add from_bytes_dict alternative constructor for HttpResponseHeaders #33
Conversation
Codecov Report
@@ Coverage Diff @@
## master #33 +/- ##
=========================================
Coverage 100.00% 100.00%
=========================================
Files 5 6 +1
Lines 127 224 +97
=========================================
+ Hits 127 224 +97
|
| alongside a plain ``bytes`` value. | ||
|
|
||
| By default, it converts the ``bytes`` value using "utf-8". However, this | ||
| can easily be overridden using the ``encoding`` parameter. |
There was a problem hiding this comment.
This utf-8 default value was based on how Scrapy was using it as the default for its Headers.
web_poet/page_inputs.py
Outdated
| @@ -74,6 +75,45 @@ def from_name_value_pairs(cls: Type[T_headers], arg: List[Dict]) -> T_headers: | |||
| """ | |||
| return cls([(pair["name"], pair["value"]) for pair in arg]) | |||
|
|
|||
| @classmethod | |||
| def from_bytes( | |||
There was a problem hiding this comment.
The name may be a bit misleading, one may think this creates Headers from HTTP response bytes which represent the headers. Maybe something like from_bytes_dict? Or maybe even more specific, from_scrapy_headers?
There was a problem hiding this comment.
Good point! Renaming it to from_scrapy_headers seems to be too specific. Other non-scrapy libs could have the same setup. Having it as from_bytes_dict makes it more clear and semantically more meaningful at the same time. Updated this on eb18427
| if isinstance(data, str): | ||
| return data | ||
| elif isinstance(data, bytes): | ||
| return data.decode(encoding) |
There was a problem hiding this comment.
I think it should raise an exception if data is not bytes or str
There was a problem hiding this comment.
Good catch! Adressed this in ddb7d20 alongside the ability to handle Tuple alongside List.
BurnzZ
changed the title
|
Thanks @BurnzZ! |
From the recent enhancement in #30, it would seem that we'll need to support insantiating the
HttpResponseHeadersfrom raw bytes values. In particular, frameworks like Scrapy uses header values in bytes format.By default, https://github.com/aio-libs/multidict only supports strings and not bytes. Thus, an alternative constructor is proposed for ease of use.