1. Fix value idx in bool rle decoder
2. Iceberg table support datetimev2(3). In the previous version, we converted hive timestamp to datetimev2(0) default.
1. Introduce hadoop libhdfs
2. For Linux-X86 platform, use the hadoop libhdfs
3. For other platform, use libhdfs3, because currently we don't have hadoop libhdfs binary for other platform
Co-authored-by: adonis0147 <adonis0147@gmail.com>
rpc common is duplicate, all its method is included in function rpc. So that I remove it.
get_field_type is never used, remove it.
---------
Co-authored-by: yiguolei <yiguolei@gmail.com>
On macOS systems with Apple silicon, the '==' operator of StringRef uses string_compare, which takes StringRef as a C-String with null-terminated chars.
In clickhouse's 4.x version of jdbc, some UInt types use special Java types, so I adapted Doris's ClickHouse JDBC External
```
com.clickhouse.data.value.UnsignedByte;
com.clickhouse.data.value.UnsignedInteger;
com.clickhouse.data.value.UnsignedLong;
com.clickhouse.data.value.UnsignedShort;
```
1. add PassNullPredicate to fix topn wrong result for NULL values
2. refactor RuntimePredicate to avoid using TCondition
3. refactor using ordering_exprs in fe and vsort_node
A framework that read data from jni scanner, which can support the data source from java ecosystem(java API).
## Java Interface
Java scanner should extends `org.apache.doris.jni.JniScanner`, implements the following methods:
```
// Initialize JniScanner
public abstract void open() throws IOException;
// Close JniScanner and release resources
public abstract void close() throws IOException;
// Scan data and save as vector table
public abstract int getNext() throws IOException;
```
See demo usage in `org.apache.doris.jni.MockJniScanner`
## c++ interface
C++ reader should use `doris::JniConnector` to get data from `org.apache.doris.jni.JniScanner`. See demo usage in `doris::MockJniReader`.
## Pushed-down predicates
Java scanner can get pushed-down predicates by `org.apache.doris.jni.vec.ScanPredicate`.
## Remaining works:
1. Implement complex nested types.
2. Read hudi MOR table as the end-to-end demo usage.
when loading big file with multi bytes line delimiter, some line record maybe incomplete because of _output_buf_limit, so this incomplete data will move to the beginning of the output buf and read more data into output buf. In this case, find line delimiter should start with no offset to avoid a bug that spilt two lines as one line.
Result is empty for query select * from person where address like '%\\\\%';, but MySQL can get a line of result.
CREATE TABLE `person` (
`id` int(11) NULL,
`name` text NULL,
`age` int(11) NULL,
`class` int(11) NULL,
`address` text NULL
) ENGINE=OLAP
UNIQUE KEY(`id`)
COMMENT 'OLAP'
DISTRIBUTED BY HASH(`id`) BUCKETS 1
PROPERTIES (
"replication_allocation" = "tag.location.default: 1",
"in_memory" = "false",
"storage_format" = "V2",
"disable_auto_compaction" = "false"
);
insert into person values (10001,'test1',30,2,'test\\\\,xxx');
Adding logs:
select * from person where address like '%\\\\%';
I0323 10:26:15.907760 2387043 like.cpp:558] arg str: %\\%, size: 4, pattern LIKE_ENDS_WITH_RE: (?:%+)(((\\%)|(\\_)|([^%_]))+), size: 30
I0323 10:26:15.907789 2387043 like.cpp:562] match 0: \\%, size: 3
I0323 10:26:15.907801 2387043 like.cpp:562] match 1: \%, size: 2
I0323 10:26:15.907811 2387043 like.cpp:562] match 2: \%, size: 2
I0323 10:26:15.907821 2387043 like.cpp:562] match 3: , size: 0
I0323 10:26:15.907830 2387043 like.cpp:562] match 4: \, size: 1
I0323 10:26:15.907842 2387043 like.cpp:615] search_string : \\%
I0323 10:26:15.907855 2387043 like.cpp:619] search_string escape removed: \%
It matchs against the LIKE_ENDS_WITH_RE which is wrong, the meaning of the sql should be: match strings that have one backslash in any place.
* Revert "[fix](function) fix AES/SM3/SM4 encrypt/ decrypt algorithm initialization vector bug (#17420)"
This reverts commit 397cc011c4f1ba5a25c770258c13f1cd3f28b47d.
* [fix-resubmit](function) fix AES/SM3/SM4 encrypt/ decrypt algorithm initialization vector bug (#17420)
ECB algorithm, block_encryption_mode does not take effect, it only takes effect when init vector is provided.
Solved: 192/256 supports calculation without init vector
For other algorithms, an error should be reported when there is no init vector
Initialization Vector. The default value for the block_encryption_mode system variable is aes-128-ecb, or ECB mode, which does not require an initialization vector. The alternative permitted block encryption modes CBC, CFB1, CFB8, CFB128, and OFB all require an initialization vector.
Reference: https://dev.mysql.com/doc/refman/8.0/en/encryption-functions.html#function_aes-decrypt
Note: This fix does not support smooth upgrades. during upgrade process, query may report error: funciton not found
Arena can replace MemPool in most scenarios. Except for memory reuse, MemPool supports reuse of previous memory chunks after clear, but Arena does not.
Some comparisons between MemPool and Arena:
1. Expansion
Arena is less than 128M index 2 alloc chunk; more than 128M memory, allocate 128M * n > `size`, n is equal to the minimum value that satisfies the expression;
MemPool less than 512K index 2 alloc chunk, greater than 512K memory, separately apply for a `size` length chunk
After Arena applied for a chunk larger than 128M last time, the minimum chunk applied for after that is 128M. Does this seem to be a waste of memory? MemPool is also similar. After the chunk of 512K was applied for last time, the minimum chunk of subsequent applications is 512K.
2. Alignment
MemPool defaults to 16 alignment, because memtable and other places that use int128 require 16 alignment;
Arena has no default alignment;
3. Memory reuse
Arena only supports `rollback`, which reuses the memory of the current chunk, usually the memory requested last time.
MemPool supports clear(), all chunks can be reused; or call ReturnPartialAllocation() to roll back the last requested memory; if the last chunk has no memory, search for the most free chunk for allocation
4. Realloc
Arena supports realloc contiguous memory; it also supports realloc contiguous memory from any position at the time of the last allocation. The difference between `alloc_continue` and `realloc` is:
1. Alloc_continue does not need to specify the old size, but the default old size = head->pos - range_start
2. alloc_continue supports expansion from range_start when additional_bytes is between head and pos, which is equivalent to reusing a part of memory, while realloc completely allocates a new memory
MemPool does not support realloc, but supports transferring or absorbing chunks between two MemPools
5. check mem limit
MemPool checks the mem limit, and Arena checks at the Allocator layer.
6. Support for ASAN
Arena does something extra
7. Error handling
MemPool supports returning the error message of application failure directly through `Status`, and Arena throws Exception.
Tests that Arena can consider
1. After the last applied chunk is larger than 128M, the minimum applied chunk is 128M, which seems to waste memory;
2. Support clear, memory multiplexing;
3. Increase the large list, alloc the memory larger than 128M, and the size is equal to `size`, so as to avoid the current chunk not being fully used, which is wasteful.
4. In some cases, it may be possible to allocate backwards to find chunks t
Follow #17586.
This PR mainly changes:
Remove env/
Remove FileUtils/FilesystemUtils
Some methods are moved to LocalFileSystem
Remove olap/file_cache
Add s3 client cache for s3 file system
In my test, the time of open s3 file can be reduced significantly
Fix cold/hot separation bug for s3 fs.
This is the last PR of #17764.
After this, all IO operation should be in io/fs.
Except for tests in #17586, I also tested some case related to fs io:
clone
concurrency query on local/s3/hdfs
load error log create and clean
disk metrics
PR(#17330) has changed the column type of kay and value from array to normal column, but orc&parquet reader still cast to array column, resulting in cast error.
In the case of rowets non-overlap and desc sorting, the logic of VCollectIterator::Level0Iterator::init_for_union will be followed. In this function, the row ref pos of the first level0 iterator is set to 0, and the row pos of other level0 iterators are all Set to -1.
But in the level1iterator, when rowets are non-overlapping and is ordering by desc, the list of rowset iterators will be reversed, causing the row ref pos of the first level0 iterator in the list to be -1, causing the block reader to think that the entire tablet has no data.
After the memory exceeds the limit, the previous query waited for memory free in the mem hook, and changed it to wait in the Allocator.
more controllable and safe
add <optional> head to solve the compilation issue
use 3.12.9 as the protoc.artifact's version, because there is no 3.12.21
See: https://repo.maven.apache.org/maven2/com/google/protobuf/protoc/
Remove --show-progress arguments of wget because it is not supported in low version wget
When using JDBC Catalog to query the Doris data, because Doris does not provide the cursor reading method (that is, fetchBatchSize is invalid), Doris will send the data to the client at one time, resulting in client OOM.
The MySQL protocol provides a stream reading method. Doris can use this method to avoid OOM. The requirements of using the stream method are setting fetchbatchsize = Integer.MIN_VALUE and setting ResultSet.TYPE_FORWARD_ONLY, ResultSet.CONCUR_READ_ONLY
#11740 , solved the problem that the query memory statistics are higher than the actual physical memory, because PODArray does not have memset 0 when allocating memory, and the query mem tracker is virtual memory.
But in extreme cases, such as csv load, PODArray frequent insert will cause performance problems. So revert part of #11740 and part of #12820.
The accuracy of the query mem tracker, there is currently no feedback, no further attention.
Problem:
1. FE will split the parquet file into split. So a file can have several splits.
2. BE will scan each split, read the footer of the parquet file.
3. If 2 splits belongs to a same parquet file, the footer of this file will be read twice.
This PR mainly changes:
1. Use kv cache to cache the footer of parquet file.
2. The kv cache is belong to a scan node, so all parquet reader belong to this scan node will share same kv cache.
3. In cache, the key is "meta_file_path", the value is parsed thrift footer.
The KV Cache is sharded into mutlti sub cache.
So that different file can use different sub cache, avoid blocking each other
In my test, a query with 26 splits can reduce the footer parse time from 4s -> 1s
