Blobxfer

Fixed
- Fix regression in blob name encoding with Python3

Added
- Emit warning when attempting to use remoteresource with a directory upload

Changed
- Update setup.py dependencies to latest available versions

Fixed
- Fix regression in single file upload and remoteresource renaming
- Replace socket exception handling with requests ConnectionError handling
- Properly handle blob names containing `?` if using SAS

Fixed
- Disable unnecessary thread daemonization
- Gracefully handle KeyboardInterrupts
- Explicitly add azure-common to setup.py install reqs

Added
- Add python environment and package info to parameter dump to aid issue/bug
reports

Changed
- Reduce number of default concurrent workers to 3x CPU count
- Change azure\_request backoff mechanism

Fixed
- Make base requirements non-optional in import process
- Update azure\_request exception handling to support new Azure Storage Python
SDK errors

Added
- Encryption support
- No file overwrite on download option
- Auto-detection of file mimetype
- Remote delete option
- Include pattern option

Changed
- Replace keeprootdir with strip-components option
- Reduce the number of default concurrent workers to 4x CPU count

Fixed
- Fix shared key upload with non-existent container
- Fix zero-byte blob download issue

Added
- File collation support

Fixed
- Fix page alignment bug
- Reduce memory usage

Old Releases
- 0.9.9.4: improve page blob upload algorithm to skip empty max size pages.
fix zero length file uploads. fix single file upload that's skipped.
- 0.9.9.3: fix downloading of blobs with content length of zero
- 0.9.9.1: fix content length > 32bit for blob lists via SAS on Python2
- 0.9.9.0: update script for compatibility with new Azure Python packages
- 0.9.8: fix blob endpoint for non-SAS input, add retry on ServerBusy
- 0.9.7: normalize SAS keys (accept keys with or without ? char prefix)
- 0.9.6: revert local resource path expansion, PEP8 fixes
- 0.9.5: fix directory creation issue
- 0.9.4: fix Python3 compatibility issues
- 0.9.3: the script supports page blob uploading. To specify local files to
upload as page blobs, specify the `--pageblob` parameter. The script also
has a feature to detect files ending in the `.vhd` extension and will
automatically upload just these files as page blobs while uploading other
files as block blobs. Specify the `--autovhd` parameter (without the
`--pageblob` parameter) to enable this behavior.
- 0.9.0: the script will automatically default to skipping files where if the
MD5 checksum of either the local file or the stored MD5 of the remote
resource respectively matches the remote resource or local file, then the
upload or download for the file will be skipped. This capability will allow
one to perform rsync-like operations where only files that have changed will
be transferred. This behavior can be forcefully disabled by specifying
`--no-skiponmatch`.
- 0.8.2: performance regression fixes

[Unreleased]: https://github.com/Azure/blobxfer/compare/1.9.2...HEAD