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

[jvm-packages] Support different types of filesystems #965

Merged
merged 1 commit into from Mar 11, 2016

Conversation

Projects
None yet
2 participants
@shaform
Contributor

shaform commented Mar 11, 2016

As noted here: http://stackoverflow.com/a/16278054/3455159, we need to change the code a little bit to support different filesystems such as S3FileSystem.

@CodingCat

This comment has been minimized.

Show comment
Hide comment
@CodingCat

CodingCat Mar 11, 2016

Member

@shaform thanks for the PR, have you meet the problem when working with S3 ?

I think in sparkContext.hadoopConfiguration, it shall return the S3FileSystem correctly?

Member

CodingCat commented Mar 11, 2016

@shaform thanks for the PR, have you meet the problem when working with S3 ?

I think in sparkContext.hadoopConfiguration, it shall return the S3FileSystem correctly?

@shaform

This comment has been minimized.

Show comment
Hide comment
@shaform

shaform Mar 11, 2016

Contributor

I tried saving the model to S3 and failed. Since the FileSystem actually depends on the path (one might mix S3 and other paths), sparkContext.hadoopConfiguration is not enough.
Another way is to use FileSystem.get(uri, context.getConfiguration()) as noted in the same stackoverflow.

Contributor

shaform commented Mar 11, 2016

I tried saving the model to S3 and failed. Since the FileSystem actually depends on the path (one might mix S3 and other paths), sparkContext.hadoopConfiguration is not enough.
Another way is to use FileSystem.get(uri, context.getConfiguration()) as noted in the same stackoverflow.

@CodingCat

This comment has been minimized.

Show comment
Hide comment
@CodingCat

CodingCat Mar 11, 2016

Member

OK, so your source/destination are just two different file systems?

BTW, would you please update the branch, once the travis says OK, I will merge it

Member

CodingCat commented Mar 11, 2016

OK, so your source/destination are just two different file systems?

BTW, would you please update the branch, once the travis says OK, I will merge it

CodingCat added a commit that referenced this pull request Mar 11, 2016

Merge pull request #965 from shaform/fs
Support the cases when user load dataset and save model to two different file systems

@CodingCat CodingCat merged commit 8e3ce90 into dmlc:master Mar 11, 2016

1 check passed

continuous-integration/travis-ci/pr The Travis CI build passed
Details
@CodingCat

This comment has been minimized.

Show comment
Hide comment
@CodingCat

CodingCat Mar 11, 2016

Member

Thanks, @shaform , merged in

Member

CodingCat commented Mar 11, 2016

Thanks, @shaform , merged in

@CodingCat CodingCat changed the title from Support different types of filesystems to [jvm-packages] Support different types of filesystems Mar 11, 2016

@shaform

This comment has been minimized.

Show comment
Hide comment
@shaform

shaform Mar 11, 2016

Contributor

Thanks!

Contributor

shaform commented Mar 11, 2016

Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment