Merge pull request #3224 from pypa/feature/distutils-e1d5c9b1f6

Merge with distutils@e1d5c9b1f6

changelog.d/3224.change.rst

@@ -0,0 +1 @@
+Merge changes from pypa/distutils@e1d5c9b1f6

setuptools/_distutils/command/build_scripts.py

@@ -2,7 +2,8 @@
 
 Implements the Distutils 'build_scripts' command."""
 
-import os, re
+import os
+import re
 from stat import ST_MODE
 from distutils import sysconfig
 from distutils.core import Command
@@ -11,8 +12,14 @@
 from distutils import log
 import tokenize
 
-# check if Python is called on the first line with this expression
-first_line_re = re.compile(b'^#!.*python[0-9.]*([ \t].*)?$')
+shebang_pattern = re.compile('^#!.*python[0-9.]*([ \t].*)?$')
+"""
+Pattern matching a Python interpreter indicated in first line of a script.
+"""
+
+# for Setuptools compatibility
+first_line_re = shebang_pattern
+
 
 class build_scripts(Command):
 
@@ -26,13 +33,11 @@ class build_scripts(Command):
 
     boolean_options = ['force']
 
-
     def initialize_options(self):
         self.build_dir = None
         self.scripts = None
         self.force = None
         self.executable = None
-        self.outfiles = None
 
     def finalize_options(self):
         self.set_undefined_options('build',
@@ -49,104 +54,117 @@ def run(self):
             return
         self.copy_scripts()
 
-
     def copy_scripts(self):
-        r"""Copy each script listed in 'self.scripts'; if it's marked as a
-        Python script in the Unix way (first line matches 'first_line_re',
-        ie. starts with "\#!" and contains "python"), then adjust the first
-        line to refer to the current Python interpreter as we copy.
+        """
+        Copy each script listed in ``self.scripts``.
+
+        If a script is marked as a Python script (first line matches
+        'shebang_pattern', i.e. starts with ``#!`` and contains
+        "python"), then adjust in the copy the first line to refer to
+        the current Python interpreter.
         """
         self.mkpath(self.build_dir)
         outfiles = []
         updated_files = []
         for script in self.scripts:
-            adjust = False
-            script = convert_path(script)
-            outfile = os.path.join(self.build_dir, os.path.basename(script))
-            outfiles.append(outfile)
-
-            if not self.force and not newer(script, outfile):
-                log.debug("not copying %s (up-to-date)", script)
-                continue
-
-            # Always open the file, but ignore failures in dry-run mode --
-            # that way, we'll get accurate feedback if we can read the
-            # script.
-            try:
-                f = open(script, "rb")
-            except OSError:
-                if not self.dry_run:
-                    raise
-                f = None
-            else:
-                encoding, lines = tokenize.detect_encoding(f.readline)
-                f.seek(0)
-                first_line = f.readline()
-                if not first_line:
-                    self.warn("%s is an empty file (skipping)" % script)
-                    continue
-
-                match = first_line_re.match(first_line)
-                if match:
-                    adjust = True
-                    post_interp = match.group(1) or b''
-
-            if adjust:
-                log.info("copying and adjusting %s -> %s", script,
-                         self.build_dir)
-                updated_files.append(outfile)
-                if not self.dry_run:
-                    if not sysconfig.python_build:
-                        executable = self.executable
-                    else:
-                        executable = os.path.join(
-                            sysconfig.get_config_var("BINDIR"),
-                           "python%s%s" % (sysconfig.get_config_var("VERSION"),
-                                           sysconfig.get_config_var("EXE")))
-                    executable = os.fsencode(executable)
-                    shebang = b"#!" + executable + post_interp + b"\n"
-                    # Python parser starts to read a script using UTF-8 until
-                    # it gets a #coding:xxx cookie. The shebang has to be the
-                    # first line of a file, the #coding:xxx cookie cannot be
-                    # written before. So the shebang has to be decodable from
-                    # UTF-8.
-                    try:
-                        shebang.decode('utf-8')
-                    except UnicodeDecodeError:
-                        raise ValueError(
-                            "The shebang ({!r}) is not decodable "
-                            "from utf-8".format(shebang))
-                    # If the script is encoded to a custom encoding (use a
-                    # #coding:xxx cookie), the shebang has to be decodable from
-                    # the script encoding too.
-                    try:
-                        shebang.decode(encoding)
-                    except UnicodeDecodeError:
-                        raise ValueError(
-                            "The shebang ({!r}) is not decodable "
-                            "from the script encoding ({})"
-                            .format(shebang, encoding))
-                    with open(outfile, "wb") as outf:
-                        outf.write(shebang)
-                        outf.writelines(f.readlines())
-                if f:
-                    f.close()
-            else:
-                if f:
-                    f.close()
-                updated_files.append(outfile)
-                self.copy_file(script, outfile)
-
-        if os.name == 'posix':
-            for file in outfiles:
-                if self.dry_run:
-                    log.info("changing mode of %s", file)
-                else:
-                    oldmode = os.stat(file)[ST_MODE] & 0o7777
-                    newmode = (oldmode | 0o555) & 0o7777
-                    if newmode != oldmode:
-                        log.info("changing mode of %s from %o to %o",
-                                 file, oldmode, newmode)
-                        os.chmod(file, newmode)
-        # XXX should we modify self.outfiles?
+            self._copy_script(script, outfiles, updated_files)
+
+        self._change_modes(outfiles)
+
         return outfiles, updated_files
+
+    def _copy_script(self, script, outfiles, updated_files):
+        shebang_match = None
+        script = convert_path(script)
+        outfile = os.path.join(self.build_dir, os.path.basename(script))
+        outfiles.append(outfile)
+
+        if not self.force and not newer(script, outfile):
+            log.debug("not copying %s (up-to-date)", script)
+            return
+
+        # Always open the file, but ignore failures in dry-run mode
+        # in order to attempt to copy directly.
+        try:
+            f = tokenize.open(script)
+        except OSError:
+            if not self.dry_run:
+                raise
+            f = None
+        else:
+            first_line = f.readline()
+            if not first_line:
+                self.warn("%s is an empty file (skipping)" % script)
+                return
+
+            shebang_match = shebang_pattern.match(first_line)
+
+        updated_files.append(outfile)
+        if shebang_match:
+            log.info("copying and adjusting %s -> %s", script,
+                     self.build_dir)
+            if not self.dry_run:
+                if not sysconfig.python_build:
+                    executable = self.executable
+                else:
+                    executable = os.path.join(
+                        sysconfig.get_config_var("BINDIR"),
+                        "python%s%s" % (
+                            sysconfig.get_config_var("VERSION"),
+                            sysconfig.get_config_var("EXE")))
+                post_interp = shebang_match.group(1) or ''
+                shebang = "#!" + executable + post_interp + "\n"
+                self._validate_shebang(shebang, f.encoding)
+                with open(outfile, "w", encoding=f.encoding) as outf:
+                    outf.write(shebang)
+                    outf.writelines(f.readlines())
+            if f:
+                f.close()
+        else:
+            if f:
+                f.close()
+            self.copy_file(script, outfile)
+
+    def _change_modes(self, outfiles):
+        if os.name != 'posix':
+            return
+
+        for file in outfiles:
+            self._change_mode(file)
+
+    def _change_mode(self, file):
+        if self.dry_run:
+            log.info("changing mode of %s", file)
+            return
+
+        oldmode = os.stat(file)[ST_MODE] & 0o7777
+        newmode = (oldmode | 0o555) & 0o7777
+        if newmode != oldmode:
+            log.info("changing mode of %s from %o to %o",
+                     file, oldmode, newmode)
+            os.chmod(file, newmode)
+
+    @staticmethod
+    def _validate_shebang(shebang, encoding):
+        # Python parser starts to read a script using UTF-8 until
+        # it gets a #coding:xxx cookie. The shebang has to be the
+        # first line of a file, the #coding:xxx cookie cannot be
+        # written before. So the shebang has to be encodable to
+        # UTF-8.
+        try:
+            shebang.encode('utf-8')
+        except UnicodeEncodeError:
+            raise ValueError(
+                "The shebang ({!r}) is not encodable "
+                "to utf-8".format(shebang))
+
+        # If the script is encoded to a custom encoding (use a
+        # #coding:xxx cookie), the shebang has to be encodable to
+        # the script encoding too.
+        try:
+            shebang.encode(encoding)
+        except UnicodeEncodeError:
+            raise ValueError(
+                "The shebang ({!r}) is not encodable "
+                "to the script encoding ({})"
+                .format(shebang, encoding))

setuptools/_distutils/command/check.py

@@ -2,6 +2,8 @@
 
 Implements the Distutils 'check' command.
 """
+from email.utils import getaddresses
+
 from distutils.core import Command
 from distutils.errors import DistutilsSetupError
 
@@ -96,19 +98,39 @@ def check_metadata(self):
 
         if missing:
             self.warn("missing required meta-data: %s"  % ', '.join(missing))
-        if metadata.author:
-            if not metadata.author_email:
-                self.warn("missing meta-data: if 'author' supplied, " +
-                          "'author_email' should be supplied too")
-        elif metadata.maintainer:
-            if not metadata.maintainer_email:
-                self.warn("missing meta-data: if 'maintainer' supplied, " +
-                          "'maintainer_email' should be supplied too")
-        else:
+        if not (
+            self._check_contact("author", metadata) or
+            self._check_contact("maintainer", metadata)
+        ):
             self.warn("missing meta-data: either (author and author_email) " +
                       "or (maintainer and maintainer_email) " +
                       "should be supplied")
 
+    def _check_contact(self, kind, metadata):
+        """
+        Returns True if the contact's name is specified and False otherwise.
+        This function will warn if the contact's email is not specified.
+        """
+        name = getattr(metadata, kind) or ''
+        email = getattr(metadata, kind + '_email') or ''
+
+        msg = ("missing meta-data: if '{}' supplied, " +
+               "'{}' should be supplied too")
+
+        if name and email:
+            return True
+
+        if name:
+            self.warn(msg.format(kind, kind + '_email'))
+            return True
+
+        addresses = [(alias, addr) for alias, addr in getaddresses([email])]
+        if any(alias and addr for alias, addr in addresses):
+            # The contact's name can be encoded in the email: `Name <email>`
+            return True
+
+        return False
+
     def check_restructuredtext(self):
         """Checks if the long string fields are reST-compliant."""
         data = self.distribution.get_long_description()

setuptools/_distutils/tests/test_check.py

@@ -71,6 +71,28 @@ def test_check_metadata(self):
         cmd = self._run(metadata)
         self.assertEqual(cmd._warnings, 0)
 
+    def test_check_author_maintainer(self):
+        for kind in ("author", "maintainer"):
+            # ensure no warning when author_email or maintainer_email is given
+            # (the spec allows these fields to take the form "Name <email>")
+            metadata = {'url': 'xxx',
+                        kind + '_email': 'Name <[email protected]>',
+                        'name': 'xxx', 'version': 'xxx'}
+            cmd = self._run(metadata)
+            self.assertEqual(cmd._warnings, 0)
+
+            # the check should warn if only email is given and it does not
+            # contain the name
+            metadata[kind + '_email'] = '[email protected]'
+            cmd = self._run(metadata)
+            self.assertEqual(cmd._warnings, 1)
+
+            # the check should warn if only the name is given
+            metadata[kind] = "Name"
+            del metadata[kind + '_email']
+            cmd = self._run(metadata)
+            self.assertEqual(cmd._warnings, 1)
+
     @unittest.skipUnless(HAS_DOCUTILS, "won't test without docutils")
     def test_check_document(self):
         pkg_info, dist = self.create_dist()