* removed packer.Cache and references since packer.Cache is never used except in the download step. The download step now uses the new func packer.CachePath(targetPath) for this, the behavior is the same.
* removed download code from packer that was reimplemented into the go-getter library: progress bar, http download restart, checksuming from file, skip already downloaded files, symlinking, make a download cancellable by context.
* on windows if packer is running without symlinking rights and we are getting a local file, the file will be copied instead to avoid errors.
* added unit tests for step_download that are now CI tested on windows, mac & linux.
* files are now downloaded under cache dir `sha1(filename + "?checksum=" + checksum) + file_extension`
* since the output dir is based on the source url and the checksum, when the checksum fails, the file is auto deleted.
* a download file is protected and locked by a file lock,
* updated docs
* updated go modules and vendors
Previously, when you built from an existing image, you were unable
to reconfigure block device mappings, as it just took them and
copied them over. This allows users to specify new, custom
block device mappings, even when building from an existing
image.
Per code review feedback, the Go template library supports
encoding variables in URL query format. Instead of exposing
two different public key formats (unmodified string and a
URL encoded string), just have the user apply the 'urlquery'
modifier to their template.
to retain current encryption settings
this changes the fields :
* ami_config.encrypt_boot
* block_device.encrypted
This also removes StepCreateEncryptedAMICopy as this step is now done in StepAMIRegionCopy
While the `kmsArnStartPattern` regexp supports `us-gov` as a region, it does not take the AWS partition into consideration.
For more info see https://docs.aws.amazon.com/govcloud-us/latest/ug-east/using-govcloud-arns.html
Added test case for AWS partition us-gov as well as a negative test case for a non-existing AWS partition.
There is some logic around how EBS mappings are handled / generated depending
on whether an image if from scratch or not. There was no testing around
this behaviour before. This strips that logic out into a separate function,
to enable testing it independentlhy.
Right now, if we have a source image that's PV, and try to build an
image with mount_partition set to not 0, it does not get picked up.
This is because under PV we only had a filesystem, not partitions,
but you can convert a PV image to an HVM image during build time.