# Proposed changes
This PR fixed lots of issues when building from source on macOS with Apple M1 chip.
## ATTENTION
The job for supporting macOS with Apple M1 chip is too big and there are lots of unresolved issues during runtime:
1. Some errors with memory tracker occur when BE (RELEASE) starts.
2. Some UT cases fail.
...
Temporarily, the following changes are made on macOS to start BE successfully.
1. Disable memory tracker.
2. Use tcmalloc instead of jemalloc.
This PR kicks off the job. Guys who are interested in this job can continue to fix these runtime issues.
## Use case
```shell
./build.sh -j 8 --be --clean
cd output/be/bin
ulimit -n 60000
./start_be.sh --daemon
```
## Something else
It takes around _**10+**_ minutes to build BE (with prebuilt third-parties) on macOS with M1 chip. We will improve the development experience on macOS greatly when we finish the adaptation job.
1. enable varchar/char type set min/max value.
take first 8 chars as long, and convert to double.
2. fix bug when set min/max value for date and datav2
This pr is used to expand the supported data type for array_min/array_max function.
Before the change , the array_min/array_max function can't support the date/datetime type.
After the change, array_min/array_max function can support the date/datetime type.
Co-authored-by: hucheng01 <hucheng01@baidu.com>
1. Refactor the file reader creation in FileFactory, for simplicity.
Previously, FileFactory had too many `create_file_reader` interfaces.
Now unified into two categories: the interface used by the previous BrokerScanNode,
and the interface used by the new FileScanNode.
And separate the creation methods of readers that read `StreamLoadPipe` and other readers that read files.
2. Modify the StreamLoadPlanner on FE side to support using ExternalFileScanNode
3. Now for generic reader, the file reader will be created inside the reader, not passed from the outside.
4. Add some test cases for csv stream load, the behavior is same as the old broker scanner.
Previously, bthread_getspecific was called every time bthread local was used. In the test at #10823, it was found that frequent calls to bthread_getspecific had performance problems.
So a cache is implemented on pthread local based on the btls key, but the btls key cannot correctly sense bthread switching.
So, based on bthread_self to get the bthread id to implement the cache.
1. this pr is used to update the json load docs for import data to array column.
when we use json to import data to array column, the Rapidjson will cause precision problems.
so we update the json-load docs to specify how to avoid these problems.
Issue Number: #7570
Co-authored-by: hucheng01 <hucheng01@baidu.com>
Doris do not support explicitly cast NULL_TYPE to ANY type .
```
mysql> select cast(NULL as int);
ERROR 1105 (HY000): errCode = 2, detailMessage = Invalid type cast of NULL from NULL_TYPE to INT
```
So we should also forbid user from casting NULL_TYPE to ARRAY type.
This commit will produce the following effect:
```
mysql> select cast(NULL as array<int>);
ERROR 1105 (HY000): errCode = 2, detailMessage = Invalid type cast of NULL from NULL_TYPE to ARRAY<INT(11)>
```
We should prevent insert while value overflow.
1. create table:
`CREATE TABLE test_array_load_test_array_int_insert_db.test_array_load_test_array_int_insert_tb ( k1 int NULL, k2 array<int> NULL ) DUPLICATE KEY(k1) DISTRIBUTED BY HASH(k1) BUCKETS 5`
2. try insert data less than INT_MIN.
`insert into test_array_load_test_array_int_insert_tb values (1005, [-2147483649])`
Before this pr, the insert will success, but the value it not correct.
Co-authored-by: cambyzju <zhuxiaoli01@baidu.com>
disable page cache by default
disable chunk allocator by default
not use chunk allocator for vectorized allocator by default
add a new config memory_linear_growth_threshold = 128Mb, not allocate memory by RoundUpToPowerOf2 if the allocated size is larger than this threshold. This config is added to MemPool, ChunkAllocator, PodArray, Arena.
* [fix](agg)the reseet the content of grouping exprs instead of replace it with original exprs
* keep old behavior if the grouping type is not GROUP_BY
Allocator of rapidjson does not release memory, this fix use allocator with local buffer and call Clear to release memory allocated beyond local buffer.
Manually drop statistics for tables or partitions. Table or partition can be specified, if neither is specified, all statistics under the current database will be deleted.
syntax:
```SQL
DROP STATS [tableName [PARTITIONS(partitionNames)]];
-- e.g.
DROP STATS; -- drop all table statistics under the current database
DROP STATS t0; -- drop t0 statistics
DROP STATS t1 PARTITIONS(p1); -- drop partition p1 statistics of t1
```
