package.debian: Re-implement debian loader within the package loader mechanism

This introduces the debian package loader.

Contrary to other package loaders, a package can have multiple artifacts (per package version).

In those terms, we had to slightly adapt the swh.loader.package.loader.PackageLoader to demonstrate this. Thus the impacts on pypi, npm, deposit modules as well.

Related #2025 (closed) Related swh-core!91 (closed) Related swh-lister!114 (closed)

Test Plan

tox

---

Migrated from D2135 (view on Phabricator)

swh/loader/package/tests/test_debian.py

@@ -3,10 +3,13 @@
 # License: GNU General Public License version 3, or any later version
 # See top-level LICENSE file for more information
 
+import copy
+import pytest
+
 from os import path
 
 from swh.loader.package.debian import (
-    DebianLoader, get_file_info, download_package
+    DebianLoader, get_file_info, download_package, dsc_information
 )
 from swh.loader.package.tests.common import check_snapshot
 
@@ -94,6 +97,39 @@ def test_download_package(datadir, tmpdir, requests_mock_http_datadir):
             'length': 96527}}
 
 
+def test_dsc_information_ok():
+    fname = 'cicero_0.7.2-3.dsc'
+    dsc_url, dsc_name = dsc_information(PACKAGE_FILES)
+
+    assert dsc_url == PACKAGE_FILES['files'][fname]['uri']
+    assert dsc_name == PACKAGE_FILES['files'][fname]['name']
+
+
+def test_dsc_information_not_found():
+    fname = 'cicero_0.7.2-3.dsc'
+    package_files = copy.deepcopy(PACKAGE_FILES)
+    package_files['files'].pop(fname)
+
+    dsc_url, dsc_name = dsc_information(package_files)
+
+    assert dsc_url is None
+    assert dsc_name is None
+
+
+def test_dsc_information_too_many_dsc_entries():
+    # craft an extra dsc file
+    fname = 'cicero_0.7.2-3.dsc'
+    package_files = copy.deepcopy(PACKAGE_FILES)
+    data = package_files['files'][fname]
+    fname2 = fname.replace('cicero', 'ciceroo')
+    package_files['files'][fname2] = data
+
+    with pytest.raises(
+            ValueError, match='Package %s_%s references several dsc' % (
+                package_files['name'], package_files['version'])):
+        dsc_information(package_files)
+
+
 def test_debian_first_visit(
         swh_config, requests_mock_http_datadir):
     """With no prior visit, load a gnu project ends up with 1 snapshot