comments and minor refactoring

.coveragerc

@@ -0,0 +1,10 @@
+[run]
+source = .
+branch = true
+omit =
+    .venv/*
+
+[report]
+fail_under = 100
+show_missing = True
+skip_covered = True

.gitignore

@@ -3,4 +3,8 @@
 # PyCharm:
 /.idea/
 # Distribution / packaging:
-*.egg-info/
+*.egg-info/
+# Python cache:
+__pycache__/
+# Tests:
+.coverage

coverage.sh

@@ -0,0 +1,3 @@
+#!/usr/bin/env sh
+
+coverage erase && coverage run -m unittest discover && coverage report

setup.cfg

@@ -19,7 +19,7 @@ keywords = webscraping, html, markup, dom, scraper, attributes, tags, stocks, fi
 package_dir =
     = src
 packages = find:
-python_requires = >=3.7
+python_requires = >=3.8
 install_requires =
     beautifulsoup4
     aiohttp

src/stocksymbolscraper/__main__.py

@@ -8,7 +8,7 @@ from pathlib import Path
 from . import get_all_data, log
 
 
-def main() -> None:
+async def main() -> None:
     parser = ArgumentParser(description="Scrape all stock symbols")
     parser.add_argument(
         '-v', '--verbose',
@@ -29,7 +29,7 @@ def main() -> None:
     if args.verbose:
         log.setLevel(logging.DEBUG)
 
-    data = asyncio.run(get_all_data(args.sequential))
+    data = await get_all_data(args.sequential)
 
     if args.to_file is None:
         csv.writer(sys.stdout).writerows(data)
@@ -39,4 +39,4 @@ def main() -> None:
 
 
 if __name__ == '__main__':
-    main()
+    asyncio.run(main())

tests/test_scrape.py

@@ -1,12 +1,13 @@
-from unittest import TestCase
+import logging
-from unittest.mock import patch, MagicMock, call
+from unittest import IsolatedAsyncioTestCase
+from unittest.mock import patch, MagicMock, AsyncMock, call
 
 from bs4 import BeautifulSoup
 
 from stocksymbolscraper import scrape
 
 
-class ScrapeTestCase(TestCase):
+class ScrapeTestCase(IsolatedAsyncioTestCase):
 
     @patch.object(scrape, 'get_single_tr_data')
     def test_extract_row_data(self, mock_get_single_tr_data: MagicMock):
@@ -53,8 +54,111 @@ class ScrapeTestCase(TestCase):
         output = scrape.get_str_from_td(test_td)
         self.assertEqual(expected_output, output)
 
+    @patch.object(scrape, 'ClientSession')
+    async def test_soup_from_url(self, mock_session_cls):
+        test_html = '<b>foo</b>'
+        mock_response = MagicMock()
+        mock_response.text = AsyncMock(return_value=test_html)
+        mock_get_return = MagicMock()
+        mock_get_return.__aenter__ = AsyncMock(return_value=mock_response)
+        mock_session_obj = MagicMock()
+        mock_session_obj.get = MagicMock(return_value=mock_get_return)
+        mock_session_cls.return_value = mock_session_obj
+        output = await scrape.soup_from_url('foo')
+        expected_output = BeautifulSoup(test_html, scrape.HTML_PARSER)
+        self.assertEqual(expected_output, output)
+
+        output = await scrape.soup_from_url('foo', mock_session_obj)
+        self.assertEqual(expected_output, output)
+
     def test_trs_from_page(self):
-        # Tested function takes URL as argument (GET request is issued)
+        tr1_text, tr2_text = '<tr>foo</tr>', '<tr>bar</tr>'
-        # HTML to be parsed could be substituted
+        test_html = f'<div id="marketsindex"><table><tbody>{tr1_text}{tr2_text}</tbody></table></div>'
-        #
+        test_soup = BeautifulSoup(test_html, scrape.HTML_PARSER)
-        pass
+        output = scrape.trs_from_page(test_soup)
+        expected_output = test_soup.find_all('tr')
+        self.assertSequenceEqual(expected_output, output)
+
+        logging.disable(logging.CRITICAL)
+        test_html = f'<div id="marketsindex"><table>garbage</table></div>'
+        test_soup = BeautifulSoup(test_html, scrape.HTML_PARSER)
+        with patch.object(scrape, 'open') as mock_open:
+            self.assertRaises(scrape.UnexpectedMarkupError, scrape.trs_from_page, test_soup)
+            mock_open.assert_called_once()
+            mock_open.return_value.__enter__.return_value.write.assert_called_once_with(test_soup.prettify())
+        logging.disable(logging.NOTSET)
+
+    @patch.object(scrape, 'soup_from_url')
+    @patch.object(scrape, 'trs_from_page')
+    @patch.object(scrape, 'extract_row_data')
+    @patch.object(scrape, 'ClientSession')
+    async def test_get_data_from_category(self, mock_session_cls, mock_extract_row_data,
+                                          mock_trs_from_page, mock_soup_from_url):
+        # We do not pass a session object into the tested function,
+        # so we mock the ClientSession class for the first two tests.
+        mock_session = MagicMock()
+        mock_session_cls.return_value = mock_session
+        mock_soup = MagicMock()
+        mock_soup_from_url.return_value = mock_soup
+        category = 'ßßß'
+        url = f'{scrape.BASE_URL}{category}'
+
+        # First test with no TRs returned, thus not entering the loop.
+        mock_trs = []
+        mock_trs_from_page.return_value = mock_trs
+        mock_extract_row_data.return_value = expected_output = []
+        output = await scrape.get_data_from_category(category)
+        self.assertListEqual(expected_output, output)
+        mock_soup_from_url.assert_called_once_with(url, mock_session)
+        mock_trs_from_page.assert_called_once_with(mock_soup)
+        mock_extract_row_data.assert_not_called()
+
+        mock_soup_from_url.reset_mock()
+        mock_trs_from_page.reset_mock()
+
+        # Second test with (fake) TRs returned, thus entering the loop.
+        # We pass a last_page argument to stop after the first iteration.
+        mock_trs = ['foo', 'bar']
+        mock_trs_from_page.return_value = mock_trs
+        mock_extract_row_data.return_value = expected_output = ['a', 'b']
+
+        # Factored out checks because the only difference in the next two tests is the presence or absence of a
+        # real session instance.
+        def check_assertions(test_output, session_obj):
+            self.assertListEqual(expected_output, test_output)
+            mock_soup_from_url.assert_has_calls([call(url, session_obj), call(f'{url}/{2}', session_obj)])
+            mock_trs_from_page.assert_has_calls([call(mock_soup), call(mock_soup)])
+            mock_extract_row_data.assert_called_once_with(*mock_trs)
+
+        output = await scrape.get_data_from_category(category, last_page=1)
+        check_assertions(output, mock_session)
+
+        mock_soup_from_url.reset_mock()
+        mock_trs_from_page.reset_mock()
+        mock_extract_row_data.reset_mock()
+
+        # Third test with (fake) TRs returned and explicitly passing a real session object.
+        async with scrape.ClientSession() as session:
+            output = await scrape.get_data_from_category(category, session, last_page=1)
+            check_assertions(output, session)
+
+    @patch.object(scrape, 'get_data_from_category')
+    @patch.object(scrape, 'ClientSession')
+    async def test_get_all_data(self, mock_session_cls, mock_get_data_from_category):
+        mock_session = MagicMock()
+        mock_session_cls.return_value.__aenter__.return_value = mock_session
+        mock_result = ['foo']
+        expected_output = len(scrape.CATEGORIES) * mock_result
+        mock_get_data_from_category.return_value = mock_result
+
+        output = await scrape.get_all_data(sequential=True)
+        self.assertListEqual(expected_output, output)
+        mock_get_data_from_category.assert_has_calls([
+            call(category, mock_session) for category in scrape.CATEGORIES
+        ])
+
+        output = await scrape.get_all_data(sequential=False)
+        self.assertListEqual(expected_output, output)
+        mock_get_data_from_category.assert_has_calls([
+            call(category, mock_session) for category in scrape.CATEGORIES
+        ])