5d6eeb5a7f
Map urllib3 ReadTimeoutError to an appropriate ReadTimeout requests exception, issue #2392
2018-01-13 03:28:02 +03:00
37f1d23c39
Merge remote-tracking branch 'upstream/master' into proposed/3.0.0
2017-10-15 13:59:50 -07:00
e2f5a135c3
remove legacy super() call
2017-07-26 19:21:45 -05:00
9766870807
Use comprehensions whenever possible
2017-07-14 11:42:32 -03:00
4847f5b8cd
Allow Requests.Response to be used as a context manager
...
This saves having to wrap the call to requests with
`contextlib.closing()`, allowing it to be used directly in a
`with` statement, like so:
```
with requests.get('http://httpbin.org/get ', stream=True) as r:
# Do things with the response here.
```
Fixes #4136 .
2017-06-06 23:30:47 +01:00
548823be83
Clean up invalid indentation from merge
2017-05-31 10:41:47 +01:00
a6f7f5dc62
Clean up flake8 errors from merge
2017-05-31 10:39:27 +01:00
c3f4a028bf
Merge branch 'master' into proposed/3.0.0
2017-05-31 10:29:19 +01:00
kennethreitz
33619268f9
cleanup models.py
2017-05-29 23:00:09 -04:00
kennethreitz
7f14db17c8
new requests namespace
2017-05-29 12:11:43 -04:00
kennethreitz
6d3d5e59c9
fix models.py
2017-05-26 22:55:55 -04:00
kennethreitz
6c41633cfd
fix models.py
2017-05-26 22:50:13 -04:00
kennethreitz
c3367d1854
remove Request.send(session)
2017-05-26 21:28:52 -04:00
kennethreitz
dfd667df96
who wrote this shit
2017-05-26 21:17:13 -04:00
kennethreitz
ce64cce14e
API improvements
2017-05-26 21:07:37 -04:00
a889b62c50
Merge branch 'master' into proposed/3.0.0
2017-05-21 16:25:55 -07:00
458df8f4f4
Fix the additional newline generated by iter_lines() caused by a '\r\n' pair being separated in two different chunks.
2017-04-26 23:02:57 +08:00
19ba9f1097
Remove some unused imports.
2017-04-25 22:10:01 +01:00
ead8fba84b
Fixes DeprecationWarnings in Python 3.6 due to invalid escape sequences. Fixes #3954
2017-04-05 15:51:58 +01:00
d491e9f9b2
use [-1] instead of endswith() to work with bytes or string
...
Also add a parametrize on decode_unicode for iter_lines() test
to check with bytestrings and str content
2017-03-15 22:29:18 +01:00
052595ffbf
add explanatory comment about skipping null chunks in iter_lines
2017-03-15 12:16:32 +01:00
cc2ac23c0d
remove useless brackets in iter_lines boolean condition
2017-03-15 01:37:55 +01:00
9881be25f3
Review markups for @Lukasa
2017-03-14 16:57:52 +01:00
9174925916
Fix bug when delimiter is split between responses
2017-03-14 16:57:52 +01:00
5a65a0dab1
use InvalidHeader instead of ConflictingHeaderError
2017-03-01 07:32:14 -07:00
a52fe6586c
consolidate super_len code and cleanup docstrings
2017-02-28 14:51:08 -07:00
f239fe754d
Change UnreachableCodeError to InvalidBodyError.
2017-02-28 14:49:52 -07:00
1003fdf0f2
Small fixes based on feedback in pull request.
2017-02-28 14:49:31 -07:00
033dfc165d
Raise an error if body is not null, has not length and is not streamable.
2017-02-28 14:49:20 -07:00
bfb202527d
Alternate fix for 3066 to refactor prepare_body to always call prepare_content_length.
...
This allows for the 'Content-Length' header to only be set in prepare_content_length.
2017-02-28 14:49:01 -07:00
8e049bd9eb
Update docstrings for Response.__{bool,nonzero}__
...
Like the docstring added to Response.ok these were misleading and
vaguely incorrect. Better to be explicit than implicit.
2017-02-16 07:21:15 -06:00
7a281e74ec
Correct docstring for Response.ok
2017-02-16 07:15:37 -06:00
3664682539
added docstring for response.ok property
2017-02-16 10:06:28 +01:00
efcbe93075
Make Response.raise_for_status() return the response object if the response is successful
...
This allows for chaining method calls in cases where we want to raise for bad codes but use the response otherwise,
e.g. requests.get(URL).raise_for_status().json()['value']
2017-02-12 11:11:40 +02:00
241f2b2a6b
Merge branch 'master' into proposed/3.0.0
2017-02-10 17:40:52 +00:00
fae3f92c34
lazily load idna library
2017-01-17 20:34:15 +05:30
ba7c464a03
Merge branch 'master' into proposed/3.0.0
2016-12-08 09:16:51 +00:00
402a55b647
Revert "Restrict URL preparation to HTTP/HTTPS"
...
This reverts commit 34af72c87d
2016-11-30 21:19:31 +00:00
5c4549493b
Merge pull request #3695 from nateprewitt/idna_bypass
...
idna bypass
2016-11-25 13:18:01 +00:00
f897be58bf
Make Response.content return None if raw is None
...
Add test.
2016-11-24 12:54:26 +00:00
4f428228d7
Make Response.content() return None if status_code is None
...
Fix #3698 .
2016-11-23 21:15:07 +00:00
34af72c87d
Restrict URL preparation to HTTP/HTTPS
...
Requests treats all URLs starting with the string 'http' as HTTP URLs.
Preparation with IDNA breaks non-standard URIs like http+unix. Requests
now prepares only URLs with prefix http:// and https://.
Signed-off-by: Christian Heimes <christian@python.org >
2016-11-21 18:46:57 +01:00
a83685ce00
modifying IDNA encoding check to allow fallback
2016-11-21 08:22:42 -07:00
0d7de50dbd
Just lstrip instead of strip because trailing whitespace already works.
2016-11-16 23:26:48 +05:30
248d3e9f0f
Fixed issue #3696 . Added a test for it as well.
2016-11-16 22:48:45 +05:30
84d99f01f5
add enforce_content_length=True default
2016-11-15 11:27:29 -07:00
7f88aca8f7
Merge branch 'master' into proposed/3.0.0
2016-11-15 17:16:04 +00:00
a49c387b7a
Document that Response.json() may raise ValueError
2016-11-10 16:26:07 -05:00
40402cd0dd
adding rewind for re-POST bodies
2016-11-02 09:56:56 -06:00
548a03acef
requests.post checked data parameters for type like dict.
...
Changed the type check to Mapping.
2016-10-27 17:57:09 +02:00
VasiliPupkin256
kennethreitz