Name: s3-upload-missing
Owner: P'unk Avenue
Description: Upload contents of given folder to given s3 path, recursively. If a file already exists, do not upload it.
Created: 2015-12-11 20:01:50.0
Updated: 2015-12-14 02:43:49.0
Pushed: 2015-12-14 01:01:04.0
Homepage: null
Size: 30
Language: JavaScript
GitHub Committers
| User | Most Recent Commit | # Commits |
|---|
Other Committers
| User | Most Recent Commit | # Commits |
|---|
ovel this folder's contents, recursively, up to mybucket.
a file is already there, it is not sent again
pload-missing . mybucketname .
it with verbose output
pload-missing --verbose . mybucketname .
ecify a prefix in the destination bucket.
slash at the end is implied if not provided
pload-missing --verbose . mybucketname uploads
low the public to read the files (web-accessible)
pload-missing --verbose . mybucketname uploads --acl=public-read
a file is unreadable, chmod it momentarily so we
n read it, then send it to S3 with "private" permissions.
en chmod it back to 000
pload-missing --verbose . mybucketname uploads --acl=public-read --chmod-if-needed
so remove any remote files that do not exist locally.
e with care
pload-missing . mybucketname . --delete
You must populate ~/.aws/credentials with your key and secret, like this:
ault]
access_key_id = xxx
secret_access_key = yyyyyy
TODO: support command line arguments for these as well.
“Why not use s3cmd?” s3cmd works fine, but we have a peculiar need to successfully upload files with permissions 000 and give them the private acl on s3 (the --chmod-if-needed option). This is very useful when transitioning from local files to s3 with uploadfs. Also, s3-upload-missing may be faster.