In previous, when querying hive table in orc format, and the file is splitted.
the result of select count(*) may be multiple of the real row number.
This is because the number of rows should be got after orc strip prune,
otherwise, it may return wrong result
1. Change from using string matching function to using Expr matching
2. Replace the `nvl` function with `ifnull` when pushed down to MySQL
3. Adapt ClickHouse's `from_unixtime` function to push down
4. Non-function filtering can still be pushed down when `enable_func_pushdown` is set to false
support insert into table values(...) for Nereids.
sql like:
insert into t values(1, 2, 3)
insert into t values(1 + 1, dayofweek(now()), 4), (4, 5, 6)
insert into t values('1', '6.5', cast(1.5 as int))
1. Change the external hive docker network mode from the bridge mode to the host mode to support the external test of the multi-node doris cluster
2. Added more hive test data in various formats
3. Added a test case with hive
now we make error if we deal with leading zeros in decimal value , type_precision >= precision will make value overflow and DCHECK will fail , so if here has leading zero we should only make type_precision > precision to make value right
Enable two phase partition topn optimization, instead of original full sort at the second phase.
E.g, partial plan of tpcds q67 is as following and a full sort after exchange will have performance impact, especially if the window column's ndv is very high and the number of window is huge.
------PhysicalTopN
--------filter((rk <= 100))
----------PhysicalWindow
------------PhysicalQuickSort
--------------PhysicalDistribute
----------------PhysicalPartitionTopN
------------------PhysicalProject
Under this scenario, the second phase full sort can be transformed to a global PhysicalPartitionTopN and reduce the cost from full sort. The plan will be optimized to the following:
------PhysicalTopN
--------filter((rk <= 100))
----------PhysicalWindow
------------PhysicalPartitionTopN
--------------PhysicalDistribute
----------------PhysicalPartitionTopN
------------------PhysicalProject
Now we can not support streamload with column which is map/array nested map/array
serde can do this now , so we can replace it
Notice. if item data in complex type data is empty we just return error, instead of makeup default value , because now we can not define right default for complex type
we generate project for all set operation's children to ensure the order
of all children are not changed. However, some rules, such as
PushDownProjectThroughLimit could remove these projects involuntarily.
When it happen, the column order is wrong and lead to BE core dump.
This PR use a new variable in SetOperation to save the output order of
children of set operation. Then the children's output order could be
changed and never affect to SetOperation at all.
Issue Number: close#24315
The root cause of this issue is that Elasticsearch's long type allows inserting floats and strings. Doris did not handle these cases when doing type conversion. The current strategy is to take the integer before the decimal point if a float or string is found.
Add file cache regression test in tpch 1g on orc&parquet format.
tpch will run 3 times:
1. running without file cache
2. running with file cache for the first time
3. running with file cache for the second time
The file cache configuration is already added in `be/conf/be.conf` on the regression test environment, and the available capacity is 100MB. After running the tpch 1g test, the metrics introduced by https://github.com/apache/doris/pull/19177 is like:
```
doris_be_file_cache_normal_queue_curr_size{path="/mnt/datadisk1/gaoxin/file_cache"} 92808933
doris_be_file_cache_normal_queue_curr_elements{path="/mnt/datadisk1/gaoxin/file_cache"} 59
doris_be_file_cache_normal_queue_max_elements{path="/mnt/datadisk1/gaoxin/file_cache"} 102400
doris_be_file_cache_normal_queue_max_size{path="/mnt/datadisk1/gaoxin/file_cache"} 89128960
doris_be_file_cache_removed_elements{path="/mnt/datadisk1/gaoxin/file_cache"} 2132
doris_be_file_cache_segment_reader_cache_size{path="/mnt/datadisk1/gaoxin/file_cache"} 54
```
CREATE EXTERNAL TABLE `dim_server` (
`col1` varchar(50) NOT NULL,
`col2` varchar(50) NOT NULL
)
create view ads_oreo_sid_report
(
`col1` ,
`col2`
)
AS
select
tmp.col1,tmp.col2
from (
select 'abc' as col1,'def' as col2
) tmp
inner join dim_server ds on tmp.col1 = ds.col1 and tmp.col2 = ds.col2;
select * from ads_oreo_sid_report where col1='abc' and col2='def';
before this pr, col1='abc' and col2='def' can't be pushed to dim_server. now the 2 predicates can be pushed to odbc table.
Fix three bugs:
1. Hudi slice maybe has log files only, so `new Path(filePath)` will throw errors.
2. Hive column names are lowercase only, so match column names in ignore-case-mode.
3. Compatible with [Spark Datasource Configs](https://hudi.apache.org/docs/configurations/#Read-Options), so users can add `hoodie.datasource.merge.type=skip_merge` in catalog properties to skip merge logs files.
before the lambda function Expr not implement toSqlImpl() function.
so it's call parent function, which is not suit for lambda function.
and will be have error when create view.
