Regression in 6.2.0 when URL source to be parsed contains a query string #2120
Description
Graph.parse in 6.2.0 is producing inconsistent statement counts when reading over HTTP versus 6.1.1 when the source URL contains a query string.
To reproduce:
$ date -u
Wed Sep 21 16:57:54 UTC 2022
$ cd /tmp
$ python3.10 --version
Python 3.10.5
$ python3.10 -m venv venv
$ cd venv
$ bin/pip install --quiet --upgrade setuptools pip wheel rdflib==6.2.0
$ bin/python
Python 3.10.5 (main, Jun 23 2022, 17:15:25) [Clang 13.1.6 (clang-1316.0.21.2.5)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import rdflib
>>> a, b = rdflib.Graph(), rdflib.Graph()
>>> a.parse('https://bmdb.jpl.nasa.gov/rdf/example')
>>> b.parse('https://bmdb.jpl.nasa.gov/rdf/example?all=right')
>>> len(a), len(b), len(a) != len(b)
(2, 2, False)This should be (2, 4, True).
Using 6.1.1 produces the correct results:
$ bin/pip install --quiet rdflib==6.1.1
$ bin/python
Python 3.10.5 (main, Jun 23 2022, 17:15:25) [Clang 13.1.6 (clang-1316.0.21.2.5)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import rdflib
>>> a, b = rdflib.Graph(), rdflib.Graph()
>>> a.parse('https://bmdb.jpl.nasa.gov/rdf/example')
>>> b.parse('https://bmdb.jpl.nasa.gov/rdf/example?all=right')
>>> len(a), len(b), len(a) != len(b)
(2, 4, True)Apparently in 6.2.2 the query string part of the URL ?all=right gets stripped out. It is necessary with certain web services APIs in order to select different sets of RDF statements.