* Fix encoding error with non-prettified encoded responses
Removed `--format-option response.as` an promote `--response-as`: using
the format option would be misleading as it is now also used by non-prettified
responses.
* Encoding refactoring
* split --response-as into --response-mime and --response-charset
* add support for Content-Type charset for requests printed to terminal
* add support charset detection for requests printed to terminal without a Content-Type charset
* etc.
* `test_unicode.py` → `test_encoding.py`
* Drop sequence length check
* Clean-up tests
* [skip ci] Tweaks
* Use the compatible release clause for `charset_normalizer` requirement
Cf. https://www.python.org/dev/peps/pep-0440/#version-specifiers
* Clean-up
* Partially revert d52a4833e4
* Changelog
* Tweak tests
* [skip ci] Better test name
* Cleanup tests and add request body charset detection
* More test suite cleanups
* Cleanup
* Fix code style in test
* Improve detect_encoding() docstring
* Uniformize pytest.mark.parametrize() calls
* [skip ci] Comment out TODOs (will be tackled in a specific PR)
Co-authored-by: Jakub Roztocil <jakub@roztocil.co>
* Allow to override the response `Content-Type` from options
* Apply suggestions from code review
Co-authored-by: Jakub Roztocil <jakub@roztocil.co>
* Rename the option from `--response.content-type` to `--response-as`
* Update CHANGELOG.md
Co-authored-by: Jakub Roztocil <jakub@roztocil.co>
In some special cases, to prevent against Cross Site Script Inclusion (XSSI)
attacks, the JSON response body starts with a magic prefix line that must be
stripped before feeding the rest of the response body to the JSON parser.
Such prefix is now simply ignored from the parser but still printed in the
terminal.
* Fix Windows tests
* Add a workflow to check documentations
* Fix markdown issues
* Install Ruby 2.7
* Finally, handle and fix GitHub templates
* Minor improvement in the feature request template
* Verbose mode to be sure all files are checked
* Convert most of the documentation from the frontend `README.rst` to `docs/REAME.md`
Also converted all reStructuredText files to Markdown.
* Tell `mdformat` to use LF for end on lines
* `--check` is not needed in the help message
* Skip tests on GitHub Windows.
Those tests pass on a real Windows machine.
Let's revisit those failure later, if needed.
* Move `mdoformat` requirement from `test` to `dev` extra
To fix Fedora CI.
* Fix the handling of cookies from session files
* Apply suggestions from code review
Co-authored-by: Jakub Roztocil <jakub@roztocil.co>
* Fix test docstring formatting
Co-authored-by: Jakub Roztocil <jakub@roztocil.co>
* Support `requests.response.raw` being a file-like object
Previously HTTPie relied on `requests.models.Response.raw` being
`urllib3.HTTPResponse`. The `requests` documentation specifies that
(requests.models.Response.raw)[https://docs.python-requests.org/en/master/api/#requests.Response.raw]
is a file-like object but allows for other types for internal use.
This change introduces graceful handling for scenarios when
`requests.models.Response.raw` is not `urllib3.HTTPResponse`. In such a scenario
HTTPie now falls back to extracting metadata from `requests.models.Response`
directly instead of direct access from protected protected members such as
`response.raw._original_response`. A side effect in this fallback procedure is
that we can no longer determine HTTP protocol version and report it as `1.1`.
This change is necessary to make it possible to implement `TransportPlugins`
without having to also needing to emulate internal behavior of `urlib3` and
`http.client`.
* Load cookies from `response.headers` instead of `response.raw._original_response.msg._headers`
`response.cookies` was not utilized as it not possible to construct original
payload from `http.cookiejar.Cookie`. Data is stored in lossy format. For example
`Cookie.secure` defaults to `False` so we cannot distinguish if `Cookie.secure` was
set to `False` or was not set at all. Same problem applies to other fields also.
* Simpler HTTP envelope data extraction
* Test cookie extraction and make cookie presentment backwards compatible
Co-authored-by: Mickaël Schoentgen <contact@tiger-222.fr>
Co-authored-by: Jakub Roztocil <jakub@roztocil.co>
* Make sure there’s no trailing \n in test files for easier output inspection
* Refactor output matching test utils
* More robust `test_http_307_allow_redirect_post_verbose()`
* Changelog
* Mention HTTP 307 Temporary Redirect re-post behaviour in README
The fix may actually be slightly more complex than I expected.
We would need, at first sight, to loose the prepared requests + responses streaming.
A potential solution will come in a near future.
https://docs.python.org/3.10/whatsnew/3.10.html#enum
Python 3.10 changed the repr of enum members, and the doctest of tests.utils.http failed.
Exact reprs are unfortunately not considered stable API between Python releases:
=================================== FAILURES ===================================
__________________________ [doctest] tests.utils.http __________________________
209
210 Example:
211
212 $ http --auth=user:password GET pie.dev/basic-auth/user/password
213
214 >>> httpbin = getfixture('httpbin')
215 >>> r = http('-a', 'user:pw', httpbin.url + '/basic-auth/user/pw')
216 >>> type(r) == StrCLIResponse
217 True
218 >>> r.exit_status
Expected:
<ExitStatus.SUCCESS: 0>
Got:
ExitStatus.SUCCESS
A simple replacement of the expected output however breaks the doctest on Python 3.9.
This is the best solution I could think of
that keeps the docstring readable and doctest working in Pythons both old and new.
* Add --raw to allow specifying the raw request body without extra processing
As an alternative to `stdin`.
Co-authored-by: Elena Lape <elapinskaite@gmail.com>
Co-authored-by: Jakub Roztocil <jakub@roztocil.co>
* Update README.rst
Co-authored-by: Jakub Roztocil <jakub@roztocil.co>
* Update README.rst
Co-authored-by: Jakub Roztocil <jakub@roztocil.co>
* Fix default HTTP method on empty data
Co-authored-by: Elena Lape <elapinskaite@gmail.com>
Co-authored-by: Jakub Roztocil <jakub@roztocil.co>
* Add skip when required TLS version unsupported
Allow tests to skip, rather than fail from SSL errors with unsupported
TLS version, e.g. if Openssl is configured with MinProtocol higher than
the version being tested.
* Regenerate test certificate and key
Regenerate these with more secure settings for the sake of future
proofing, regenerate the key using RSA 4096 and sign the certificate
with SHA512.
This fixes test failures in tests/test_ssl.py when the user's OpenSSL
security level is set to a value greater than 1 and resolves issue #948
* Suppress SSL warnings in no verify tests
* Add test to test auth plugin reused in session
* Remove unnecessary assertion in auth-plugin test
* Fixed auth test to use same session file
* Add test for password prompt behaviour in session
* Edit auth readme for plugin clarity
* added a test for expiring cookies
* updated tests
* set up util for extracting expired cookies from response header
* Revert "updated tests"
This reverts commit a4eb5c4498.
* Revert "Revert "updated tests""
This reverts commit d242e21bce.
* added more functionality to get-expired-cookies
* add 'clear expired cookies' from session.json files
* refactored get_expired_cookies
* fixed formatting issues
* ensured key exists in cookie_header dict
* fixed linting errors
* removed unused import
* Added tests for get_expired_cookies util
* Added additional test for get_expired_cookies
* added remove_expired_cookies method directly to sessions class
* extracted logic to clear cookies to sessions.py
* refactored utils
* added tests to check expired cookies being removed from session obj
* added type annotations for methods
* Refactored test_sessions
* Seperated out expiry related tests into own class
* Refactored get_expired_cookies in utils
* Refactored remove cookie methods
* fixed linting errors
* fixed indentation and also pluralized test class name
* removed inheritance from SessionTestbase class
* Moved related test to TestExpiredCookies class
Co-authored-by: kbanc <katherine.bancoft@gmail.com>
* Support curl-like syntax for custom MIME type for files
In order to specify a custom MIME type for file uploads, a syntax
similar to that used by cURL is used so that
http -F test_file@/path/to/file.bin;type=application/zip https://...
forwards the user-provided file type if provided, otherwise falling
back to the usual guesswork out of the file extension.
On Unix-like systems, the configuration file now lives in
$XDG_CONFIG_HOME/httpie/ by default, not ~/.httpie/ (the behaviour on
Windows is unchanged). The previous location is still checked, in order
to support existing installations.
Searching $XDG_CONFIG_DIRS is still not supported.
Fixes#145; supersedes #436.
