[Feature] Add pyproject.toml (#1542)

* add pyproject.toml

* fix

.github/workflows/publish.yaml

@@ -22,7 +22,10 @@ jobs:
       - name: Install wheel
         run: pip install wheel && pip install -r requirements/framework.txt
       - name: Build ModelScope
-        run: python setup.py sdist bdist_wheel
+        # Build AST template before packaging
+        run: |
+          python -c "from modelscope.utils.ast_utils import generate_ast_template; generate_ast_template()"
+          python setup.py sdist bdist_wheel
       - name: Publish package to PyPI
         run: |
           pip install twine

MANIFEST.in

@@ -1,3 +1,3 @@
-recursive-include modelscope/configs *.py *.cu *.h *.cpp
+recursive-include modelscope/ops *.py *.cu *.h *.cpp
 recursive-include modelscope/cli/template *.tpl
 recursive-include modelscope/utils *.json

modelscope/cli/cli.py

@@ -16,6 +16,7 @@ from modelscope.cli.server import ServerCMD
 from modelscope.cli.upload import UploadCMD
 from modelscope.hub.constants import MODELSCOPE_ASCII
 from modelscope.utils.logger import get_logger
+from modelscope.version import __version__
 
 logger = get_logger(log_level=logging.WARNING)
 
@@ -24,6 +25,11 @@ def run_cmd():
     print(MODELSCOPE_ASCII)
     parser = argparse.ArgumentParser(
         'ModelScope Command Line tool', usage='modelscope <command> [<args>]')
+    parser.add_argument(
+        '-V',
+        '--version',
+        action='version',
+        version=f'ModelScope CLI {__version__}')
     parser.add_argument(
         '--token', default=None, help='Specify ModelScope SDK token.')
     subparsers = parser.add_subparsers(help='modelscope commands helpers')

modelscope/utils/ast_utils.py

@@ -768,7 +768,7 @@ def load_from_prebuilt(file_path=None):
     if os.path.exists(file_path):
         index = _load_index(file_path, with_template=True)
     else:
-        index = None
+        index = generate_ast_template()
     return index
 
 

pyproject.toml

@@ -0,0 +1,76 @@
+[project]
+name = "modelscope"
+dynamic = ["version", "dependencies", "optional-dependencies"]
+description = "ModelScope: bring the notion of Model-as-a-Service to life."
+readme = {file = "README.md", content-type = "text/markdown"}
+license = "Apache-2.0"
+license-files = ["LICENSE"]
+authors = [
+  {name = "ModelScope team"},
+  {email = "contact@modelscope.cn"}
+]
+keywords = ["python", "nlp", "science", "cv", "speech", "multi-modal"]
+requires-python = ">=3.9"
+classifiers = [
+    'Development Status :: 4 - Beta',
+    'Operating System :: OS Independent',
+    'Programming Language :: Python :: 3',
+    'Programming Language :: Python :: 3.9',
+    'Programming Language :: Python :: 3.10',
+    'Programming Language :: Python :: 3.11',
+]
+
+[project.urls]
+Homepage = "https://github.com/modelscope/modelscope"
+
+[project.scripts]
+modelscope = "modelscope.cli.cli:run_cmd"
+
+[build-system]
+requires = ["setuptools>=69", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools]
+include-package-data = true
+
+[tool.setuptools.packages.find]
+where = ["."]
+include = ["modelscope*"]
+
+[tool.setuptools.dynamic]
+version = {attr = "modelscope.version.__version__"}
+dependencies = {file = ["requirements/hub.txt"]}
+
+[tool.setuptools.dynamic.optional-dependencies]
+hub        = {file = ["requirements/hub.txt"]}
+datasets   = {file = ["requirements/datasets.txt"]}
+framework  = {file = ["requirements/framework.txt"]}
+server     = {file = ["requirements/server.txt"]}
+
+# domain specific with framework base
+cv          = {file = ["requirements/framework.txt", "requirements/cv.txt"]}
+nlp         = {file = ["requirements/framework.txt", "requirements/nlp.txt"]}
+multi-modal = {file = ["requirements/framework.txt", "requirements/multi-modal.txt"]}
+science     = {file = ["requirements/framework.txt", "requirements/science.txt"]}
+
+# audio specific with framework base
+audio_asr     = {file = ["requirements/framework.txt", "requirements/audio/audio_asr.txt"]}
+audio_codec   = {file = ["requirements/framework.txt", "requirements/audio/audio_codec.txt"]}
+audio_tts     = {file = ["requirements/framework.txt", "requirements/audio/audio_tts.txt"]}
+audio_kws     = {file = ["requirements/framework.txt", "requirements/audio/audio_kws.txt"]}
+audio_signal  = {file = ["requirements/framework.txt", "requirements/audio/audio_signal.txt"]}
+
+docs       = {file = ["requirements/docs.txt"]}
+tests      = {file = ["requirements/tests.txt"]}
+
+# skip audio requirements due to its hard dependency which may cause installation failure
+all = {file = [
+  "requirements/hub.txt",
+  "requirements/datasets.txt",
+  "requirements/framework.txt",
+  "requirements/cv.txt",
+  "requirements/nlp.txt",
+  "requirements/multi-modal.txt",
+  "requirements/science.txt",
+  "requirements/server.txt",
+]}

requirements/multi-modal.txt

@@ -2,11 +2,11 @@ accelerate
 cloudpickle
 decord>=0.6.0
 diffusers>=0.25.0
-ftfy>=6.0.3
 # 0.12.1 has issue of No such file or directory: 'fairseq/version.txt'
 # 0.12.2 not support py311
 #fairseq==0.12.2
-https://github.com/liyaodev/fairseq/releases/download/v0.12.3.1/fairseq-0.12.3.1-cp311-cp311-linux_x86_64.whl
+fairseq @ https://github.com/liyaodev/fairseq/releases/download/v0.12.3.1/fairseq-0.12.3.1-cp311-cp311-linux_x86_64.whl ; python_version == "3.11" and platform_system == "Linux" and platform_machine == "x86_64"
+ftfy>=6.0.3
 librosa==0.10.1
 opencv-python
 pycocoevalcap>=1.2

setup.py

@@ -1,239 +1,6 @@
 # Copyright (c) Alibaba, Inc. and its affiliates.
-# !/usr/bin/env python
+# -*- coding: utf-8 -*-
-import os
-import shutil
-import subprocess
-from setuptools import find_packages, setup
 
-from modelscope.utils.constant import Fields
+from setuptools import setup
 
-
+setup()
-def readme():
-    with open('README.md', encoding='utf-8') as f:
-        content = f.read()
-    return content
-
-
-version_file = 'modelscope/version.py'
-
-
-def get_git_hash():
-
-    def _minimal_ext_cmd(cmd):
-        # construct minimal environment
-        env = {}
-        for k in ['SYSTEMROOT', 'PATH', 'HOME']:
-            v = os.environ.get(k)
-            if v is not None:
-                env[k] = v
-        # LANGUAGE is used on win32
-        env['LANGUAGE'] = 'C'
-        env['LANG'] = 'C'
-        env['LC_ALL'] = 'C'
-        out = subprocess.Popen(
-            cmd, stdout=subprocess.PIPE, env=env).communicate()[0]
-        return out
-
-    try:
-        out = _minimal_ext_cmd(['git', 'rev-parse', 'HEAD'])
-        sha = out.strip().decode('ascii')
-    except OSError:
-        sha = 'unknown'
-
-    return sha
-
-
-def get_hash():
-    assert os.path.exists('.git'), '.git directory does not exist'
-    sha = get_git_hash()[:7]
-    return sha
-
-
-def get_version():
-    with open(version_file, 'r', encoding='utf-8') as f:
-        exec(compile(f.read(), version_file, 'exec'))
-    return locals()['__version__']
-
-
-def parse_requirements(fname='requirements.txt', with_version=True):
-    """
-    Parse the package dependencies listed in a requirements file but strips
-    specific versioning information.
-
-    Args:
-        fname (str): path to requirements file
-        with_version (bool, default=False): if True include version specs
-
-    Returns:
-        List[str]: list of requirements items
-
-    CommandLine:
-        python -c "import setup; print(setup.parse_requirements())"
-    """
-    import re
-    import sys
-    from os.path import exists
-    require_fpath = fname
-
-    def parse_line(line):
-        """
-        Parse information from a line in a requirements text file
-        """
-        if line.startswith('-r '):
-            # Allow specifying requirements in other files
-            target = line.split(' ')[1]
-            relative_base = os.path.dirname(fname)
-            absolute_target = os.path.join(relative_base, target)
-            for info in parse_require_file(absolute_target):
-                yield info
-        else:
-            info = {'line': line}
-            if line.startswith('-e '):
-                info['package'] = line.split('#egg=')[1]
-            else:
-                # Remove versioning from the package
-                pat = '(' + '|'.join(['>=', '==', '>']) + ')'
-                parts = re.split(pat, line, maxsplit=1)
-                parts = [p.strip() for p in parts]
-
-                info['package'] = parts[0]
-                if len(parts) > 1:
-                    op, rest = parts[1:]
-                    if ';' in rest:
-                        # Handle platform specific dependencies
-                        # http://setuptools.readthedocs.io/en/latest/setuptools.html#declaring-platform-specific-dependencies
-                        version, platform_deps = map(str.strip,
-                                                     rest.split(';'))
-                        info['platform_deps'] = platform_deps
-                    else:
-                        version = rest  # NOQA
-                    info['version'] = (op, version)
-            yield info
-
-    def parse_require_file(fpath):
-        with open(fpath, 'r', encoding='utf-8') as f:
-            for line in f.readlines():
-                line = line.strip()
-                if line.startswith('http'):
-                    print('skip http requirements %s' % line)
-                    continue
-                if line and not line.startswith('#') and not line.startswith(
-                        '--'):
-                    for info in parse_line(line):
-                        yield info
-                elif line and line.startswith('--find-links'):
-                    eles = line.split()
-                    for e in eles:
-                        e = e.strip()
-                        if 'http' in e:
-                            info = dict(dependency_links=e)
-                            yield info
-
-    def gen_packages_items():
-        items = []
-        deps_link = []
-        if exists(require_fpath):
-            for info in parse_require_file(require_fpath):
-                if 'dependency_links' not in info:
-                    parts = [info['package']]
-                    if with_version and 'version' in info:
-                        parts.extend(info['version'])
-                    if not sys.version.startswith('3.4'):
-                        # apparently package_deps are broken in 3.4
-                        platform_deps = info.get('platform_deps')
-                        if platform_deps is not None:
-                            parts.append(';' + platform_deps)
-                    item = ''.join(parts)
-                    items.append(item)
-                else:
-                    deps_link.append(info['dependency_links'])
-        return items, deps_link
-
-    return gen_packages_items()
-
-
-def pack_resource():
-    # pack resource such as configs and tools
-    root_dir = 'package/'
-    if os.path.isdir(root_dir):
-        shutil.rmtree(root_dir)
-    os.makedirs(root_dir)
-
-    proj_dir = root_dir + 'modelscope/'
-    shutil.copytree('./modelscope', proj_dir)
-    shutil.copytree('./configs', proj_dir + 'configs')
-    shutil.copytree('./requirements', 'package/requirements')
-    shutil.copy('./requirements.txt', 'package/requirements.txt')
-    shutil.copy('./MANIFEST.in', 'package/MANIFEST.in')
-    shutil.copy('./README.md', 'package/README.md')
-
-
-if __name__ == '__main__':
-    # write_version_py()
-    from modelscope.utils.ast_utils import generate_ast_template
-    generate_ast_template()
-    pack_resource()
-    os.chdir('package')
-    install_requires, deps_link = parse_requirements('requirements.txt')
-    extra_requires = {}
-    all_requires = []
-    for field in dir(Fields):
-        if field.startswith('_'):
-            continue
-        field = getattr(Fields, field)
-        extra_requires[field], _ = parse_requirements(
-            f'requirements/{field}.txt')
-
-        # skip audio requirements due to its hard dependency which
-        # result in mac/windows compatibility problems
-        if field != Fields.audio:
-            all_requires.append(extra_requires[field])
-    for subfiled in ['asr', 'kws', 'signal', 'tts']:
-        filed_name = f'audio_{subfiled}'
-        extra_requires[filed_name], _ = parse_requirements(
-            f'requirements/audio/{filed_name}.txt')
-    framework_requires = extra_requires['framework']
-    # add framework dependencies to every field
-    for field, requires in extra_requires.items():
-        if field not in [
-                'server', 'framework', 'hub', 'datasets'
-        ]:  # server need install model's field dependencies before.
-            extra_requires[field] = framework_requires + extra_requires[field]
-    extra_requires['all'] = all_requires
-
-    setup(
-        name='modelscope',
-        version=get_version(),
-        description=
-        'ModelScope: bring the notion of Model-as-a-Service to life.',
-        long_description=readme(),
-        long_description_content_type='text/markdown',
-        author='ModelScope team',
-        author_email='contact@modelscope.cn',
-        keywords='python,nlp,science,cv,speech,multi-modal',
-        url='https://github.com/modelscope/modelscope',
-        packages=find_packages(exclude=('configs', 'demo')),
-        include_package_data=True,
-        package_data={
-            '': ['*.h', '*.cpp', '*.cu'],
-        },
-        classifiers=[
-            'Development Status :: 4 - Beta',
-            'License :: OSI Approved :: Apache Software License',
-            'Operating System :: OS Independent',
-            'Programming Language :: Python :: 3',
-            'Programming Language :: Python :: 3.7',
-            'Programming Language :: Python :: 3.8',
-            'Programming Language :: Python :: 3.9',
-            'Programming Language :: Python :: 3.10',
-            'Programming Language :: Python :: 3.11',
-        ],
-        license='Apache License 2.0',
-        tests_require=parse_requirements('requirements/tests.txt'),
-        install_requires=install_requires,
-        extras_require=extra_requires,
-        entry_points={
-            'console_scripts': ['modelscope=modelscope.cli.cli:run_cmd']
-        },
-        dependency_links=deps_link,
-        zip_safe=False)