Issue 5250 - Add copyright update script

update-copyright/tests/test_update_copyright.py

+#!/usr/bin/env python3
+
+import os
+import re
+import datetime
+import subprocess
+import shutil
+import urllib.parse
+
+import pytest
+
+from update_copyright import extract_urls, text_replace, hg_commit, main
+
+data_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'data')
+
+
+def create_repo(path):
+    subprocess.check_call(['hg', 'init', path])
+    with open(os.path.join(path, '.hg', 'hgrc'), 'w+') as hgrc:
+        set_user = '[ui]\nusername = Test User <test@example.com>'
+        hgrc.write(set_user)
+    shutil.copy(os.path.join(data_path, 'sample_file.py'), path)
+    subprocess.check_call(['hg', 'commit', '-Am', 'Initial commit',
+                           '--repository', path])
+
+
+@pytest.fixture()
+def temp_dir(tmpdir):
+    temp_dir = tmpdir.mkdir('temp_dir')
+    return temp_dir
+
+
+@pytest.fixture()
+def temp_repo(tmpdir):
+    """"Returns a path to a temporary repo containing one sample file"""
+    temp_repo = tmpdir.mkdir('tmp_dir')
+    create_repo(str(temp_repo))
+    return temp_repo
+
+
+@pytest.fixture()
+def base_dir(tmpdir):
+    """Returns a temporary directory that contains one html page and two
+    repositories (one with push access, the other without)"""
+    repo = tmpdir.mkdir('tmp_dir')
+    temp_dir = str(repo)
+    subprocess.check_call(['cp', os.path.join(data_path, 'hg_page.html'),
+                           temp_dir])
+    repo_1 = temp_dir + '/repo_1'

[Vasily Kuznetsov, 2017/06/27 16:18:51]

Any reason why you're not using `os.path.join` here? Perhaps that would be more
portable and consistent.

[rosie, 2017/07/03 15:33:44]

Done.

[rosie, 2017/07/03 15:33:44]

repo_1 and repo_2 were only needed as strings, so this was a bit shorter, but
you're right about switching to os.path.join being more portable and consistent.

+    repo_2 = temp_dir + '/repo_2'
+    subprocess.check_call(['mkdir', repo_1])

[Vasily Kuznetsov, 2017/06/27 16:18:51]

You can also use `os.mkdir` for this -- simpler than calling an external
process. Also, can this call be moved into `create_repo` function to avoid
duplication here?

[rosie, 2017/07/03 15:33:43]

Done.

[rosie, 2017/07/03 15:33:44]

Switched it to `os.mkdir`. The `create_repo` function is used in both fixtures,
once in `temp_repo`, and twice in `base_dir`. But the `temp_repo` function
doesn't need both `repo_1` and `repo_2`, so I left that part in the fixture.

[Vasily Kuznetsov, 2017/07/03 19:25:45]

It looks like the same pattern is followed in the other fixture, first a
directory is created and then `create_repo` is called on it. Creating the
directory inside of `create_repo` instead would replace these three lines
(creating "repo_1", "repo_2" and "tmp_repo") with just one line inside of the
function. Or am I missing something?

+    subprocess.check_call(['mkdir', repo_2])
+    create_repo(repo_1)
+    create_repo(repo_2)
+
+    # Make repo_2 read-only
+    subprocess.check_call(['touch', os.path.join(repo_2, '.hg/hgrc')])

[Vasily Kuznetsov, 2017/06/27 16:18:51]

Do we actually need to touch `hgrc` before we write into it? Wouldn't the next
line just create the file anyway?

[rosie, 2017/07/03 15:33:43]

Done.

+    with open(os.path.join(repo_2, '.hg/hgrc'), 'w') as hgrc:
+        hook = '[hooks]\npretxnchangegroup = return True'
+        hgrc.write(hook)
+    return repo
+
+
+def test_extract_urls():
+    data_url = urllib.parse.urljoin('file:///', data_path)
+    urls = [os.path.join(data_url + '/repo_1/'),

[Vasily Kuznetsov, 2017/06/27 16:18:51]

Seems like we're passing only one argument to `os.path.join` here. Wouldn't it
do nothing in this case?

[rosie, 2017/07/03 15:33:44]

Done.

+            os.path.join(data_url + '/repo_2/')]
+    assert urls == extract_urls(os.path.join(data_url, 'hg_page.html'))
+
+
+def test_text_replacement(temp_repo):
+    filename = temp_repo.join('sample_file.py').strpath
+    text_replace(temp_repo.strpath, filename)
+    with open(filename) as file:
+        text = file.read()
+        pattern = re.compile(r'(copyright.*?\d{4})(?:-\d{4})?\s+eyeo gmbh',
+                             re.I)
+        for year in re.finditer(pattern, text):
+            dates = re.search(r'(\d{4})-(\d{4})', year.group(0))
+            assert dates.group(2) == str(datetime.datetime.now().year)
+
+
+def test_hg_commit(temp_repo, temp_dir):
+    directory = str(temp_dir)
+    subprocess.check_call(['hg', 'clone', temp_repo.strpath, directory])

[Vasily Kuznetsov, 2017/06/27 16:18:51]

You could also assign `temp_repo.strpath` to a variable, like you do with
`directory` and reduce the duplication a bit.

[rosie, 2017/07/03 15:33:43]

Done.

+    open(os.path.join(directory, 'foo'), 'w').close()
+    subprocess.check_call(['hg', 'add', '--repository', directory])
+    hg_commit(directory, temp_repo.strpath)
+
+    # Make sure both files contain the commmit message from hg log
+    log_1 = subprocess.run(['hg', 'log', '--repository', temp_repo.strpath],
+                           stdout=subprocess.PIPE)
+    assert 'Noissue - Updated copyright year' in str(log_1.stdout)
+
+
+def test_all(base_dir):
+    main(urllib.parse.urljoin('file:///',
+         base_dir.join('hg_page.html').strpath),
+         base_dir)
+
+    # assert hg log for repo_1
+    log_1 = subprocess.run(['hg', 'log', '--repository',
+                            base_dir.join('repo_1').strpath],
+                           stdout=subprocess.PIPE)
+    assert 'Noissue - Updated copyright year' in str(log_1.stdout)
+
+    # assert the .patch file for repo_2
+    assert'Noissue - Updated copyright year' in open('repo_2.patch').read()
+    subprocess.call(['rm', 'repo_2.patch'])  # cleanup