cpython/Lib/test/test_tools/test_i18n.py

"""Tests to cover the Tools/i18n package"""

import os
import sys
import unittest
import textwrap

from test.support.script_helper import assert_python_ok
from test.test_tools import skip_if_missing, toolsdir
from test.support import temp_cwd


skip_if_missing()


class Test_pygettext(unittest.TestCase):
    """Tests for the pygettext.py tool"""

    script = os.path.join(toolsdir,'i18n', 'pygettext.py')

    def get_header(self, data):
        """ utility: return the header of a .po file as a dictionary """
        headers = {}
        for line in data.split('\n'):
            if not line or line.startswith(('#', 'msgid','msgstr')):
                continue
            line = line.strip('"')
            key, val = line.split(':',1)
            headers[key] = val.strip()
        return headers

    def get_msgids(self, data):
        """ utility: return all msgids in .po file as a list of strings """
        msgids = []
        reading_msgid = False
        cur_msgid = []
        for line in data.split('\n'):
            if reading_msgid:
                if line.startswith('"'):
                    cur_msgid.append(line.strip('"'))
                else:
                    msgids.append('\n'.join(cur_msgid))
                    cur_msgid = []
                    reading_msgid = False
                    continue
            if line.startswith('msgid '):
                line = line[len('msgid '):]
                cur_msgid.append(line.strip('"'))
                reading_msgid = True
        else:
            if reading_msgid:
                msgids.append('\n'.join(cur_msgid))

        return msgids

    def extract_docstrings_from_str(self, module_content):
        """ utility: return all msgids extracted from module_content """
        filename = 'test_docstrings.py'
        with temp_cwd(None) as cwd:
            with open(filename, 'w') as fp:
                fp.write(module_content)
            assert_python_ok(self.script, '-D', filename)
            with open('messages.pot') as fp:
                data = fp.read()
        return self.get_msgids(data)

    def test_header(self):
        """Make sure the required fields are in the header, according to:
           http://www.gnu.org/software/gettext/manual/gettext.html#Header-Entry
        """
        with temp_cwd(None) as cwd:
            assert_python_ok(self.script)
            with open('messages.pot') as fp:
                data = fp.read()
            header = self.get_header(data)

            self.assertIn("Project-Id-Version", header)
            self.assertIn("POT-Creation-Date", header)
            self.assertIn("PO-Revision-Date", header)
            self.assertIn("Last-Translator", header)
            self.assertIn("Language-Team", header)
            self.assertIn("MIME-Version", header)
            self.assertIn("Content-Type", header)
            self.assertIn("Content-Transfer-Encoding", header)
            self.assertIn("Generated-By", header)

            # not clear if these should be required in POT (template) files
            #self.assertIn("Report-Msgid-Bugs-To", header)
            #self.assertIn("Language", header)

            #"Plural-Forms" is optional

    @unittest.skipIf(sys.platform.startswith('aix'),
                     'bpo-29972: broken test on AIX')
    def test_POT_Creation_Date(self):
        """ Match the date format from xgettext for POT-Creation-Date """
        from datetime import datetime
        with temp_cwd(None) as cwd:
            assert_python_ok(self.script)
            with open('messages.pot') as fp:
                data = fp.read()
            header = self.get_header(data)
            creationDate = header['POT-Creation-Date']

            # peel off the escaped newline at the end of string
            if creationDate.endswith('\\n'):
                creationDate = creationDate[:-len('\\n')]

            # This will raise if the date format does not exactly match.
            datetime.strptime(creationDate, '%Y-%m-%d %H:%M%z')

    def test_funcdocstring_annotated_args(self):
        """ Test docstrings for functions with annotated args """
        msgids = self.extract_docstrings_from_str(textwrap.dedent('''\
        def foo(bar: str):
            """doc"""
        '''))
        self.assertIn('doc', msgids)

    def test_funcdocstring_annotated_return(self):
        """ Test docstrings for functions with annotated return type """
        msgids = self.extract_docstrings_from_str(textwrap.dedent('''\
        def foo(bar) -> str:
            """doc"""
        '''))
        self.assertIn('doc', msgids)

    def test_funcdocstring_defvalue_args(self):
        """ Test docstring for functions with default arg values """
        msgids = self.extract_docstrings_from_str(textwrap.dedent('''\
        def foo(bar=()):
            """doc"""
        '''))
        self.assertIn('doc', msgids)

    def test_funcdocstring_multiple_funcs(self):
        """ Test docstring extraction for multiple functions combining
        annotated args, annotated return types and default arg values
        """
        msgids = self.extract_docstrings_from_str(textwrap.dedent('''\
        def foo1(bar: tuple=()) -> str:
            """doc1"""

        def foo2(bar: List[1:2]) -> (lambda x: x):
            """doc2"""

        def foo3(bar: 'func'=lambda x: x) -> {1: 2}:
            """doc3"""
        '''))
        self.assertIn('doc1', msgids)
        self.assertIn('doc2', msgids)
        self.assertIn('doc3', msgids)

    def test_classdocstring_early_colon(self):
        """ Test docstring extraction for a class with colons occuring within
        the parentheses.
        """
        msgids = self.extract_docstrings_from_str(textwrap.dedent('''\
        class D(L[1:2], F({1: 2}), metaclass=M(lambda x: x)):
            """doc"""
        '''))
        self.assertIn('doc', msgids)