gawel/pyquery

Is it OK to pass bytes in as the first arguement to PyQuery()?

Closed this issue · 3 comments

cclauss commented

We have code that is sometimes calling pyquery.PyQuery("a string (str)") and is other times calling pyquery.PyQuery(b"a bytes string (bytes)"). Is this acceptable or should the first argument always be a str instead of bytes?

The question comes up because basestring is defined as (str, bytes) which is different than the usual definition of basestring which in Python 2 was defined as (str, unicode) and in Python 3 is usually defined as (str, ).

>>> from pyquery import PyQuery as pq
>>> pq("a").encoding, pq(b"a").encoding
('ISO-8859-1', 'UTF-8')
gawel commented

As I remember lxml deals with both cases

cclauss commented

Agreed. Beyond the .encoding difference mentioned above, is there any other reason to prefer bytes over str?

gawel commented

As I remember lmxl detect the encoding based on the charset specified in the xml/html when you provides bytes. So you don't have to deal with this when you're scraping a large amount of web pages. I guess it's the only reason

  • 👍1