1
0
Fork 0
mirror of https://github.com/fog/fog.git synced 2022-11-09 13:51:43 -05:00

[storage|aws] Fix docs to say files.each, not each_file.

each_file was an API that was discussed, but dismissed as a bad idea.
The documentation still had a reference to each_file, though.
This commit is contained in:
Samuel Merritt 2011-04-28 11:57:31 -07:00
parent 746f39a526
commit 5aed5ee93e

View file

@ -86,17 +86,34 @@ and maybe some pictures of your cat doing funny stuff. Since this is
all of vital importance, you need to back it up.
# copy each file to local disk
directory.each_file do |s3_file|
directory.files.each do |s3_file|
File.open(s3_file.key, 'w') do |local_file|
local_file.write(s3_file.body)
end
end
One caveat: it's tempting to just write `directory.files.each` here,
but that only works until you get a large number of files. S3's API
for listing files forces pagination on you. `directory.each_file`
takes pagination into account; `directory.files.each` will only
operate on the first page.
One caveat: it's way more efficient to do this:
# do two things per file
directory.files.each do |file|
do_one_thing(file)
do_another_thing(file)
end
than it is to do this:
# do two things per file
directory.files.each do |file|
do_one_thing(file)
end.each do |file|
do_another_thing(file)
end
The reason is that the list of files might be large. Really
large. Eat-all-your-RAM-and-ask-for-more large. Therefore, every time
you say `files.each`, fog makes a fresh set of API calls to Amazon to
list the available files (Amazon's API returns a page at a time, so
fog works a page at a time in order to keep its memory requirements sane).
## Sending it out