From: Scott Worley Date: Tue, 19 Dec 2023 09:28:31 +0000 (-0800) Subject: Use the new encapsulated fetchers X-Git-Url: http://git.scottworley.com/paperdoorknob/commitdiff_plain/a64403ac570d0049c7b5a616d19fab37ab5cb4e8?ds=sidebyside Use the new encapsulated fetchers --- diff --git a/paperdoorknob.py b/paperdoorknob.py index 8ab62f1..f83c9f8 100644 --- a/paperdoorknob.py +++ b/paperdoorknob.py @@ -14,10 +14,10 @@ from typing import IO, Iterable from bs4 import BeautifulSoup from bs4.element import Tag -import requests -import requests_cache from xdg_base_dirs import xdg_cache_home +from fetch import CachingFetcher, Fetcher + def command_line_parser() -> ArgumentParser: parser = ArgumentParser(prog='paperdoorknob', description='Print glowfic') @@ -40,10 +40,8 @@ def command_line_parser() -> ArgumentParser: return parser -def fetch(url: str, session: requests.Session, timeout: int) -> BeautifulSoup: - with session.get(url, timeout=timeout) as r: - r.raise_for_status() - return BeautifulSoup(r.text, 'html.parser') +def fetch(url: str, fetcher: Fetcher) -> BeautifulSoup: + return BeautifulSoup(fetcher.fetch(url), 'html.parser') def clean(html: BeautifulSoup) -> BeautifulSoup: @@ -79,13 +77,11 @@ def html_to_tex(pandoc: str, tag: Tag) -> bytes: def process( url: str, - session: requests.Session, - timeout: int, + fetcher: Fetcher, texout: IO[bytes], - pandoc: str, -) -> None: + pandoc: str) -> None: texout.write(b'\\documentclass{article}\n\\begin{document}\n') - html = clean(fetch(url, session, timeout)) + html = clean(fetch(url, fetcher)) for r in replies(html): texout.write(html_to_tex(pandoc, r)) texout.write(b'\\end{document}\n') @@ -93,14 +89,9 @@ def process( def main() -> None: args = command_line_parser().parse_args() - with requests_cache.CachedSession(args.cache_path, cache_control=True) as session: + with CachingFetcher(args.cache_path, args.timeout) as fetcher: with open(args.out + '.tex', 'wb') as texout: - process( - args.url, - session, - args.timeout, - texout, - args.pandoc or 'pandoc') + process(args.url, fetcher, texout, args.pandoc or 'pandoc') if __name__ == '__main__': diff --git a/paperdoorknob_test.py b/paperdoorknob_test.py index c896810..49bef9f 100644 --- a/paperdoorknob_test.py +++ b/paperdoorknob_test.py @@ -8,70 +8,31 @@ import unittest import io import subprocess -import requests -import requests_cache import paperdoorknob from testing.fakeserver import FakeGlowficServer +from fetch import DirectFetcher TIMEOUT = 8 -class TestFetch(unittest.TestCase): +class TestPaperDoorknob(unittest.TestCase): def setUp(self) -> None: self._server = self.enterContext(FakeGlowficServer()) self._port = self._server.port() - def testFetch(self) -> None: - with requests.session() as s: - paperdoorknob.fetch(f"http://localhost:{self._port}", s, TIMEOUT) - self.assertEqual(self._server.request_count(), 1) - paperdoorknob.fetch(f"http://localhost:{self._port}", s, TIMEOUT) - self.assertEqual(self._server.request_count(), 2) - - def testFetchCaching(self) -> None: - with requests_cache.CachedSession() as s: - paperdoorknob.fetch(f"http://localhost:{self._port}", s, TIMEOUT) - self.assertEqual(self._server.request_count(), 1) - paperdoorknob.fetch(f"http://localhost:{self._port}", s, TIMEOUT) - self.assertEqual(self._server.request_count(), 1) - - def testFetchPersistentCaching(self) -> None: - with requests_cache.CachedSession() as s: - paperdoorknob.fetch(f"http://localhost:{self._port}", s, TIMEOUT) - self.assertEqual(self._server.request_count(), 1) - with requests_cache.CachedSession() as s: - paperdoorknob.fetch(f"http://localhost:{self._port}", s, TIMEOUT) - self.assertEqual(self._server.request_count(), 1) - def testReplies(self) -> None: - with requests.session() as s: + with DirectFetcher(TIMEOUT) as f: replies = paperdoorknob.replies( paperdoorknob.clean( - paperdoorknob.fetch( - f"http://localhost:{self._port}", - s, - TIMEOUT))) + paperdoorknob.fetch(f"http://localhost:{self._port}", f))) self.assertEqual([r.text.strip() for r in replies], ["This is glowfic", "You sure?", "Pretty sure."]) - def testFetchErrors(self) -> None: - with requests.session() as s: - with self.assertRaises(requests.HTTPError): - paperdoorknob.fetch( - f"http://localhost:{self._port}/not_found", s, TIMEOUT) - with self.assertRaises(requests.HTTPError): - paperdoorknob.fetch( - f"http://localhost:{self._port}/server_error", s, TIMEOUT) - def testProcess(self) -> None: - with requests.session() as s: + with DirectFetcher(TIMEOUT) as f: buf = io.BytesIO() paperdoorknob.process( - f"http://localhost:{self._port}", - s, - TIMEOUT, - buf, - 'pandoc') + f"http://localhost:{self._port}", f, buf, 'pandoc') self.assertEqual(buf.getvalue(), b'''\\documentclass{article} \\begin{document} This is glowfic @@ -81,10 +42,10 @@ Pretty sure. ''') def testPDF(self) -> None: - with requests.session() as s: + with DirectFetcher(TIMEOUT) as f: with open("test.tex", 'wb') as out: paperdoorknob.process( - f"http://localhost:{self._port}", s, TIMEOUT, out, 'pandoc') + f"http://localhost:{self._port}", f, out, 'pandoc') subprocess.run(['pdflatex', 'test.tex'], stdin=subprocess.DEVNULL, check=True)