Skip to content
Commits on Sep 14, 2011
  1. gitignore

    committed Sep 14, 2011
  2. readme is in md now

    committed Sep 14, 2011
Commits on Apr 23, 2010
  1. ok, so like, bsd and linux find are different: bsd can accept granula…

    …r times like 30s and 1m, linux can't, but they both have 'files newer than file x', so we're saving the state of the sync in an actual file with touch and cp and junk, this'll work cross platform and ain't too shabby. i don't care that this is starting to look like a perl script, coz linux is the bomb, shut up!
    committed Apr 23, 2010
Commits on Apr 22, 2010
  1. now i'm positively confused, find is working differently on two opera…

    …ting systems? not sure, it might be borked, more likely i'm borked
    committed Apr 22, 2010
  2. damn it, got args around the wrong way, worked on osx, not on linux, …

    …put them around the right way and using display to show time since last sync so you can see in the output
    committed Apr 22, 2010
  3. add -ctime based comparisions, so sync saves when it was last run, sh…

    …ould be way faster than running full checksum comparisons (also fixing bug in error handling, had wrong method name)
    committed Apr 22, 2010
Commits on Apr 12, 2010
  1. figured out how to get a full list of objects from s3 in a determinis…

    …tic manner, hooray, now can do backups more reliably, and now using find with -ctime is actually a decent thing to do...
    committed Apr 12, 2010
Commits on Apr 1, 2010
  1. wrong numbers in the readme

    committed Apr 1, 2010
  2. remote, not origin

    committed Apr 1, 2010
  3. more updates to the readme

    committed Apr 1, 2010
  4. ok, so amazon's list of objects request is just like, so annoying. in…

    …stead of trying to traverse through all objects with multiple damn requests with markers and prefixes (what were they thinking?), we're gonna push up our local manifest to the remote for comparison with a later local manifest. -cmin based comparisons are out for the moment (because the checksum doesn't perform poorly on 500,000 files i think), let's see how this goes
    committed Apr 1, 2010
Commits on Mar 31, 2010
  1. damn, you surely don't have more than one million files? this is a bi…

    …t lame, might have to store a manifest on s3...
    committed Apr 1, 2010
  2. need a chomp

    committed Apr 1, 2010
  3. here's a readmelol

    committed Apr 1, 2010
  4. license yay

    committed Apr 1, 2010
  5. sync start is pushed out of the sync method, which is correct, aws st…

    …oring objects correctly, and we have checksum backups working, hurrah, so next we gotta make sure the ctime stuff works and we're done
    committed Mar 31, 2010
  6. implementation of setup method (creation of buckets and verification …

    …of credentials and whatnot), locking is put in the right place and junk
    committed Mar 31, 2010
  7. logic fleshed out for sync, but data and op methods not implemented, …

    …added locking to sync and setup
    committed Mar 31, 2010
Commits on Mar 30, 2010
  1. gets doesn't work for some reason, using stdin.readline, bug in local…

    … file path check, adding find options to config so we can filter out .git and wp-cache files in our specific use case, dry up messages referring to the name of the file
    committed Mar 31, 2010
Something went wrong with that request. Please try again.