pyrox.dev
1{ stdenv, buildPythonPackage, fetchPypi, glibcLocales, mock, pytest, botocore,
2 testfixtures, pillow, six, twisted, w3lib, lxml, queuelib, pyopenssl,
3 service-identity, parsel, pydispatcher, cssselect, lib }:
4buildPythonPackage rec {
5 version = "1.5.1";
6 pname = "Scrapy";
7
8 checkInputs = [ glibcLocales mock pytest botocore testfixtures pillow ];
9 propagatedBuildInputs = [
10 six twisted w3lib lxml cssselect queuelib pyopenssl service-identity parsel pydispatcher
11 ];
12
13 # Scrapy is usually installed via pip where copying all
14 # permissions makes sense. In Nix the files copied are owned by
15 # root and readonly. As a consequence scrapy can't edit the
16 # project templates.
17 patches = [ ./permissions-fix.patch ];
18
19 LC_ALL="en_US.UTF-8";
20
21 checkPhase = ''
22 py.test --ignore=tests/test_linkextractors_deprecated.py --ignore=tests/test_proxy_connect.py ${lib.optionalString stdenv.isDarwin "--ignore=tests/test_utils_iterators.py"}
23 # The ignored tests require mitmproxy, which depends on protobuf, but it's disabled on Python3
24 # Ignore iteration test, because lxml can't find encodings on darwin https://bugs.launchpad.net/lxml/+bug/707396
25 '';
26
27 src = fetchPypi {
28 inherit pname version;
29 sha256 = "5a398bf6818f87dcc817c919408a195f19ba46414ae12f259119336cfa862bb6";
30 };
31
32 meta = with lib; {
33 description = "A fast high-level web crawling and web scraping framework, used to crawl websites and extract structured data from their pages";
34 homepage = https://scrapy.org/;
35 license = licenses.bsd3;
36 maintainers = with maintainers; [ drewkett ];
37 platforms = platforms.unix;
38 };
39}