sunchao opened a new pull request #32777:
URL: https://github.com/apache/spark/pull/32777
<!--
Thanks for sending a pull request! Here are some tips for you:
1. If this is your first time, please read our contributor guidelines:
https://spark.apache.org/contributing.html
2. Ensure you have added or run the appropriate tests for your PR:
https://spark.apache.org/developer-tools.html
3. If the PR is unfinished, add '[WIP]' in your PR title, e.g.,
'[WIP][SPARK-XXXX] Your PR title ...'.
4. Be sure to keep the PR description updated to reflect all changes.
5. Please write your PR title to summarize what this PR proposes.
6. If possible, provide a concise example to reproduce the issue for a
faster review.
7. If you want to add a new configuration, please read the guideline first
for naming configurations in
'core/src/main/scala/org/apache/spark/internal/config/ConfigEntry.scala'.
8. If you want to add or modify an error message, please read the
guideline first:
https://spark.apache.org/error-message-guidelines.html
-->
### What changes were proposed in this pull request?
<!--
Please clarify what changes you are proposing. The purpose of this section
is to outline the changes and how this PR fixes the issue.
If possible, please consider writing useful notes for better and faster
reviews in your PR. See the examples below.
1. If you refactor some codes with changing classes, showing the class
hierarchy will help reviewers.
2. If you fix some SQL features, you can provide some references of other
DBMSes.
3. If there is design documentation, please add the link.
4. If there is a discussion in the mailing list, please add the link.
-->
1. Remove duplicated code in the form of `readXXX` in
`VectorizedRleValuesReader`. For instance:
```java
public void readIntegers(
int total,
WritableColumnVector c,
int rowId,
int level,
VectorizedValuesReader data) throws IOException {
int left = total;
while (left > 0) {
if (this.currentCount == 0) this.readNextGroup();
int n = Math.min(left, this.currentCount);
switch (mode) {
case RLE:
if (currentValue == level) {
data.readIntegers(n, c, rowId);
} else {
c.putNulls(rowId, n);
}
break;
case PACKED:
for (int i = 0; i < n; ++i) {
if (currentBuffer[currentBufferIdx++] == level) {
c.putInt(rowId + i, data.readInteger());
} else {
c.putNull(rowId + i);
}
}
break;
}
rowId += n;
left -= n;
currentCount -= n;
}
}
```
and replace with:
```java
public void readBatch(
int total,
int offset,
WritableColumnVector values,
int maxDefinitionLevel,
VectorizedValuesReader valueReader,
ParquetVectorUpdater updater) throws IOException {
int left = total;
while (left > 0) {
if (this.currentCount == 0) this.readNextGroup();
int n = Math.min(left, this.currentCount);
switch (mode) {
case RLE:
if (currentValue == maxDefinitionLevel) {
updater.updateBatch(n, offset, values, valueReader);
} else {
values.putNulls(offset, n);
}
break;
case PACKED:
for (int i = 0; i < n; ++i) {
if (currentBuffer[currentBufferIdx++] == maxDefinitionLevel) {
updater.update(offset + i, values, valueReader);
} else {
values.putNull(offset + i);
}
}
break;
}
offset += n;
left -= n;
currentCount -= n;
}
}
```
where the `ParquetVectorUpdater` is type specific, and has different
implementations under `updateBatch` and `update`. Together, this also improves
code paths handling timestamp types to use the batch read API for decoding
definition levels.
2. Similar to the above, this removes code duplication in
`VectorizedColumnReader.decodeDictionaryIds`. Now different implementations are
under `ParquetVectorUpdater.decodeSingleDictionaryId`.
### Why are the changes needed?
<!--
Please clarify why the changes are needed. For instance,
1. If you propose a new API, clarify the use case for a new API.
2. If you fix a bug, you can clarify why it is a bug.
-->
The above code duplication makes it hard to maintain as any change on this
will need to be replicated in 20+ places. The issue becomes more serious when
we are going to implement column index (for instance, see how the change change
[here](https://github.com/apache/spark/pull/32753/files#diff-a01e174e178366aadf07f64ee690d47d343b2ca416a4a2b2ea735887c22d5934R191)
has to be replicated multiple times) and complex type support for the
vectorized path.
The original intention is for performance. However now days JIT compilers
tend to be smart on this and will inline virtual calls as much as possible.
I've also done benchmarks using `DataSourceReadBenchmark` and the result is
almost exact the same before and after the change. The results can be found
[here](https://gist.github.com/sunchao/674afbf942ccc2370bdcfa33efb4471c).
### Does this PR introduce _any_ user-facing change?
<!--
Note that it means *any* user-facing change including all aspects such as
the documentation fix.
If yes, please clarify the previous behavior and the change this PR proposes
- provide the console output, description and/or an example to show the
behavior difference if possible.
If possible, please also clarify if this is a user-facing change compared to
the released Spark versions or within the unreleased branches such as master.
If no, write 'No'.
-->
No
### How was this patch tested?
<!--
If tests were added, say they were added here. Please make sure to add some
test cases that check the changes thoroughly including negative and positive
cases if possible.
If it was tested in a way different from regular unit tests, please clarify
how you tested step by step, ideally copy and paste-able, so that other
reviewers can test and check, and descendants can verify in the future.
If tests were not added, please describe why they were not added and/or why
it was difficult to add.
-->
Existing tests.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
