-
Notifications
You must be signed in to change notification settings - Fork 461
/
CHANGELOG
105 lines (53 loc) · 5.13 KB
/
CHANGELOG
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
head:
0.6.2:
- Apparently need to use custom __method__ in Ruby versions right up to 1.8.7.
0.6.1:
- Use custom __method__ in Ruby versions *prior* to 1.8.7 not *up to* 1.8.7.
- Rename Kernel#memoize to Kernel#expirable_memoize so that it doesn't conflict with memoize method in ActiveSupport which has an entirely different API and semantics. Reported by [Florian Dütsc ([email protected])].
0.6.0:
- Full 1.9 compatibility (all tests passing against 1.9 & 1.8.6). Thanks to [David ([email protected]), Cyril David ([email protected])]
0.5.1:
- For now just redefine __method__ to take arguments so we don't break 1.8.7 use today
0.5.0:
- Bug #17458 fixed. Normalize the :expires_in option to always be an integer even if the actual object passed in is a proxy to an integer, such as is the case with 2.hours from ActiveSupport which is actually an instance of ActiveSupport::Duration. Reported by [Steve Kickert [email protected]]
- Bug #19158 fixed. Don't prepend leading slash onto bucket name when deleting a bucket with the :force => true option.
- Bug #17628 fixed. Don't ignore :use_ssl => false in url_for when the connection is established over ssl. Reported by [Tom Fixed (tkat11)]
- Bug #13052 fixed. Appease some proxies by always including a Content-Length with all requests. Reported by [James Murty (jmurty)]
- Bug #13756 fixed. Attributes that are false should not raise NoMethodError in Base#method_missing. Fixed by [Scott Patten]
- Bug #19189 fixed. No longer reference Date::ABBR_MONTHS constant which was removed in Ruby 1.8.6. Reported by [Khurram Virani (kvirani)]
- Bug #20487 fixed. If a request fails and is retried, only escape the request path the first time. Reported by [anonymous]
- Replace ad-hoc S3Object.copy method with newly support built in API call.
- Do not make connections persistent by default. This "feature" causes far more broken pipes than it is worth. Use with caution.
0.4.0:
- Various adjustments to connection handling to try to mitigate exceptions raised from deep within Net::HTTP.
- Don't coerce numbers that start with a zero because the zero will be lost. If a bucket, for example, has a name like '0815', all operation it will fail. Closes ticket #10089 [reported anonymously]"
- Add ability to connect through a proxy using the :proxy option when establishing a connection. Suggested by [Simon Horne <[email protected]>]
- Add :authenticated option to url_for. When passing false, don't generate signature parameters for query string.
- Make url_for accept custom port settings. [Rich Olson]
0.3.0:
- Ensure content type is eventually set to account for changes made to Net::HTTP in Ruby version 1.8.5. Reported by [David Hanson, Stephen Caudill, Tom Mornini <[email protected]>]
- Add :persistent option to connections which keeps a persistent connection rather than creating a new one per request, defaulting to true. Based on a patch by [Metalhead <[email protected]>]
- If we are retrying a request after rescuing one of the retry exceptions, rewind the body if its an IO stream so it starts at the beginning. [Jamis Buck]
- Ensure that all paths being submitted to S3 are valid utf8. If they are not, we remove the extended characters. Ample help from [Jamis Buck]
- Wrap logs in Log objects which exposes each line as a Log::Line that has accessors by name for each field.
- Various performance optimizations for the extensions code. [Roman LE NEGRATE <[email protected]>]
- Make S3Object.copy more efficient by streaming in both directions in parallel.
- Open up Net:HTTPGenericRequest to make the chunk size 1 megabyte, up from 1 kilobyte.
- Add S3Object.exists?
0.2.1:
- When the bucket name argument (for e.g. Bucket.objects) is being used as the option hash, reassign it to the options variable and set the bucket to nil so bucket inference + options works.
- Don't call CGI.escape on query string parameters in Hash#to_query_string since all paths get passed through URI.escape right before the request is made. Paths were getting double escaped. Bug spotted by [David Hanson]
- Make s3sh exec irb.bat if on Windows. Bug spotted by [N. Sathish Kumar <[email protected]>]
- Avoid class_variable_(get|set) since it was only recently added to Ruby. Spotted by [N. Sathish Kumar <[email protected]>]
- Raise NoSuchKey if S3Object.about requests a key that does not exist.
- If the response body is an empty string, don't try to parse it as xml.
- Don't reject every body type save for IO and String at the door when making a request. Suggested by [Alex MacCaw <[email protected]>]
- Allow dots in bucket names. [Jesse Newland]
0.2.0:
- Infer content type for an object when calling S3Object.store without explicitly passing in the :content_type option.
0.1.2:
- Scrap (overly) fancy generator based version of CoercibleString with a much simpler and clearer case statement. Continuations are really slow and the specific use of the generator was leaking memory. Bug spotted by [Remco van't Veer]
0.1.1:
- Don't add the underscore method to String if it is already defined (like, for example, from ActiveSupport). Bug spotted by [Matt White <[email protected]>]
0.1.0:
- Initial public release