Skip to content

Latest commit

 

History

History
171 lines (102 loc) · 4.74 KB

NEWS.md

File metadata and controls

171 lines (102 loc) · 4.74 KB
Raw

Version 0.2.47

  • Fix bug in s3exists when using aws-cli.

Version 0.2.46

  • Fix typo in last_modified.

Version 0.2.45

  • Compatibility with aws-cli.

Version 0.2.44

  • Bugfixes and better compatibility with system environment variables.

Version 0.2.43

  • Allowing s3mpi.path and s3mpi.cache to be called via system environment variables.

Version 0.2.42

  • Better error output in s3read if path does not exist. (see issue #72)

Version 0.2.41

  • Fix incorrect strptime version specification.

Version 0.2.40

  • Remove overly complicated metaprogramming.

Version 0.2.33-9

  • Other fixes for s4cmd.

Version 0.2.32

  • allow choice of storage format in s3read and s3store. Defaults to RDS, and now you can chooose CSV or table for data frames.

Version 0.2.31

  • Make s3store work with s4cmd.

Version 0.2.30

  • Don't set --bucket-location flag if s4cmd is detected as the s3cmd binary.

Version 0.2.29

  • Revert the change in 0.2.21 in favor of using s3cmd info over s3cmd ls to obtain updated_at information on files.

Version 0.2.28

  • Add exponential backoff logic to s3.put function.

Version 0.2.27

  • Turn off the LRU cache too when cache = FALSE in s3read.

Version 0.2.26

  • options(s3mpi.num_retries) now allows you to specify default number of retries globally.

Version 0.2.25

  • Automatically adds ending slashes to paths if they are missing when using s3store, s3exists, or s3delete.

Version 0.2.24

  • Add the ability to delete an object in s3 using s3delete.

Version 0.2.23

  • s3path() is exported.

Version 0.2.22

  • s3read() (with no arguments) is no longer supported.

Version 0.2.21

  • Fixed an issue where reading files that have the same prefix as another file on the S3 bucket generates a warning.
  • Fix a more serious problem where writing and reading within the same minute produces incorrect results due to the s3cmd utility having minute-level rather than second-level granularity.

Version 0.2.20

  • Workaround for the silent but oh-so-deadly sporadic failure of s3cmd's put. By default we now check for the existence of the object when issuing a put, with the option to retry a number of times.

Version 0.2.19

  • Keep AWS.tools on a remote.

Version 0.2.18

  • Add remotes to DESCRIPTION.

Version 0.2.17

  • Explicitly create the directory of a file given by tempfile() to prevent rare errors wherein the directory does not exist and yields a file connection error. (#41)

Version 0.2.16

  • Introduce an s3mpi.disable_lru_cache option as well as silently fail if storage to LRU does not succeed.

Version 0.2.15

  • Switch to system2, which should be more windows friendly, and allow the user to specify path to executable of s3cmd, by setting options(s3mpi.s3cmd_path = '/usr/local/bin/s3cmd')

Version 0.2.13

  • Fixup LRU cache to actually use size parameter option.

Version 0.2.11

  • A stylistic refactor of the package. The .path argument has been deprecated in s3read and s3store in favor of simply path.

Version 0.2.9-10

  • Remove the need to type a trailing slash in .path.

Version 0.2.8

  • A hotfix for cache corruption, where data exists without metadata. It can happen if writing metadata ever fails.

Version 0.2.7

  • Remove the s3mpi.memoize_cache global option, since it makes no sense. A user could have overwritten an S3 key in a different R session.

  • s3exists(NULL) now returns FALSE. Fixes issue #22.

Version 0.2.5-6

  • The s3mpi.memoize_cache global controls whether or not caching is memoised. If set to TRUE, it would have the effect of keeping a common object in the R session instead of retrieving it from the cached file for each given s3 key. This can significantly speed up code that reads from the same S3 key multiple times within a single R session.

Version 0.2.4

  • The safety check on s3store uses safe = FALSE by default now. The new function s3put is equivalent to s3store and should be used going forward if one does not wish to overwrite existing keys. The other approach was causing too many breaking changes to existing codebases.

Version 0.2.2

  • Added a safety check for s3store. Now if you want to overwrite a key inside a bucket, you need to use s3store(key, safe = FALSE). By default safe is set to TRUE.

Version 0.2.0

  • Added a caching mechanism that will keep copies of files downloaded and uploaded to S3. Useful if local storage constraints are not an issue. To enable, set options(s3mpi.cache = '~/.s3cache') in your ~/.Rprofile (or replace '~/.s3cache' with a directory of your choice).