Upload folder using huggingface_hub
9823a7e verified 9 months ago
51.5 kB
import cython

import collections
import os
import re, sys, time
from glob import iglob
from io import StringIO
from os.path import relpath as _relpath
from .Cache import Cache, FingerprintFlags

from collections.abc import Iterable

try:
    import pythran
except:
    pythran = None

from .. import Utils
from ..Utils import (cached_function, cached_method, path_exists,
    safe_makedirs, copy_file_to_dir_if_newer, is_package_dir, write_depfile)
from ..Compiler import Errors
from ..Compiler.Main import Context
from ..Compiler import Options
from ..Compiler.Options import (CompilationOptions, default_options,
    get_directive_defaults)

join_path = cached_function(os.path.join)
copy_once_if_newer = cached_function(copy_file_to_dir_if_newer)
safe_makedirs_once = cached_function(safe_makedirs)


def _make_relative(file_paths, base=None):
    if not base:
        base = os.getcwd()
    if base[-1] != os.path.sep:
        base += os.path.sep
    return [_relpath(path, base) if path.startswith(base) else path
            for path in file_paths]


def extended_iglob(pattern):
    if '{' in pattern:
        m = re.match('(.*){([^}]+)}(.*)', pattern)
        if m:
            before, switch, after = m.groups()
            for case in switch.split(','):
                for path in extended_iglob(before + case + after):
                    yield path
            return

    # We always accept '/' and also '\' on Windows,
    # because '/' is generally common for relative paths.
    if '**/' in pattern or os.sep == '\\' and '**\\' in pattern:
        seen = set()
        first, rest = re.split(r'\*\*[%s]' % ('/\\\\' if os.sep == '\\' else '/'), pattern, 1)
        if first:
            first = iglob(first + os.sep)
        else:
            first = ['']
        for root in first:
            for path in extended_iglob(join_path(root, rest)):
                if path not in seen:
                    seen.add(path)
                    yield path
            for path in extended_iglob(join_path(root, '*', '**', rest)):
                if path not in seen:
                    seen.add(path)
                    yield path
    else:
        for path in iglob(pattern):
            yield path


def nonempty(it, error_msg="expected non-empty iterator"):
    empty = True
    for value in it:
        empty = False
        yield value
    if empty:
        raise ValueError(error_msg)


def update_pythran_extension(ext):
    if pythran is None:
        raise RuntimeError("You first need to install Pythran to use the np_pythran directive.")
    try:
        pythran_ext = pythran.config.make_extension(python=True)
    except TypeError:  # older pythran version only
        pythran_ext = pythran.config.make_extension()

    ext.include_dirs.extend(pythran_ext['include_dirs'])
    ext.extra_compile_args.extend(pythran_ext['extra_compile_args'])
    ext.extra_link_args.extend(pythran_ext['extra_link_args'])
    ext.define_macros.extend(pythran_ext['define_macros'])
    ext.undef_macros.extend(pythran_ext['undef_macros'])
    ext.library_dirs.extend(pythran_ext['library_dirs'])
    ext.libraries.extend(pythran_ext['libraries'])
    ext.language = 'c++'

    # These options are not compatible with the way normal Cython extensions work
    for bad_option in ["-fwhole-program", "-fvisibility=hidden"]:
        try:
            ext.extra_compile_args.remove(bad_option)
        except ValueError:
            pass


def parse_list(s):
    """
    >>> parse_list("")
    []
    >>> parse_list("a")
    ['a']
    >>> parse_list("a b c")
    ['a', 'b', 'c']
    >>> parse_list("[a, b, c]")
    ['a', 'b', 'c']
    >>> parse_list('a " " b')
    ['a', ' ', 'b']
    >>> parse_list('[a, ",a", "a,", ",", ]')
    ['a', ',a', 'a,', ',']
    """
    if len(s) >= 2 and s[0] == '[' and s[-1] == ']':
        s = s[1:-1]
        delimiter = ','
    else:
        delimiter = ' '
    s, literals = strip_string_literals(s)
    def unquote(literal):
        literal = literal.strip()
        if literal[0] in "'\"":
            return literals[literal[1:-1]]
        else:
            return literal
    return [unquote(item) for item in s.split(delimiter) if item.strip()]


transitive_str = object()
transitive_list = object()
bool_or = object()

distutils_settings = {
    'name':                 str,
    'sources':              list,
    'define_macros':        list,
    'undef_macros':         list,
    'libraries':            transitive_list,
    'library_dirs':         transitive_list,
    'runtime_library_dirs': transitive_list,
    'include_dirs':         transitive_list,
    'extra_objects':        list,
    'extra_compile_args':   transitive_list,
    'extra_link_args':      transitive_list,
    'export_symbols':       list,
    'depends':              transitive_list,
    'language':             transitive_str,
    'np_pythran':           bool_or
}


def _legacy_strtobool(val):
    # Used to be "distutils.util.strtobool", adapted for deprecation warnings.
    if val == "True":
        return True
    elif val == "False":
        return False

    import warnings
    warnings.warn("The 'np_python' option requires 'True' or 'False'", category=DeprecationWarning)
    val = val.lower()
    if val in ('y', 'yes', 't', 'true', 'on', '1'):
        return True
    elif val in ('n', 'no', 'f', 'false', 'off', '0'):
        return False
    else:
        raise ValueError("invalid truth value %r" % (val,))


class DistutilsInfo:

    def __init__(self, source=None, exn=None):
        self.values = {}
        if source is not None:
            source_lines = StringIO(source) if isinstance(source, str) else source
            for line in source_lines:
                line = line.lstrip()
                if not line:
                    continue
                if line[0] != '#':
                    break
                line = line[1:].lstrip()
                kind = next((k for k in ("distutils:","cython:") if line.startswith(k)), None)
                if kind is not None:
                    key, _, value = [s.strip() for s in line[len(kind):].partition('=')]
                    type = distutils_settings.get(key, None)
                    if line.startswith("cython:") and type is None: continue
                    if type in (list, transitive_list):
                        value = parse_list(value)
                        if key == 'define_macros':
                            value = [tuple(macro.split('=', 1))
                                     if '=' in macro else (macro, None)
                                     for macro in value]
                    if type is bool_or:
                        value = _legacy_strtobool(value)
                    self.values[key] = value
        elif exn is not None:
            for key in distutils_settings:
                if key in ('name', 'sources','np_pythran'):
                    continue
                value = getattr(exn, key, None)
                if value:
                    self.values[key] = value

    def merge(self, other):
        if other is None:
            return self
        for key, value in other.values.items():
            type = distutils_settings[key]
            if type is transitive_str and key not in self.values:
                self.values[key] = value
            elif type is transitive_list:
                if key in self.values:
                    # Change a *copy* of the list (Trac #845)
                    all = self.values[key][:]
                    for v in value:
                        if v not in all:
                            all.append(v)
                    value = all
                self.values[key] = value
            elif type is bool_or:
                self.values[key] = self.values.get(key, False) | value
        return self

    def subs(self, aliases):
        if aliases is None:
            return self
        resolved = DistutilsInfo()
        for key, value in self.values.items():
            type = distutils_settings[key]
            if type in [list, transitive_list]:
                new_value_list = []
                for v in value:
                    if v in aliases:
                        v = aliases[v]
                    if isinstance(v, list):
                        new_value_list += v
                    else:
                        new_value_list.append(v)
                value = new_value_list
            else:
                if value in aliases:
                    value = aliases[value]
            resolved.values[key] = value
        return resolved

    def apply(self, extension):
        for key, value in self.values.items():
            type = distutils_settings[key]
            if type in [list, transitive_list]:
                value = getattr(extension, key) + list(value)
            setattr(extension, key, value)


_FIND_TOKEN = cython.declare(object, re.compile(r"""
    (?P<comment> [#] ) |
    (?P<brace> [{}] ) |
    (?P<fstring> f )? (?P<quote> '+ | "+ )
""", re.VERBOSE).search)

_FIND_STRING_TOKEN = cython.declare(object, re.compile(r"""
    (?P<escape> [\\]+ ) (?P<escaped_quote> ['"] ) |
    (?P<fstring> f )? (?P<quote> '+ | "+ )
""", re.VERBOSE).search)

_FIND_FSTRING_TOKEN = cython.declare(object, re.compile(r"""
    (?P<braces> [{]+ | [}]+ ) |
    (?P<escape> [\\]+ ) (?P<escaped_quote> ['"] ) |
    (?P<fstring> f )? (?P<quote> '+ | "+ )
""", re.VERBOSE).search)


def strip_string_literals(code: str, prefix: str = '__Pyx_L'):
    """
    Normalizes every string literal to be of the form '__Pyx_Lxxx',
    returning the normalized code and a mapping of labels to
    string literals.
    """
    new_code: list = []
    literals: dict = {}
    counter: cython.Py_ssize_t = 0
    find_token = _FIND_TOKEN

    def append_new_label(literal):
        nonlocal counter
        counter += 1
        label = f"{prefix}{counter}_"
        literals[label] = literal
        new_code.append(label)

    def parse_string(quote_type: str, start: cython.Py_ssize_t, is_fstring: cython.bint) -> cython.Py_ssize_t:
        charpos: cython.Py_ssize_t = start

        find_token = _FIND_FSTRING_TOKEN if is_fstring else _FIND_STRING_TOKEN

        while charpos != -1:
            token = find_token(code, charpos)
            if token is None:
                # This probably indicates an unclosed string literal, i.e. a broken file.
                append_new_label(code[start:])
                charpos = -1
                break
            charpos = token.end()

            if token['escape']:
                if len(token['escape']) % 2 == 0 and token['escaped_quote'] == quote_type[0]:
                    # Quote is not actually escaped and might be part of a terminator, look at it next.
                    charpos -= 1

            elif is_fstring and token['braces']:
                # Formats or brace(s) in fstring.
                if len(token['braces']) % 2 == 0:
                    # Normal brace characters in string.
                    continue
                if token['braces'][-1] == '{':
                    if start < charpos-1:
                        append_new_label(code[start : charpos-1])
                    new_code.append('{')
                    start = charpos = parse_code(charpos, in_fstring=True)

            elif token['quote'].startswith(quote_type):
                # Closing quote found (potentially together with further, unrelated quotes).
                charpos = token.start('quote')
                if charpos > start:
                    append_new_label(code[start : charpos])
                new_code.append(quote_type)
                charpos += len(quote_type)
                break

        return charpos

    def parse_code(start: cython.Py_ssize_t, in_fstring: cython.bint = False) -> cython.Py_ssize_t:
        charpos: cython.Py_ssize_t = start
        end: cython.Py_ssize_t
        quote: str

        while charpos != -1:
            token = find_token(code, charpos)
            if token is None:
                new_code.append(code[start:])
                charpos = -1
                break
            charpos = end = token.end()

            if token['quote']:
                quote = token['quote']
                if len(quote) >= 6:
                    # Ignore empty tripple-quoted strings: '''''' or """"""
                    quote = quote[:len(quote) % 6]
                if quote and len(quote) != 2:
                    if len(quote) > 3:
                        end -= len(quote) - 3
                        quote = quote[:3]
                    new_code.append(code[start:end])
                    start = charpos = parse_string(quote, end, is_fstring=token['fstring'])

            elif token['comment']:
                new_code.append(code[start:end])
                charpos = code.find('\n', end)
                append_new_label(code[end : charpos if charpos != -1 else None])
                if charpos == -1:
                    break  # EOF
                start = charpos

            elif in_fstring and token['brace']:
                if token['brace'] == '}':
                    # Closing '}' of f-string.
                    charpos = end = token.start() + 1
                    new_code.append(code[start:end])  # with '}'
                    break
                else:
                    # Starting a calculated format modifier inside of an f-string format.
                    end = token.start() + 1
                    new_code.append(code[start:end])  # with '{'
                    start = charpos = parse_code(end, in_fstring=True)

        return charpos

    parse_code(0)
    return "".join(new_code), literals


# We need to allow spaces to allow for conditional compilation like
# IF ...:
#     cimport ...
dependency_regex = re.compile(
    r"(?:^ [ \t\f]* from     [ \t\f]+ cython\.cimports\.([\w.]+) [ \t\f]+ c?import ) |"
    r"(?:^ [ \t\f]* from     [ \t\f]+ ([\w.]+) [ \t\f]+ cimport ) |"
    r"(?:^ [ \t\f]* c?import [ \t\f]+ cython\.cimports\.([\w.]+) ) |"
    r"(?:^ [ \t\f]* cimport  [ \t\f]+ ([\w.]+ (?:[ \t\f]* , [ \t\f]* [\w.]+)*) ) |"
    r"(?:^ [ \t\f]* cdef     [ \t\f]+ extern [ \t\f]+ from [ \t\f]+ ['\"] ([^'\"]+) ['\"] ) |"
    r"(?:^ [ \t\f]* include  [ \t\f]+ ['\"] ([^'\"]+) ['\"] )",
    re.MULTILINE | re.VERBOSE)
dependency_after_from_regex = re.compile(
    r"(?:^ [ \t\f]+ \( ([\w., \t\f]*) \) [ \t\f]* [#\n]) |"
    r"(?:^ [ \t\f]+    ([\w., \t\f]*)    [ \t\f]* [#\n])",
    re.MULTILINE | re.VERBOSE)


def normalize_existing(base_path, rel_paths):
    return normalize_existing0(os.path.dirname(base_path), tuple(set(rel_paths)))


@cached_function
def normalize_existing0(base_dir, rel_paths):
    """
    Given some base directory ``base_dir`` and a list of path names
    ``rel_paths``, normalize each relative path name ``rel`` by
    replacing it by ``os.path.join(base, rel)`` if that file exists.
    Return a couple ``(normalized, needed_base)`` where ``normalized``
    if the list of normalized file names and ``needed_base`` is
    ``base_dir`` if we actually needed ``base_dir``. If no paths were
    changed (for example, if all paths were already absolute), then
    ``needed_base`` is ``None``.
    """
    normalized = []
    needed_base = None
    for rel in rel_paths:
        if os.path.isabs(rel):
            normalized.append(rel)
            continue
        path = join_path(base_dir, rel)
        if path_exists(path):
            normalized.append(os.path.normpath(path))
            needed_base = base_dir
        else:
            normalized.append(rel)
    return (normalized, needed_base)


def resolve_depends(depends, include_dirs):
    include_dirs = tuple(include_dirs)
    resolved = []
    for depend in depends:
        path = resolve_depend(depend, include_dirs)
        if path is not None:
            resolved.append(path)
    return resolved


@cached_function
def resolve_depend(depend, include_dirs):
    if depend[0] == '<' and depend[-1] == '>':
        return None
    for dir in include_dirs:
        path = join_path(dir, depend)
        if path_exists(path):
            return os.path.normpath(path)
    return None


@cached_function
def package(filename):
    dir = os.path.dirname(os.path.abspath(str(filename)))
    if dir != filename and is_package_dir(dir):
        return package(dir) + (os.path.basename(dir),)
    else:
        return ()


@cached_function
def fully_qualified_name(filename):
    module = os.path.splitext(os.path.basename(filename))[0]
    return '.'.join(package(filename) + (module,))


@cached_function
def parse_dependencies(source_filename):
    # Actual parsing is way too slow, so we use regular expressions.
    # The only catch is that we must strip comments and string
    # literals ahead of time.
    with Utils.open_source_file(source_filename, error_handling='ignore') as fh:
        source = fh.read()
    distutils_info = DistutilsInfo(source)
    source, literals = strip_string_literals(source)
    source = source.replace('\\\n', ' ').replace('\t', ' ')

    # TODO: pure mode
    cimports = []
    includes = []
    externs  = []
    for m in dependency_regex.finditer(source):
        pycimports_from, cimport_from, pycimports_list, cimport_list, extern, include = m.groups()
        if pycimports_from:
            cimport_from = pycimports_from
        if pycimports_list:
            cimport_list = pycimports_list

        if cimport_from:
            cimports.append(cimport_from)
            m_after_from = dependency_after_from_regex.search(source, pos=m.end())
            if m_after_from:
                multiline, one_line = m_after_from.groups()
                subimports = multiline or one_line
                cimports.extend("{}.{}".format(cimport_from, s.strip())
                                for s in subimports.split(','))

        elif cimport_list:
            cimports.extend(x.strip() for x in cimport_list.split(","))
        elif extern:
            externs.append(literals[extern])
        else:
            includes.append(literals[include])
    return cimports, includes, externs, distutils_info


class DependencyTree:

    def __init__(self, context, quiet=False):
        self.context = context
        self.quiet = quiet
        self._transitive_cache = {}

    def parse_dependencies(self, source_filename):
        if path_exists(source_filename):
            source_filename = os.path.normpath(source_filename)
        return parse_dependencies(source_filename)

    @cached_method
    def included_files(self, filename):
        # This is messy because included files are textually included, resolving
        # cimports (but not includes) relative to the including file.
        all = set()
        for include in self.parse_dependencies(filename)[1]:
            include_path = join_path(os.path.dirname(filename), include)
            if not path_exists(include_path):
                include_path = self.context.find_include_file(include, source_file_path=filename)
            if include_path:
                if '.' + os.path.sep in include_path:
                    include_path = os.path.normpath(include_path)
                all.add(include_path)
                all.update(self.included_files(include_path))
            elif not self.quiet:
                print("Unable to locate '%s' referenced from '%s'" % (filename, include))
        return all

    @cached_method
    def cimports_externs_incdirs(self, filename):
        # This is really ugly. Nested cimports are resolved with respect to the
        # includer, but includes are resolved with respect to the includee.
        cimports, includes, externs = self.parse_dependencies(filename)[:3]
        cimports = set(cimports)
        externs = set(externs)
        incdirs = set()
        for include in self.included_files(filename):
            included_cimports, included_externs, included_incdirs = self.cimports_externs_incdirs(include)
            cimports.update(included_cimports)
            externs.update(included_externs)
            incdirs.update(included_incdirs)
        externs, incdir = normalize_existing(filename, externs)
        if incdir:
            incdirs.add(incdir)
        return tuple(cimports), externs, incdirs

    def cimports(self, filename):
        return self.cimports_externs_incdirs(filename)[0]

    def package(self, filename):
        return package(filename)

    def fully_qualified_name(self, filename):
        return fully_qualified_name(filename)

    @cached_method
    def find_pxd(self, module, filename=None):
        is_relative = module[0] == '.'
        if is_relative and not filename:
            raise NotImplementedError("New relative imports.")
        if filename is not None:
            module_path = module.split('.')
            if is_relative:
                module_path.pop(0)  # just explicitly relative
            package_path = list(self.package(filename))
            while module_path and not module_path[0]:
                try:
                    package_path.pop()
                except IndexError:
                    return None   # FIXME: error?
                module_path.pop(0)
            relative = '.'.join(package_path + module_path)
            pxd = self.context.find_pxd_file(relative, source_file_path=filename)
            if pxd:
                return pxd
        if is_relative:
            return None   # FIXME: error?
        return self.context.find_pxd_file(module, source_file_path=filename)

    @cached_method
    def cimported_files(self, filename):
        filename_root, filename_ext = os.path.splitext(filename)
        if filename_ext in ('.pyx', '.py') and path_exists(filename_root + '.pxd'):
            pxd_list = [filename_root + '.pxd']
        else:
            pxd_list = []
        # Cimports generates all possible combinations package.module
        # when imported as from package cimport module.
        for module in self.cimports(filename):
            if module[:7] == 'cython.' or module == 'cython':
                continue
            pxd_file = self.find_pxd(module, filename)
            if pxd_file is not None:
                pxd_list.append(pxd_file)
        return tuple(pxd_list)

    @cached_method
    def immediate_dependencies(self, filename):
        all_deps = {filename}
        all_deps.update(self.cimported_files(filename))
        all_deps.update(self.included_files(filename))
        return all_deps

    def all_dependencies(self, filename):
        return self.transitive_merge(filename, self.immediate_dependencies, set.union)

    @cached_method
    def timestamp(self, filename):
        return os.path.getmtime(filename)

    def extract_timestamp(self, filename):
        return self.timestamp(filename), filename

    def newest_dependency(self, filename):
        return max([self.extract_timestamp(f) for f in self.all_dependencies(filename)])

    def distutils_info0(self, filename):
        info = self.parse_dependencies(filename)[3]
        kwds = info.values
        cimports, externs, incdirs = self.cimports_externs_incdirs(filename)
        basedir = os.getcwd()
        # Add dependencies on "cdef extern from ..." files
        if externs:
            externs = _make_relative(externs, basedir)
            if 'depends' in kwds:
                kwds['depends'] = list(set(kwds['depends']).union(externs))
            else:
                kwds['depends'] = list(externs)
        # Add include_dirs to ensure that the C compiler will find the
        # "cdef extern from ..." files
        if incdirs:
            include_dirs = list(kwds.get('include_dirs', []))
            for inc in _make_relative(incdirs, basedir):
                if inc not in include_dirs:
                    include_dirs.append(inc)
            kwds['include_dirs'] = include_dirs
        return info

    def distutils_info(self, filename, aliases=None, base=None):
        return (self.transitive_merge(filename, self.distutils_info0, DistutilsInfo.merge)
            .subs(aliases)
            .merge(base))

    def transitive_merge(self, node, extract, merge):
        try:
            seen = self._transitive_cache[extract, merge]
        except KeyError:
            seen = self._transitive_cache[extract, merge] = {}
        return self.transitive_merge_helper(
            node, extract, merge, seen, {}, self.cimported_files)[0]

    def transitive_merge_helper(self, node, extract, merge, seen, stack, outgoing):
        if node in seen:
            return seen[node], None
        deps = extract(node)
        if node in stack:
            return deps, node
        try:
            stack[node] = len(stack)
            loop = None
            for next in outgoing(node):
                sub_deps, sub_loop = self.transitive_merge_helper(next, extract, merge, seen, stack, outgoing)
                if sub_loop is not None:
                    if loop is not None and stack[loop] < stack[sub_loop]:
                        pass
                    else:
                        loop = sub_loop
                deps = merge(deps, sub_deps)
            if loop == node:
                loop = None
            if loop is None:
                seen[node] = deps
            return deps, loop
        finally:
            del stack[node]


_dep_tree = None

def create_dependency_tree(ctx=None, quiet=False):
    global _dep_tree
    if _dep_tree is None:
        if ctx is None:
            ctx = Context(["."], get_directive_defaults(),
                          options=CompilationOptions(default_options))
        _dep_tree = DependencyTree(ctx, quiet=quiet)
    return _dep_tree


# If this changes, change also docs/src/reference/compilation.rst
# which mentions this function
def default_create_extension(template, kwds):
    if 'depends' in kwds:
        include_dirs = kwds.get('include_dirs', []) + ["."]
        depends = resolve_depends(kwds['depends'], include_dirs)
        kwds['depends'] = sorted(set(depends + template.depends))

    t = template.__class__
    ext = t(**kwds)
    if hasattr(template, "py_limited_api"):
        ext.py_limited_api = template.py_limited_api
    metadata = dict(distutils=kwds, module_name=kwds['name'])
    return (ext, metadata)


# This may be useful for advanced users?
def create_extension_list(patterns, exclude=None, ctx=None, aliases=None, quiet=False, language=None,
                          exclude_failures=False):
    if language is not None:
        print('Warning: passing language={0!r} to cythonize() is deprecated. '
              'Instead, put "# distutils: language={0}" in your .pyx or .pxd file(s)'.format(language))
    if exclude is None:
        exclude = []
    if patterns is None:
        return [], {}
    elif isinstance(patterns, str) or not isinstance(patterns, Iterable):
        patterns = [patterns]

    from distutils.extension import Extension
    if 'setuptools' in sys.modules:
        # Support setuptools Extension instances as well.
        extension_classes = (
            Extension,  # should normally be the same as 'setuptools.extension._Extension'
            sys.modules['setuptools.extension']._Extension,
            sys.modules['setuptools'].Extension,
        )
    else:
        extension_classes = (Extension,)

    explicit_modules = {m.name for m in patterns if isinstance(m, extension_classes)}
    deps = create_dependency_tree(ctx, quiet=quiet)
    shared_utility_qualified_name = ctx.shared_utility_qualified_name

    to_exclude = set()
    if not isinstance(exclude, list):
        exclude = [exclude]
    for pattern in exclude:
        to_exclude.update(map(os.path.abspath, extended_iglob(pattern)))

    module_list = []
    module_metadata = {}

    # if no create_extension() function is defined, use a simple
    # default function.
    create_extension = ctx.options.create_extension or default_create_extension

    seen = set()
    for pattern in patterns:
        if isinstance(pattern, str):
            filepattern = pattern
            template = Extension(pattern, [])  # Fake Extension without sources
            name = '*'
            base = None
            ext_language = language
        elif isinstance(pattern, extension_classes):
            cython_sources = [s for s in pattern.sources
                              if os.path.splitext(s)[1] in ('.py', '.pyx')]
            if cython_sources:
                filepattern = cython_sources[0]
                if len(cython_sources) > 1:
                    print("Warning: Multiple cython sources found for extension '%s': %s\n"
                          "See https://cython.readthedocs.io/en/latest/src/userguide/sharing_declarations.html "
                          "for sharing declarations among Cython files." % (pattern.name, cython_sources))
            elif shared_utility_qualified_name and pattern.name == shared_utility_qualified_name:
                # This is the shared utility code file.
                m, _ = create_extension(pattern, dict(
                    name=shared_utility_qualified_name,
                    sources=pattern.sources or [
                        shared_utility_qualified_name.replace('.', os.sep) + ('.cpp' if pattern.language == 'c++' else '.c')],
                    language=pattern.language,
                ))
                m.np_pythran = False
                m.shared_utility_qualified_name = None
                module_list.append(m)
                continue
            else:
                # ignore non-cython modules
                module_list.append(pattern)
                continue
            template = pattern
            name = template.name
            base = DistutilsInfo(exn=template)
            ext_language = None  # do not override whatever the Extension says
        else:
            msg = str("pattern is not of type str nor subclass of Extension (%s)"
                      " but of type %s and class %s" % (repr(Extension),
                                                        type(pattern),
                                                        pattern.__class__))
            raise TypeError(msg)

        for file in nonempty(sorted(extended_iglob(filepattern)), "'%s' doesn't match any files" % filepattern):
            if os.path.abspath(file) in to_exclude:
                continue
            module_name = deps.fully_qualified_name(file)
            if '*' in name:
                if module_name in explicit_modules:
                    continue
            elif name:
                module_name = name

            Utils.raise_error_if_module_name_forbidden(module_name)

            if module_name not in seen:
                try:
                    kwds = deps.distutils_info(file, aliases, base).values
                except Exception:
                    if exclude_failures:
                        continue
                    raise
                if base is not None:
                    for key, value in base.values.items():
                        if key not in kwds:
                            kwds[key] = value

                kwds['name'] = module_name

                sources = [file] + [m for m in template.sources if m != filepattern]
                if 'sources' in kwds:
                    # allow users to add .c files etc.
                    for source in kwds['sources']:
                        if source not in sources:
                            sources.append(source)
                kwds['sources'] = sources

                if ext_language and 'language' not in kwds:
                    kwds['language'] = ext_language

                np_pythran = kwds.pop('np_pythran', False)

                # Create the new extension
                m, metadata = create_extension(template, kwds)
                m.np_pythran = np_pythran or getattr(m, 'np_pythran', False)
                m.shared_utility_qualified_name = shared_utility_qualified_name
                if m.np_pythran:
                    update_pythran_extension(m)
                module_list.append(m)

                # Store metadata (this will be written as JSON in the
                # generated C file but otherwise has no purpose)
                module_metadata[module_name] = metadata

                if file not in m.sources:
                    # Old setuptools unconditionally replaces .pyx with .c/.cpp
                    target_file = os.path.splitext(file)[0] + ('.cpp' if m.language == 'c++' else '.c')
                    try:
                        m.sources.remove(target_file)
                    except ValueError:
                        # never seen this in the wild, but probably better to warn about this unexpected case
                        print("Warning: Cython source file not found in sources list, adding %s" % file)
                    m.sources.insert(0, file)
                seen.add(name)
    return module_list, module_metadata


# This is the user-exposed entry point.
def cythonize(module_list, exclude=None, nthreads=0, aliases=None, quiet=False, force=None, language=None,
              exclude_failures=False, show_all_warnings=False, **options):
    """
    Compile a set of source modules into C/C++ files and return a list of distutils
    Extension objects for them.
    :param module_list: As module list, pass either a glob pattern, a list of glob
                        patterns or a list of Extension objects.  The latter
                        allows you to configure the extensions separately
                        through the normal distutils options.
                        You can also pass Extension objects that have
                        glob patterns as their sources. Then, cythonize
                        will resolve the pattern and create a
                        copy of the Extension for every matching file.
    :param exclude: When passing glob patterns as ``module_list``, you can exclude certain
                    module names explicitly by passing them into the ``exclude`` option.
    :param nthreads: The number of concurrent builds for parallel compilation
                     (requires the ``multiprocessing`` module).
    :param aliases: If you want to use compiler directives like ``# distutils: ...`` but
                    can only know at compile time (when running the ``setup.py``) which values
                    to use, you can use aliases and pass a dictionary mapping those aliases
                    to Python strings when calling :func:`cythonize`. As an example, say you
                    want to use the compiler
                    directive ``# distutils: include_dirs = ../static_libs/include/``
                    but this path isn't always fixed and you want to find it when running
                    the ``setup.py``. You can then do ``# distutils: include_dirs = MY_HEADERS``,
                    find the value of ``MY_HEADERS`` in the ``setup.py``, put it in a python
                    variable called ``foo`` as a string, and then call
                    ``cythonize(..., aliases={'MY_HEADERS': foo})``.
    :param quiet: If True, Cython won't print error, warning, or status messages during the
                  compilation.
    :param force: Forces the recompilation of the Cython modules, even if the timestamps
                  don't indicate that a recompilation is necessary.
    :param language: To globally enable C++ mode, you can pass ``language='c++'``. Otherwise, this
                     will be determined at a per-file level based on compiler directives.  This
                     affects only modules found based on file names.  Extension instances passed
                     into :func:`cythonize` will not be changed. It is recommended to rather
                     use the compiler directive ``# distutils: language = c++`` than this option.
    :param exclude_failures: For a broad 'try to compile' mode that ignores compilation
                             failures and simply excludes the failed extensions,
                             pass ``exclude_failures=True``. Note that this only
                             really makes sense for compiling ``.py`` files which can also
                             be used without compilation.
    :param show_all_warnings: By default, not all Cython warnings are printed.
                              Set to true to show all warnings.
    :param annotate: If ``True``, will produce a HTML file for each of the ``.pyx`` or ``.py``
                     files compiled. The HTML file gives an indication
                     of how much Python interaction there is in
                     each of the source code lines, compared to plain C code.
                     It also allows you to see the C/C++ code
                     generated for each line of Cython code. This report is invaluable when
                     optimizing a function for speed,
                     and for determining when to :ref:`release the GIL <nogil>`:
                     in general, a ``nogil`` block may contain only "white" code.
                     See examples in :ref:`determining_where_to_add_types` or
                     :ref:`primes`.
    :param annotate-fullc: If ``True`` will produce a colorized HTML version of
                           the source which includes entire generated C/C++-code.
    :param compiler_directives: Allow to set compiler directives in the ``setup.py`` like this:
                                ``compiler_directives={'embedsignature': True}``.
                                See :ref:`compiler-directives`.
    :param depfile: produce depfiles for the sources if True.
    :param cache: If ``True`` the cache enabled with default path. If the value is a path to a directory,
                  then the directory is used to cache generated ``.c``/``.cpp`` files. By default cache is disabled.
                  See :ref:`cython-cache`.
    """
    if exclude is None:
        exclude = []
    if 'include_path' not in options:
        options['include_path'] = ['.']
    if 'common_utility_include_dir' in options:
        safe_makedirs(options['common_utility_include_dir'])

    depfile = options.pop('depfile', None)

    if pythran is None:
        pythran_options = None
    else:
        pythran_options = CompilationOptions(**options)
        pythran_options.cplus = True
        pythran_options.np_pythran = True

    if force is None:
        force = os.environ.get("CYTHON_FORCE_REGEN") == "1"  # allow global overrides for build systems

    c_options = CompilationOptions(**options)
    cpp_options = CompilationOptions(**options); cpp_options.cplus = True
    ctx = Context.from_options(c_options)
    options = c_options
    shared_utility_qualified_name = ctx.shared_utility_qualified_name
    module_list, module_metadata = create_extension_list(
        module_list,
        exclude=exclude,
        ctx=ctx,
        quiet=quiet,
        exclude_failures=exclude_failures,
        language=language,
        aliases=aliases)

    fix_windows_unicode_modules(module_list)

    deps = create_dependency_tree(ctx, quiet=quiet)
    build_dir = getattr(options, 'build_dir', None)
    if options.cache and not (options.annotate or Options.annotate):
        # cache is enabled when:
        # * options.cache is True (the default path to the cache base dir is used)
        # * options.cache is the explicit path to the cache base dir
        # * annotations are not generated
        cache_path = None if options.cache is True else options.cache
        cache = Cache(cache_path, getattr(options, 'cache_size', None))
    else:
        cache = None

    def copy_to_build_dir(filepath, root=os.getcwd()):
        filepath_abs = os.path.abspath(filepath)
        if os.path.isabs(filepath):
            filepath = filepath_abs
        if filepath_abs.startswith(root):
            # distutil extension depends are relative to cwd
            mod_dir = join_path(build_dir,
                                os.path.dirname(_relpath(filepath, root)))
            copy_once_if_newer(filepath_abs, mod_dir)

    def file_in_build_dir(c_file):
        if not build_dir:
            return c_file
        if os.path.isabs(c_file):
            c_file = os.path.splitdrive(c_file)[1]
            c_file = c_file.split(os.sep, 1)[1]
        c_file = os.path.join(build_dir, c_file)
        dir = os.path.dirname(c_file)
        safe_makedirs_once(dir)
        return c_file

    modules_by_cfile = collections.defaultdict(list)
    to_compile = []
    for m in module_list:
        if build_dir:
            for dep in m.depends:
                copy_to_build_dir(dep)

        cy_sources = [
            source for source in m.sources
            if os.path.splitext(source)[1] in ('.pyx', '.py')]
        if len(cy_sources) == 1:
            # normal "special" case: believe the Extension module name to allow user overrides
            full_module_name = m.name
        else:
            # infer FQMN from source files
            full_module_name = None

        np_pythran = getattr(m, 'np_pythran', False)
        py_limited_api = getattr(m, 'py_limited_api', False)

        if np_pythran:
            options = pythran_options
        elif m.language == 'c++':
            options = cpp_options
        else:
            options = c_options

        new_sources = []
        for source in m.sources:
            base, ext = os.path.splitext(source)
            if ext in ('.pyx', '.py'):
                c_file = base + ('.cpp' if m.language == 'c++' or np_pythran else '.c')

                # setup for out of place build directory if enabled
                c_file = file_in_build_dir(c_file)

                # write out the depfile, if requested
                if depfile:
                    dependencies = deps.all_dependencies(source)
                    write_depfile(c_file, source, dependencies)

                # Missing files and those generated by other Cython versions should always be recreated.
                if Utils.file_generated_by_this_cython(c_file):
                    c_timestamp = os.path.getmtime(c_file)
                else:
                    c_timestamp = -1

                # Priority goes first to modified files, second to direct
                # dependents, and finally to indirect dependents.
                if c_timestamp < deps.timestamp(source):
                    dep_timestamp, dep = deps.timestamp(source), source
                    priority = 0
                else:
                    dep_timestamp, dep = deps.newest_dependency(source)
                    priority = 2 - (dep in deps.immediate_dependencies(source))
                if force or c_timestamp < dep_timestamp:
                    if not quiet and not force:
                        if source == dep:
                            print("Compiling %s because it changed." % Utils.decode_filename(source))
                        else:
                            print("Compiling %s because it depends on %s." % (
                                Utils.decode_filename(source),
                                Utils.decode_filename(dep),
                            ))
                    if not force and cache:
                        fingerprint = cache.transitive_fingerprint(
                                source, deps.all_dependencies(source), options,
                                FingerprintFlags(m.language or 'c', py_limited_api, np_pythran)
                        )
                    else:
                        fingerprint = None
                    to_compile.append((
                        priority, source, c_file, fingerprint, quiet,
                        options, not exclude_failures, module_metadata.get(m.name),
                        full_module_name, show_all_warnings))
                modules_by_cfile[c_file].append(m)
            elif shared_utility_qualified_name and m.name == shared_utility_qualified_name:
                # Generate shared utility code module now.
                c_file = file_in_build_dir(source)
                module_options = CompilationOptions(
                    options, shared_c_file_path=c_file, shared_utility_qualified_name=None)
                if not Utils.is_cython_generated_file(c_file):
                    print(f"Warning: Shared module source file is not a Cython file - not creating '{m.name}' as '{c_file}'")
                elif force or not Utils.file_generated_by_this_cython(c_file):
                    from .SharedModule import generate_shared_module
                    if not quiet:
                        print(f"Generating shared module '{m.name}'")
                    generate_shared_module(module_options)
            else:
                c_file = source
                if build_dir:
                    copy_to_build_dir(source)

            new_sources.append(c_file)

        m.sources = new_sources

    to_compile.sort()
    N = len(to_compile)

    # Drop "priority" sorting component of "to_compile" entries
    # and add a simple progress indicator and the remaining arguments.
    build_progress_indicator = ("[{0:%d}/%d] " % (len(str(N)), N)).format
    to_compile = [
        task[1:] + (build_progress_indicator(i), cache)
        for i, task in enumerate(to_compile, 1)
    ]

    if N <= 1:
        nthreads = 0
    try:
        from concurrent.futures import ProcessPoolExecutor
    except ImportError:
        nthreads = 0

    if nthreads:
        with ProcessPoolExecutor(
            max_workers=nthreads,
            initializer=_init_multiprocessing_helper,
        ) as proc_pool:
            try:
                list(proc_pool.map(cythonize_one_helper, to_compile, chunksize=1))
            except KeyboardInterrupt:
                proc_pool.terminate_workers()
                proc_pool.shutdown(cancel_futures=True)
                raise
    else:
        for args in to_compile:
            cythonize_one(*args)

    if exclude_failures:
        failed_modules = set()
        for c_file, modules in modules_by_cfile.items():
            if not os.path.exists(c_file):
                failed_modules.update(modules)
            elif os.path.getsize(c_file) < 200:
                f = open(c_file, 'r', encoding='iso8859-1')
                try:
                    if f.read(len('#error ')) == '#error ':
                        # dead compilation result
                        failed_modules.update(modules)
                finally:
                    f.close()
        if failed_modules:
            for module in failed_modules:
                module_list.remove(module)
            print("Failed compilations: %s" % ', '.join(sorted([
                module.name for module in failed_modules])))

    if cache:
        cache.cleanup_cache()

    # cythonize() is often followed by the (non-Python-buffered)
    # compiler output, flush now to avoid interleaving output.
    sys.stdout.flush()
    return module_list


def fix_windows_unicode_modules(module_list):
    # Hack around a distutils 3.[5678] bug on Windows for unicode module names.
    # https://bugs.python.org/issue39432
    if sys.platform != "win32":
        return
    if sys.version_info >= (3, 8, 2):
        return

    def make_filtered_list(ignored_symbol, old_entries):
        class FilteredExportSymbols(list):
            # export_symbols for unicode filename cause link errors on Windows
            # Cython doesn't need them (it already defines PyInit with the correct linkage)
            # so use this class as a temporary fix to stop them from being generated
            def __contains__(self, val):
                # so distutils doesn't "helpfully" add PyInit_<name>
                return val == ignored_symbol or list.__contains__(self, val)

        filtered_list = FilteredExportSymbols(old_entries)
        if old_entries:
            filtered_list.extend(name for name in old_entries if name != ignored_symbol)
        return filtered_list

    for m in module_list:
        if m.name.isascii():
            continue
        m.export_symbols = make_filtered_list(
            "PyInit_" + m.name.rsplit(".", 1)[-1],
            m.export_symbols,
        )


if os.environ.get('XML_RESULTS'):
    compile_result_dir = os.environ['XML_RESULTS']
    def record_results(func):
        def with_record(*args):
            t = time.time()
            success = True
            try:
                try:
                    func(*args)
                except:
                    success = False
            finally:
                t = time.time() - t
                module = fully_qualified_name(args[0])
                name = "cythonize." + module
                failures = 1 - success
                if success:
                    failure_item = ""
                else:
                    failure_item = "failure"
                output = open(os.path.join(compile_result_dir, name + ".xml"), "w")
                output.write("""
                    <?xml version="1.0" ?>
                    <testsuite name="%(name)s" errors="0" failures="%(failures)s" tests="1" time="%(t)s">
                    <testcase classname="%(name)s" name="cythonize">
                    %(failure_item)s
                    </testcase>
                    </testsuite>
                """.strip() % locals())
                output.close()
        return with_record
else:
    def record_results(func):
        return func


# TODO: Share context? Issue: pyx processing leaks into pxd module
@record_results
def cythonize_one(pyx_file, c_file,
                  fingerprint=None, quiet=False, options=None,
                  raise_on_failure=True, embedded_metadata=None,
                  full_module_name=None, show_all_warnings=False,
                  progress="", cache=None):
    from ..Compiler.Main import compile_single, default_options
    from ..Compiler.Errors import CompileError, PyrexError

    if not quiet:
        if cache and fingerprint and cache.lookup_cache(c_file, fingerprint):
            print(f"{progress}Found compiled {pyx_file} in cache")
        else:
            print(f"{progress}Cythonizing {Utils.decode_filename(pyx_file)}")
    if options is None:
        options = CompilationOptions(default_options)
    options.output_file = c_file
    options.embedded_metadata = embedded_metadata

    old_warning_level = Errors.LEVEL
    if show_all_warnings:
        Errors.LEVEL = 0

    any_failures = 0
    try:
        result = compile_single(pyx_file, options, full_module_name=full_module_name, cache=cache, fingerprint=fingerprint)
        if result.num_errors > 0:
            any_failures = 1
    except (OSError, PyrexError) as e:
        sys.stderr.write('%s\n' % e)
        any_failures = 1
        # XXX
        import traceback
        traceback.print_exc()
    except Exception:
        if raise_on_failure:
            raise
        import traceback
        traceback.print_exc()
        any_failures = 1
    finally:
        if show_all_warnings:
            Errors.LEVEL = old_warning_level

    if any_failures:
        if raise_on_failure:
            raise CompileError(None, pyx_file)
        elif os.path.exists(c_file):
            os.remove(c_file)


def cythonize_one_helper(m):
    import traceback
    try:
        return cythonize_one(*m)
    except Exception:
        traceback.print_exc()
        raise


def _init_multiprocessing_helper():
    # KeyboardInterrupt kills workers, so don't let them get it
    import signal
    signal.signal(signal.SIGINT, signal.SIG_IGN)