-
Notifications
You must be signed in to change notification settings - Fork 72
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Use map_partitions
to compute LIMIT / OFFSET
#517
Merged
ayushdg
merged 2 commits into
dask-contrib:main
from
charlesbluca:use-map-partitions-limit
May 13, 2022
Merged
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains 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
This file was deleted.
Oops, something went wrong.
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Outdated suggestions cannot be applied.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.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
iiuc, the
partition_index
attribute associates a partition with a number/value. Is there a reason whyBlockIndex
is preferred in this situation?There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Correct,
partition_index
is expected to be the index of the partition within the overall Dask dataframe - the first partition should havepartition_index=0
, the secondpartition_index=1
, and so on.My understanding here is the "blocks" of a Dask dataframe are its partitions, making
BlockIndex
roughly synonymous with partition index here.Chatting with @rjzamora, I now know that another approach to getting partition index would be to add kwarg
partition_info
to the function, which would then be populated with the partition index, among some other info. Perhaps that approach might be a little clearer to future developers?There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't have a strong preference here and
BlockIndex
getting the same info from dask should be good for the sake of this pr.For me personally, explicitly mentioning the argument name when calling
map_partitions
should be enough to give an idea of what this is, something likedf.map_partitions(..., partition_index=BlockIndex(numblocks=(df.npartitions,))
.More generally, the reason for my confusion was primarily because I've seen
Block
like terminology used in the context of IO in dask andpartition
like terminology used in the context of dataframes.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Unfortunately,
map_partitions
has different handling for args versus kwargs that makes it so we must supply theBlockIndex
as an argument in order for it to be computed for the partition functions.I'll try out
partition_info
, since it makes it somewhat clearer what is being done and seems to be the more documented way of achieving what we want (getting partition info within the function).There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Using
partition_info
is fine. Under the hood, it will just add on aBlockwiseDep
argument (one that is a bit heavier weight thanBlockIndex
). I think you should feel comfortable going that route here.If it turns out that you need to use
BlockIndex
directly in the future, you can probably make the code a bit more intuitive by naming the argument more clearly: