: &'a str, /// The partition to commit the offset for pub partition: i32, } impl<'a> CommitOffset<'a> { pub fn new(topic: &'a str, partition: i32, offset: i64) -> Self { CommitOffset { topic: topic, partition: partition, offset: offset } } } impl<'a> AsRef<CommitOffset<'a>> for CommitOffset<'a> { fn as_ref(&self) -> &Self { self } } // -------------------------------------------------------------------- /// Message data to be sent/produced to a particular topic partition. /// See `KafkaClient::produce_messages` and `Producer::send`. #[derive(Debug)] pub struct ProduceMessage<'a, 'b> { /// The "key" data of this message. pub key: Option<&'b [u8]>, /// The "value" data of this message. pub value: Option<&'b [u8]>, /// The topic to produce this message to. pub topic: &'a str, /// The partition (of the corresponding topic) to produce this /// message to. pub partition: i32, } impl<'a, 'b> AsRef<ProduceMessage<'a, 'b>> for ProduceMessage<'a, 'b> { fn as_ref(&self) -> &Self { self } } impl<'a, 'b> ProduceMessage<'a, 'b> { /// A convenient constructor method to create a new produce /// message with all attributes specified. pub fn new(topic: &'a str, partition: i32, key: Option<&'b [u8]>, value: Option<&'b [u8]>) -> Self { ProduceMessage { key: key, value: value, topic: topic, partition: partition } } } // -------------------------------------------------------------------- /// Partition related request data for fetching messages. /// See `KafkaClient::fetch_messages`. #[derive(Debug)] pub struct FetchPartition<'a> { /// The topic to fetch messages from. pub topic: &'a str, /// The offset as of which to fetch messages. pub offset: i64, /// The partition to fetch messasges from. pub partition: i32, /// Specifies the max. amount of data to fetch (for this /// partition.) This implicitely defines the biggest message the /// client can accept. If this value is too small, no messages /// can be delivered. Setting this size should be in sync with /// the producers to the partition. /// /// Zero or negative values are treated as "unspecified". pub max_bytes: i32, } impl<'a> FetchPartition<'a> { /// Creates a new "fetch messages" request structure with an /// unspecified `max_bytes`. pub fn new(topic: &'a str, partition: i32, offset: i64) -> Self { FetchPartition { topic: topic, partition: partition, offset: offset, max_bytes: -1, } } /// Sets the `max_bytes` value for the "fetch messages" request. pub fn with_max_bytes(mut self, max_bytes: i32) -> Self { self.max_bytes = max_bytes; self } } impl<'a> AsRef<FetchPartition<'a>> for FetchPartition<'a> { fn as_ref(&self) -> &Self { self } } // -------------------------------------------------------------------- /// This will be expanded in the future. See #51. #[derive(Debug)] pub struct SecurityConfig(SslContext); impl SecurityConfig { /// In the future this will also support a kerbos via #51. pub fn new(ssl: SslContext) -> SecurityConfig { SecurityConfig(ssl) } } // -------------------------------------------------------------------- impl KafkaClient { /// Creates a new instance of KafkaClient. Before being able to /// successfully use the new client, you'll have to load metadata. /// /// # Examples /// /// ```no_run /// let mut client = kafka::client::KafkaClient::new(vec!("localhost:9092".to_owned())); /// client.load_metadata_all().unwrap(); /// ``` pub fn new(hosts: Vec<String>) -> KafkaClient { KafkaClient { config: ClientConfig { client_id: CLIENTID.to_owned(), hosts: hosts, compression: DEFAULT_COMPRESSION, fetch_max_wait_time: DEFAULT_FETCH_MAX_WAIT_TIME, fetch_min_bytes: DEFAULT_FETCH_MIN_BYTES, fetch_max_bytes_per_partition: DEFAULT_FETCH_MAX_BYTES_PER_PARTITION, fetch_crc_validation: DEFAULT_FETCH_CRC_VALIDATION, }, conn_pool: ConnectionPool::new(DEFAULT_SO_TIMEOUT_SECS, None), state: state::ClientState::new(), } } /// Creates a new secure instance of KafkaClient. Before being able to /// successfully use the new client, you'll have to load metadata. /// /// # Examples /// /// ```no_run /// extern crate openssl; /// extern crate kafka; /// /// use openssl::ssl::{Ssl, SslContext, SslStream, SslMethod, SSL_VERIFY_PEER}; /// use openssl::x509::X509FileType; /// use kafka::client::{KafkaClient, SecurityConfig}; /// /// fn main() { /// let (key, cert) = ("client.key".to_string(), "client.crt".to_string()); /// /// // OpenSSL offers a variety of complex configurations. Here is an example: /// let mut ctx = SslContext::new(SslMethod::Sslv23).unwrap(); /// ctx.set_cipher_list("DEFAULT").unwrap(); /// ctx.set_certificate_file(&cert, X509FileType::PEM).unwrap(); /// ctx.set_private_key_file(&key, X509FileType::PEM).unwrap(); /// ctx.set_default_verify_paths().unwrap(); /// ctx.set_verify(SSL_VERIFY_PEER, None); /// /// let mut client = KafkaClient::new_secure(vec!("localhost:9092".to_owned()), /// SecurityConfig::new(ctx)); /// client.load_metadata_all().unwrap(); /// } /// ``` /// /// See also `KafkaClient::load_metadatata_all` and /// `KafkaClient::load_metadata` methods, the creates /// [openssl](https://crates.io/crates/openssl) /// and [openssl_verify](https://crates.io/crates/openssl-verify), /// as well as [Kafka's documentation](https://kafka.apache.org/documentation.html#security_ssl). pub fn new_secure(hosts: Vec<String>, security: SecurityConfig) -> KafkaClient { KafkaClient { config: ClientConfig { client_id: CLIENTID.to_owned(), hosts: hosts, compression: DEFAULT_COMPRESSION, fetch_max_wait_time: DEFAULT_FETCH_MAX_WAIT_TIME, fetch_min_bytes: DEFAULT_FETCH_MIN_BYTES, fetch_max_bytes_per_partition: DEFAULT_FETCH_MAX_BYTES_PER_PARTITION, fetch_crc_validation: DEFAULT_FETCH_CRC_VALIDATION, }, conn_pool: ConnectionPool::new(DEFAULT_SO_TIMEOUT_SECS, Some(security)), state: state::ClientState::new(), } } /// Exposes the hosts used for discovery of the target kafka /// cluster. This set of hosts corresponds to the values supplied /// to `KafkaClient::new`. #[inline] pub fn hosts(&self) -> &[String] { &self.config.hosts } /// Sets the compression algorithm to use when sending out messages. /// /// # Example /// /// ```no_run /// let mut client = kafka::client::KafkaClient::new(vec!("localhost:9092".to_owned())); /// client.load_metadata_all().unwrap(); /// client.set_compression(kafka::client::Compression::SNAPPY); /// ``` #[inline] pub fn set_compression(&mut self, compression: Compression) { self.config.compression = compression; } /// Retrieves the current `KafkaClient::set_compression` setting. #[inline] pub fn compression(&self) -> Compression { self.config.compression } /// Sets the maximum time in milliseconds to wait for insufficient /// data to become available when fetching messages. /// /// See also `KafkaClient::set_fetch_min_bytes(..)` and /// `KafkaClient::set_fetch_max_bytes_per_partition(..)`. #[inline] pub fn set_fetch_max_wait_time(&mut self, max_wait_time: i32) { self.config.fetch_max_wait_time = max_wait_time; } /// Retrieves the current `KafkaClient::set_fetch_max_wait_time` /// setting. #[inline] pub fn fetch_max_wait_time(&self) -> i32 { self.config.fetch_max_wait_time } /// Sets the minimum number of bytes of available data to wait for /// as long as specified by `KafkaClient::set_fetch_max_wait_time` /// when fetching messages. /// /// By setting higher values in combination with the timeout the /// consumer can tune for throughput and trade a little additional /// latency for reading only large chunks of data (e.g. setting /// MaxWaitTime to 100 ms and setting MinBytes to 64k would allow /// the server to wait up to 100ms to try to accumulate 64k of /// data before responding). /// /// # Example /// /// ```no_run /// use kafka::client::{KafkaClient, FetchPartition}; /// /// let mut client = KafkaClient::new(vec!["localhost:9092".to_owned()]); /// client.load_metadata_all().unwrap(); /// client.set_fetch_max_wait_time(100); /// client.set_fetch_min_bytes(64 * 1024); /// let r = client.fetch_messages(&[FetchPartition::new("my-topic", 0, 0)]); /// ``` /// /// See also `KafkaClient::set_fetch_max_wait_time(..)` and /// `KafkaClient::set_fetch_max_bytes_per_partition(..)`. #[inline] pub fn set_fetch_min_bytes(&mut self, min_bytes: i32) { self.config.fetch_min_bytes = min_bytes; } /// Retrieves the current `KafkaClient::set_fetch_min_bytes` /// setting. #[inline] pub fn fetch_min_bytes(&self) -> i32 { self.config.fetch_min_bytes } /// Sets the default maximum number of bytes to obtain from _a /// single kafka partition_ when fetching messages. /// /// This basically determines the maximum message size this client /// will be able to fetch. If a topic partition contains a /// message larger than this specified number of bytes, the server /// will not deliver it. /// /// Note that this setting is related to a single partition. The /// overall potential data size in a fetch messages response will /// thus be determined by the number of partitions in the fetch /// messages request times this "max bytes per partitions." /// /// This client will use this setting by default for all queried /// partitions, however, `fetch_messages` does allow you to /// override this setting for a particular partition being /// queried. /// /// See also `KafkaClient::set_fetch_max_wait_time`, /// `KafkaClient::set_fetch_min_bytes`, and `KafkaClient::fetch_messages`. #[inline] pub fn set_fetch_max_bytes_per_partition(&mut self, max_bytes: i32) { self.config.fetch_max_bytes_per_partition = max_bytes; } /// Retrieves the current /// `KafkaClient::set_fetch_max_bytes_per_partition` setting. #[inline] pub fn fetch_max_bytes_per_partition(&self) -> i32 { self.config.fetch_max_bytes_per_partition } /// Specifies whether the to perform CRC validation on fetched /// messages. /// /// This ensures detection of on-the-wire or on-disk corruption to /// fetched messages. This check adds some overhead, so it may be /// disabled in cases seeking extreme performance. #[inline] pub fn set_fetch_crc_validation(&mut self, validate_crc: bool) { self.config.fetch_crc_validation = validate_crc; } /// Retrieves the current `KafkaClient::set_fetch_crc_validation` /// setting. #[inline] pub fn fetch_crc_validation(&self) -> bool { self.config.fetch_crc_validation } /// Provides a view onto the currently loaded metadata of known . /// /// # Examples /// ```no_run /// use kafka::client::KafkaClient; /// use kafka::client::metadata::Broker; /// /// let mut client = KafkaClient::new(vec!["localhost:9092".to_owned()]); /// client.load_metadata_all().unwrap(); /// for topic in client.topics() { /// for partition in topic.partitions() { /// println!("{} #{} => {}", topic.name(), partition.id(), /// partition.leader() /// .map(Broker::host) /// .unwrap_or("no-leader!")); /// } /// } /// ``` #[inline] pub fn topics(&self) -> metadata::Topics { metadata::Topics::new(self) } /// Resets and loads metadata for all topics from the underlying /// brokers. /// /// # Examples /// /// ```no_run /// let mut client = kafka::client::KafkaClient::new(vec!("localhost:9092".to_owned())); /// client.load_metadata_all().unwrap(); /// for topic in client.topics().names() { /// println!("topic: {}", topic); /// } /// ``` /// /// Returns the metadata for all loaded topics underlying this /// client. #[inline] pub fn load_metadata_all(&mut self) -> Result<()> { self.reset_metadata(); self.load_metadata::<&str>(&[]) } /// Reloads metadata for a list of supplied topics. /// /// Note: if any of the specified topics does not exist yet on the /// underlying brokers and these have the [configuration for "auto /// create topics" /// enabled](https://kafka.apache.org/documentation.html#configuration), /// the remote kafka instance will create the yet missing topics /// on the fly as a result of explicitely loading their metadata. /// This is in contrast to other methods of this `KafkaClient` /// which will silently filter out requests to /// not-yet-loaded/not-yet-known topics and, thus, not cause /// topics to be automatically created. /// /// # Examples /// /// ```no_run /// let mut client = kafka::client::KafkaClient::new(vec!("localhost:9092".to_owned())); /// let _ = client.load_metadata(&["my-topic"]).unwrap(); /// ``` /// /// Returns the metadata for _all_ loaded topics underlying this /// client (this might be more topics than specified right to this /// method call.) #[inline] pub fn load_metadata<T: AsRef<str>>(&mut self, topics: &[T]) -> Result<()> { let resp = try!(self.fetch_metadata(topics)); self.state.update_metadata(resp) } /// Clears metadata stored in the client. You must load metadata /// after this call if you want to use the client. #[inline] pub fn reset_metadata(&mut self) { self.state.clear_metadata(); } /// Fetches metadata about the specified topics from all of the /// underlying brokers (`self.hosts`). fn fetch_metadata<T: AsRef<str>>(&mut self, topics: &[T]) -> Result<protocol::MetadataResponse> { let correlation = self.state.next_correlation_id(); for host in &self.config.hosts { debug!("Attempting to fetch metadata from {}", host); match self.conn_pool.get_conn(host) { Ok(conn) => { let req = protocol::MetadataRequest::new(correlation, &self.config.client_id, topics); match __send_request(conn, req) { Ok(_) => return __get_response::<protocol::MetadataResponse>(conn), Err(e) => debug!("Failed to request metadata from {}: {}", host, e), } } Err(e) => { debug!("Failed to connect to {}: {}", host, e); } } } Err(Error::NoHostReachable) } /// Fetch offsets for a list of topics /// /// # Examples /// /// ```no_run /// use kafka::client::KafkaClient; /// /// let mut client = KafkaClient::new(vec!["localhost:9092".to_owned()]); /// client.load_metadata_all().unwrap(); /// let topics: Vec<String> = client.topics().names().map(ToOwned::to_owned).collect(); /// let offsets = client.fetch_offsets(&topics, kafka::client::FetchOffset::Latest).unwrap(); /// ``` /// /// Returns a mapping of topic name to `PartitionOffset`s for each /// currently available partition of the corresponding topic. pub fn fetch_offsets<T: AsRef<str>>(&mut self, topics: &[T], offset: FetchOffset) -> Result<HashMap<String, Vec<PartitionOffset>>> { let time = offset.to_kafka_value(); let n_topics = topics.len(); let state = &mut self.state; let correlation = state.next_correlation_id(); // Map topic and partition to the corresponding broker let config = &self.config; let mut reqs: HashMap<&str, protocol::OffsetRequest> = HashMap::with_capacity(n_topics); for topic in topics { let topic = topic.as_ref(); if let Some(ps) = state.partitions_for(topic) { for (id, host) in ps.iter().filter_map(|(id, p)| p.broker(&state).map(|b| (id, b.host()))) { let entry = reqs.entry(host) .or_insert_with(|| protocol::OffsetRequest::new(correlation, &config.client_id)); entry.add(topic, id, time); } } } // Call each broker with the request formed earlier let mut res: HashMap<String, Vec<PartitionOffset>> = HashMap::with_capacity(n_topics); for (host, req) in reqs { let resp = try!(__send_receive::<protocol::OffsetRequest, protocol::OffsetResponse>(&mut self.conn_pool, &host, req)); for tp in resp.topic_partitions { let e = res.entry(tp.topic).or_insert(vec!()); for p in tp.partitions { e.push(p.into_offset()); } } } Ok(res) } /// Fetch offset for a single topic. /// /// # Examples /// /// ```no_run /// use kafka::client::{KafkaClient, FetchOffset}; /// /// let mut client = KafkaClient::new(vec!["localhost:9092".to_owned()]); /// client.load_metadata_all().unwrap(); /// let offsets = client.fetch_topic_offsets("my-topic", FetchOffset::Latest).unwrap(); /// ``` /// /// Returns a vector of the offset data for each available partition. /// See also `KafkaClient::fetch_offsets`. pub fn fetch_topic_offsets<T: AsRef<str>>(&mut self, topic: T, offset: FetchOffset) -> Result<Vec<PartitionOffset>> { let topic = topic.as_ref(); let mut m = try!(self.fetch_offsets(&[topic], offset)); let offs = m.remove(topic).unwrap_or(vec!()); if offs.is_empty() { Err(Error::Kafka(KafkaCode::UnknownTopicOrPartition)) } else { Ok(offs) } } /// Fetch messages from Kafka (multiple topic, partitions). /// /// It takes a vector specifying the topic partitions and their /// offsets as of which to fetch messages. Additionally, the /// default "max fetch size per partition" can be explicitely /// overriden if it is "defined" - this is, if `max_bytes` is /// greater than zero. /// /// The result is exposed in a raw, complicated manner but allows /// for very efficient consumption possibilities. All of the data /// available through the returned fetch responses is bound to /// their lifetime as that data is merely a "view" into parts of /// the response structs. If you need to keep individual messages /// for a longer time than the whole fetch responses, you'll need /// to make a copy of the message data. /// /// * This method transparently uncompresses messages (while Kafka /// might sent them in compressed format.) /// /// * This method ensures to skip messages with a lower offset /// than requested (while Kafka might for efficiency reasons sent /// messages with a lower offset.) /// /// Note: before using this method consider using /// `kafka::consumer::Consumer` instead which provides an easier /// to use API for the regular use-case of fetching messesage from /// Kafka. /// /// # Example /// /// This example demonstrates iterating all fetched messages from /// two topic partitions. From one partition we allow Kafka to /// deliver to us the default number bytes as defined by /// `KafkaClient::set_fetch_max_bytes_per_partition`, from the /// other partition we allow Kafka to deliver up to 1MiB of /// messages. /// /// ```no_run /// use kafka::client::{KafkaClient, FetchPartition}; /// /// let mut client = KafkaClient::new(vec!("localhost:9092".to_owned())); /// client.load_metadata_all().unwrap(); /// let reqs = &[FetchPartition::new("my-topic", 0, 0), /// FetchPartition::new("my-topic-2", 0, 0).with_max_bytes(1024*1024)]; /// let resps = client.fetch_messages(reqs).unwrap(); /// for resp in resps { /// for t in resp.topics() { /// for p in t.partitions() { /// match p.data() { /// &Err(ref e) => { /// println!("partition error: {}:{}: {}", t.topic(), p.partition(), e) /// } /// &Ok(ref data) => { /// println!("topic: {} / partition: {} / latest available message offset: {}", /// t.topic(), p.partition(), data.highwatermark_offset()); /// for msg in data.messages() { /// println!("topic: {} / partition: {} / message.offset: {} / message.len: {}", /// t.topic(), p.partition(), msg.offset, msg.value.len()); /// } /// } /// } /// } /// } /// } /// ``` /// See also `kafka::consumer`. /// See also `KafkaClient::set_fetch_max_bytes_per_partition`. pub fn fetch_messages<'a, I, J>(&mut self, input: I) -> Result<Vec<fetch::Response>> where J: AsRef<FetchPartition<'a>>, I: IntoIterator<Item=J> { let state = &mut self.state; let config = &self.config; let correlation = state.next_correlation_id(); // Map topic and partition to the corresponding broker let mut reqs: HashMap<&str, protocol::FetchRequest> = HashMap::new(); for inp in input { let inp = inp.as_ref(); if let Some(broker) = state.find_broker(inp.topic, inp.partition) { reqs.entry(broker) .or_insert_with(|| { protocol::FetchRequest::new( correlation, &config.client_id, config.fetch_max_wait_time, config.fetch_min_bytes) }) .add(inp.topic, inp.partition, inp.offset, if inp.max_bytes > 0 { inp.max_bytes } else { config.fetch_max_bytes_per_partition }); } } __fetch_messages(&mut self.conn_pool, config, reqs) } /// Fetch messages from a single kafka partition. /// /// See `KafkaClient::fetch_messages`. pub fn fetch_messages_for_partition<'a>(&mut self, req: &FetchPartition<'a>) -> Result<Vec<fetch::Response>> { self.fetch_messages(&[req]) } /// Send a message to Kafka /// /// `required_acks` - indicates how many acknowledgements the /// servers should receive before responding to the request. If it /// is 0 the server will not send any response (this is the only /// case where the server will not reply to a request). If it is /// 1, the server will wait the data is written to the local log /// before sending a response. If it is -1 the server will block /// until the message is committed by all in sync replicas before /// sending a response. For any number > 1 the server will block /// waiting for this number of acknowledgements to occur (but the /// server will never wait for more acknowledgements than there /// are in-sync replicas). /// /// `ack_timeout` - This provides a maximum time in milliseconds /// the server can await the receipt of the number of /// acknowledgements in `required_acks` /// /// `input` - A set of `ProduceMessage`s /// /// Note: Unlike the higher-level `Producer` API, this method will /// *not* automatically determine the partition to deliver the /// message to. It will strictly try to send the message to the /// specified partition. /// /// Note: Trying to send messages to non-existing topics or /// non-existing partitions will result in an error. /// /// # Example /// /// ```no_run /// use kafka::client::{KafkaClient, ProduceMessage}; /// /// let mut client = KafkaClient::new(vec!("localhost:9092".to_owned())); /// client.load_metadata_all().unwrap(); /// let req = vec![ProduceMessage::new("my-topic", 0, None, Some("a".as_bytes())), /// ProduceMessage::new("my-topic-2", 0, None, Some("b".as_bytes()))]; /// println!("{:?}", client.produce_messages(1, 100, req)); /// ``` /// /// The return value will contain a vector of topic, partition, /// offset and error if any OR error:Error. // XXX rework signaling an error; note that we need to either return the // messages which kafka failed to accept or otherwise tell the client about them pub fn produce_messages<'a, 'b, I, J>(&mut self, required_acks: i16, ack_timeout: i32, messages: I) -> Result<Vec<TopicPartitionOffset>> where J: AsRef<ProduceMessage<'a, 'b>>, I: IntoIterator<Item=J> { let state = &mut self.state; let correlation = state.next_correlation_id(); // ~ map topic and partition to the corresponding brokers let config = &self.config; let mut reqs: HashMap<&str, protocol::ProduceRequest> = HashMap::new(); for msg in messages { let msg = msg.as_ref(); match state.find_broker(msg.topic, msg.partition) { None => return Err(Error::Kafka(KafkaCode::UnknownTopicOrPartition)), Some(broker) => reqs.entry(broker) .or_insert_with( || protocol::ProduceRequest::new( required_acks, ack_timeout, correlation, &config.client_id, config.compression)) .add(msg.topic, msg.partition, msg.key, msg.value), } } __produce_messages(&mut self.conn_pool, reqs, required_acks == 0) } /// Commit offset for a topic partitions on behalf of a consumer group. /// /// # Examples /// /// ```no_run /// use kafka::client::{KafkaClient, CommitOffset}; /// /// let mut client = KafkaClient::new(vec!["localhost:9092".to_owned()]); /// client.load_metadata_all().unwrap(); /// client.commit_offsets("my-group", /// &[CommitOffset::new("my-topic", 0, 100), /// CommitOffset::new("my-topic", 1, 99)]) /// .unwrap(); /// ``` /// /// In this example, we commit the offset 100 for the topic /// partition "my-topic:0" and 99 for the topic partition /// "my-topic:1". Once successfully committed, these can then be /// retrieved using `fetch_group_offsets` even from another /// process or at much later point in time to resume comusing the /// topic partitions as of these offsets. pub fn commit_offsets<'a, J, I>(&mut self, group: &str, offsets: I) -> Result<()> where J: AsRef<CommitOffset<'a>>, I: IntoIterator<Item=J> { let state = &mut self.state; let correlation = state.next_correlation_id(); // Map topic and partition to the corresponding broker let config = &self.config; let mut reqs: HashMap<&str, protocol::OffsetCommitRequest> = HashMap:: new(); for tp in offsets { let tp = tp.as_ref(); if let Some(broker) = state.find_broker(&tp.topic, tp.partition) { reqs.entry(broker) .or_insert(protocol::OffsetCommitRequest::new(group, correlation, &config.client_id)) .add(tp.topic, tp.partition, tp.offset, ""); } } __commit_offsets(&mut self.conn_pool, reqs) } /// Commit offset of a particular topic partition on behalf of a /// consumer group. /// /// # Examples /// /// ```no_run /// use kafka::client::KafkaClient; /// /// let mut client = KafkaClient::new(vec!["localhost:9092".to_owned()]); /// client.load_metadata_all().unwrap(); /// client.commit_offset("my-group", "my-topic", 0, 100).unwrap(); /// ``` /// /// See also `KafkaClient::commit_offsets`. pub fn commit_offset(&mut self, group: &str, topic: &str, partition: i32, offset: i64) -> Result<()> { self.commit_offsets(group, &[CommitOffset::new(topic, partition, offset)]) } /// Fetch offset for a specified list of topic partitions of a consumer group /// /// # Examples /// /// ```no_run /// use kafka::client::{KafkaClient, FetchGroupOffset}; /// /// let mut client = KafkaClient::new(vec!["localhost:9092".to_owned()]); /// client.load_metadata_all().unwrap(); /// /// let offsets = /// client.fetch_group_offsets("my-group", /// &[FetchGroupOffset::new("my-topic", 0), /// FetchGroupOffset::new("my-topic", 1)]) /// .unwrap(); /// ``` /// /// See also `KafkaClient::fetch_group_topic_offsets`. pub fn fetch_group_offsets<'a, J, I>(&mut self, group: &str, partitions: I) -> Result<Vec<TopicPartitionOffset>> where J: AsRef<FetchGroupOffset<'a>>, I: IntoIterator<Item=J> { let correlation = self.state.next_correlation_id(); // Map topic and partition to the corresponding broker let mut reqs: HashMap<&str, protocol::OffsetFetchRequest> = HashMap:: new(); for tp in partitions { let tp = tp.as_ref(); if let Some(broker) = self.state.find_broker(tp.topic, tp.partition) { reqs.entry(broker) .or_insert(protocol::OffsetFetchRequest::new(group, correlation, &self.config.client_id)) .add(tp.topic, tp.partition); } } __fetch_group_offsets(&mut self.conn_pool, reqs) } /// Fetch offset for all partitions of a particular topic of a consumer group /// /// # Examples /// /// /// ```no_run /// use kafka::client::KafkaClient; /// /// let mut client = KafkaClient::new(vec!["localhost:9092".to_owned()]); /// client.load_metadata_all().unwrap(); /// let offsets = client.fetch_group_topic_offsets("my-group", "my-topic").unwrap(); /// ``` pub fn fetch_group_topic_offsets(&mut self, group: &str, topic: &str) -> Result<Vec<TopicPartitionOffset>> { let tps: Vec<_> = match self.state.partitions_for(topic) { None => return Err(Error::Kafka(KafkaCode::UnknownTopicOrPartition)), Some(tp) => tp.iter().map(|(id, _)| FetchGroupOffset::new(topic, id)).collect(), }; self.fetch_group_offsets(group, tps) } } fn __commit_offsets(conn_pool: &mut ConnectionPool, reqs: HashMap<&str, protocol::OffsetCommitRequest>) -> Result<()> { // Call each broker with the request formed earlier for (host, req) in reqs { try!(__send_receive::<protocol::OffsetCommitRequest, protocol::OffsetCommitResponse>(conn_pool, host, req)); } Ok(()) } fn __fetch_group_offsets(conn_pool: &mut ConnectionPool, reqs: HashMap<&str, protocol::OffsetFetchRequest>) -> Result<Vec<TopicPartitionOffset>> { // Call each broker with the request formed earlier let mut res = vec!(); for (host, req) in reqs { let resp = try!(__send_receive::<protocol::OffsetFetchRequest, protocol::OffsetFetchResponse>(conn_pool, host, req)); let o = resp.get_offsets(); for tpo in o { res.push(tpo); } } Ok(res) } /// ~ carries out the given fetch requests and returns the response fn __fetch_messages(conn_pool: &mut ConnectionPool, config: &ClientConfig, reqs: HashMap<&str, protocol::FetchRequest>) -> Result<Vec<fetch::Response>> { // Call each broker with the request formed earlier let mut res = Vec::with_capacity(reqs.len()); for (host, req) in reqs { let p = protocol::fetch::ResponseParser { validate_crc: config.fetch_crc_validation, requests: Some(&req), }; res.push(try!(__z_send_receive::<&protocol::FetchRequest, _>(conn_pool, host, &req, &p))); } Ok(res) } /// ~ carries out the given produce requests and returns the reponse fn __produce_messages(conn_pool: &mut ConnectionPool, reqs: HashMap<&str, protocol::ProduceRequest>, no_acks: bool) -> Result<Vec<TopicPartitionOffset>> { // Call each broker with the request formed earlier if no_acks { for (host, req) in reqs { try!(__send_noack::<protocol::ProduceRequest, protocol::ProduceResponse>(conn_pool, host, req)); } Ok(vec!()) } else { let mut res: Vec<TopicPartitionOffset> = vec![]; for (host, req) in reqs { let resp = try!(__send_receive::<protocol::ProduceRequest, protocol::ProduceResponse>(conn_pool, &host, req)); for tpo in resp.get_response() { res.push(tpo); } } Ok(res) } } fn __send_receive<T: ToByte, V: FromByte>(conn_pool: &mut ConnectionPool, host: &str, req: T) -> Result<V::R> { let mut conn = try!(conn_pool.get_conn(host)); try!(__send_request(&mut conn, req)); __get_response::<V>(&mut conn) } fn __send_noack<T: ToByte, V: FromByte>(conn_pool: &mut ConnectionPool, host: &str, req: T) -> Result<usize> { let mut conn = try!(conn_pool.get_conn(&host)); __send_request(&mut conn, req) } fn __send_request<T: ToByte>(conn: &mut KafkaConnection, request: T) -> Result<usize> { // ~ buffer to receive data to be sent let mut buffer = Vec::with_capacity(4); // ~ reserve bytes for the actual request size (we'll fill in that later) buffer.extend_from_slice(&[0, 0, 0, 0]); // ~ encode the request data try!(request.encode(&mut buffer)); // ~ put the size of the request data into the reseved area let size = buffer.len() as i32 - 4; try!(size.encode(&mut &mut buffer[..])); // ~ send the prepared buffer conn.send(&buffer) } fn __get_response<T: FromByte>(conn: &mut KafkaConnection) -> Result<T::R> { let v = try!(conn.read_exact(4)); let size = try!(i32::decode_new(&mut Cursor::new(v))); let resp = try!(conn.read_exact(size as u64)); // { // use std::fs::OpenOptions; // use std::io::Write; // let mut f = OpenOptions::new() // .write(true) // .truncate(true) // .create(true) // .open("/tmp/dump.dat") // .unwrap(); // f.write_all(&resp[..]).unwrap(); // } T::decode_new(&mut Cursor::new(resp)) } fn __z_send_receive<R, P>(conn_pool: &mut ConnectionPool, host: &str, req: R, parser: &P) -> Result<P::T> where R: ToByte, P: ResponseParser { let mut conn = try!(conn_pool.get_conn(host)); try!(__send_request(&mut conn, req)); __z_get_response(&mut conn, parser) } fn __z_get_response<P>(conn: &mut KafkaConnection, parser: &P) -> Result<P::T> where P: ResponseParser { let v = try!(conn.read_exact(4)); let size = try!(i32::decode_new(&mut Cursor::new(v))); let resp = try!(conn.read_exact(size as u64)); // { // use std::fs::OpenOptions; // use std::io::Write; // let mut f = OpenOptions::new() // .write(true) // .truncate(true) // .create(true) // .open("/tmp/dump.dat") // .unwrap(); // f.write_all(&resp[..]).unwrap(); // } parser.parse(resp) }