PARQUET-1388: Nanosecond precision time and timestamp - parquet-mr #519
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Contributor
|
I would suggest implement the refactor part as a separate JIRA. It is more a follow-up task for PARQUET-1253. |
Contributor
Author
|
Ok, created a separate PR for the refactor: #520 |
Rewrite Pig primitive type creation to builders
With wrapper RuntimeException the underlying checked FrontendException could be rethrown outside of the of the visitor, this way retaining backward compatibility
zivanfi
reviewed
Oct 3, 2018
| "TIMESTAMP_NANOS_STRINGIFIER", "yyyy-MM-dd'T'HH:mm:ss.SSS") { | ||
| @Override | ||
| public String stringify(long value) { | ||
| return super.stringify(value) + String.format("%06d", Math.abs(value % 1000_000)); |
Contributor
There was a problem hiding this comment.
Nit: 1_000_000 (also a few lines below)
| cal.clear(); | ||
| cal.set(2053, Calendar.JULY, 10, 22, 13, 24); | ||
| cal.set(Calendar.MILLISECOND, 84); | ||
| long micros = cal.getTimeInMillis() * 1000_000 + 1900; |
Contributor
There was a problem hiding this comment.
Please rename the variable to nanos and use 1_000_000 instead of 1000_000.
| cal.set(1848, Calendar.MARCH, 15, 9, 23, 59); | ||
| cal.set(Calendar.MILLISECOND, 765); | ||
| micros = cal.getTimeInMillis() * 1000_000 - 1; | ||
| assertEquals("1848-03-15T09:23:59.765000001", stringifier.stringify(micros)); |
Contributor
There was a problem hiding this comment.
The timestamp from the calendar is 1848-03-15T09:23:59.765. You subtract 1 nanosec from it, so it should be 1 nanosec earlier than 1848-03-15T09:23:59.765, which is 1848-03-15T09:23:59.764999999 and not 1848-03-15T09:23:59.765000001. Why does this pass?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR introduces nanosecond precision time and timestamp types to parquet-mr. It depends on apache/parquet-format#102, should be committed once this format change is released.
Nanosecond precision is introduced only into the new logical type API, the old version (ConvertedType in format and OriginalType in mr) doesn't have a corresponding enum value for nanoseconds. Therefore in the Thrift schema these fields will be null, only new releases can interpret nanosecond precision, older readers can only see the physical type.
In addition to nanosecond precision, I also refactored the modules to use the new logical type API for internal decisions (e.g.: replaced the OriginalType-based switch cases, for example in type builder when checking if the proper annotation is present on physical type). This part (
commit with title: Refactor modules to use the new logical type API) doesn't need new format release, and can be split out from this PR if needed.