spaCy/spacy/cli/package.py

249 lines
8.6 KiB
Python
Raw Normal View History

2020-06-21 19:35:01 +00:00
from typing import Optional, Union, Any, Dict
2017-03-21 01:06:29 +00:00
import shutil
from pathlib import Path
2020-06-21 19:35:01 +00:00
from wasabi import Printer, get_raw_input
import srsly
2020-06-21 19:35:01 +00:00
import sys
from ._util import app, Arg, Opt
2020-06-21 19:35:01 +00:00
from ..schemas import validate, ModelMetaSchema
from .. import util
2017-05-07 21:25:29 +00:00
from .. import about
@app.command("package")
2020-06-21 19:35:01 +00:00
def package_cli(
# fmt: off
input_dir: Path = Arg(..., help="Directory with pipeline data", exists=True, file_okay=False),
2020-06-21 19:35:01 +00:00
output_dir: Path = Arg(..., help="Output parent directory", exists=True, file_okay=False),
meta_path: Optional[Path] = Opt(None, "--meta-path", "--meta", "-m", help="Path to meta.json", exists=True, dir_okay=False),
create_meta: bool = Opt(False, "--create-meta", "-c", "-C", help="Create meta.json, even if one exists"),
version: Optional[str] = Opt(None, "--version", "-v", help="Package version to override meta"),
no_sdist: bool = Opt(False, "--no-sdist", "-NS", help="Don't build .tar.gz sdist, can be set if you want to run this step manually"),
force: bool = Opt(False, "--force", "-f", "-F", help="Force overwriting existing data in output directory"),
# fmt: on
):
"""
Generate an installable Python package for a pipeline. Includes binary data,
2020-07-12 11:53:49 +00:00
meta and required installation files. A new directory will be created in the
specified output directory, and the data will be copied over. If
2020-07-12 11:53:49 +00:00
--create-meta is set and a meta.json already exists in the output directory,
the existing values will be used as the defaults in the command-line prompt.
After packaging, "python setup.py sdist" is run in the package directory,
which will create a .tar.gz archive that can be installed via "pip install".
"""
2020-06-21 19:35:01 +00:00
package(
input_dir,
output_dir,
meta_path=meta_path,
version=version,
2020-06-21 19:35:01 +00:00
create_meta=create_meta,
create_sdist=not no_sdist,
2020-06-21 19:35:01 +00:00
force=force,
silent=False,
)
def package(
input_dir: Path,
output_dir: Path,
meta_path: Optional[Path] = None,
version: Optional[str] = None,
2020-06-21 19:35:01 +00:00
create_meta: bool = False,
create_sdist: bool = True,
2020-06-21 19:35:01 +00:00
force: bool = False,
silent: bool = True,
) -> None:
msg = Printer(no_print=silent, pretty=not silent)
2017-05-07 21:25:29 +00:00
input_path = util.ensure_path(input_dir)
output_path = util.ensure_path(output_dir)
meta_path = util.ensure_path(meta_path)
2017-05-07 21:25:29 +00:00
if not input_path or not input_path.exists():
msg.fail("Can't locate pipeline data", input_path, exits=1)
2017-05-07 21:25:29 +00:00
if not output_path or not output_path.exists():
msg.fail("Output directory not found", output_path, exits=1)
2017-05-07 21:25:29 +00:00
if meta_path and not meta_path.exists():
msg.fail("Can't find pipeline meta.json", meta_path, exits=1)
2020-06-21 19:35:01 +00:00
meta_path = meta_path or input_dir / "meta.json"
if not meta_path.exists() or not meta_path.is_file():
msg.fail("Can't load pipeline meta.json", meta_path, exits=1)
2020-06-21 19:35:01 +00:00
meta = srsly.read_json(meta_path)
meta = get_meta(input_dir, meta)
if version is not None:
meta["version"] = version
2020-06-21 19:35:01 +00:00
if not create_meta: # only print if user doesn't want to overwrite
msg.good("Loaded meta.json from file", meta_path)
else:
meta = generate_meta(meta, msg)
2020-06-21 19:35:01 +00:00
errors = validate(ModelMetaSchema, meta)
if errors:
msg.fail("Invalid pipeline meta.json")
print("\n".join(errors))
sys.exit(1)
💫 New JSON helpers, training data internals & CLI rewrite (#2932) * Support nowrap setting in util.prints * Tidy up and fix whitespace * Simplify script and use read_jsonl helper * Add JSON schemas (see #2928) * Deprecate Doc.print_tree Will be replaced with Doc.to_json, which will produce a unified format * Add Doc.to_json() method (see #2928) Converts Doc objects to JSON using the same unified format as the training data. Method also supports serializing selected custom attributes in the doc._. space. * Remove outdated test * Add write_json and write_jsonl helpers * WIP: Update spacy train * Tidy up spacy train * WIP: Use wasabi for formatting * Add GoldParse helpers for JSON format * WIP: add debug-data command * Fix typo * Add missing import * Update wasabi pin * Add missing import * 💫 Refactor CLI (#2943) To be merged into #2932. ## Description - [x] refactor CLI To use [`wasabi`](https://github.com/ines/wasabi) - [x] use [`black`](https://github.com/ambv/black) for auto-formatting - [x] add `flake8` config - [x] move all messy UD-related scripts to `cli.ud` - [x] make converters function that take the opened file and return the converted data (instead of having them handle the IO) ### Types of change enhancement ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [x] My changes don't require a change to the documentation, or if they do, I've added all required information. * Update wasabi pin * Delete old test * Update errors * Fix typo * Tidy up and format remaining code * Fix formatting * Improve formatting of messages * Auto-format remaining code * Add tok2vec stuff to spacy.train * Fix typo * Update wasabi pin * Fix path checks for when train() is called as function * Reformat and tidy up pretrain script * Update argument annotations * Raise error if model language doesn't match lang * Document new train command
2018-11-30 19:16:14 +00:00
model_name = meta["lang"] + "_" + meta["name"]
model_name_v = model_name + "-" + meta["version"]
2020-06-21 19:35:01 +00:00
main_path = output_dir / model_name_v
package_path = main_path / model_name
if package_path.exists():
if force:
shutil.rmtree(str(package_path))
else:
💫 New JSON helpers, training data internals & CLI rewrite (#2932) * Support nowrap setting in util.prints * Tidy up and fix whitespace * Simplify script and use read_jsonl helper * Add JSON schemas (see #2928) * Deprecate Doc.print_tree Will be replaced with Doc.to_json, which will produce a unified format * Add Doc.to_json() method (see #2928) Converts Doc objects to JSON using the same unified format as the training data. Method also supports serializing selected custom attributes in the doc._. space. * Remove outdated test * Add write_json and write_jsonl helpers * WIP: Update spacy train * Tidy up spacy train * WIP: Use wasabi for formatting * Add GoldParse helpers for JSON format * WIP: add debug-data command * Fix typo * Add missing import * Update wasabi pin * Add missing import * 💫 Refactor CLI (#2943) To be merged into #2932. ## Description - [x] refactor CLI To use [`wasabi`](https://github.com/ines/wasabi) - [x] use [`black`](https://github.com/ambv/black) for auto-formatting - [x] add `flake8` config - [x] move all messy UD-related scripts to `cli.ud` - [x] make converters function that take the opened file and return the converted data (instead of having them handle the IO) ### Types of change enhancement ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [x] My changes don't require a change to the documentation, or if they do, I've added all required information. * Update wasabi pin * Delete old test * Update errors * Fix typo * Tidy up and format remaining code * Fix formatting * Improve formatting of messages * Auto-format remaining code * Add tok2vec stuff to spacy.train * Fix typo * Update wasabi pin * Fix path checks for when train() is called as function * Reformat and tidy up pretrain script * Update argument annotations * Raise error if model language doesn't match lang * Document new train command
2018-11-30 19:16:14 +00:00
msg.fail(
"Package directory already exists",
"Please delete the directory and try again, or use the "
"`--force` flag to overwrite existing directories.",
💫 New JSON helpers, training data internals & CLI rewrite (#2932) * Support nowrap setting in util.prints * Tidy up and fix whitespace * Simplify script and use read_jsonl helper * Add JSON schemas (see #2928) * Deprecate Doc.print_tree Will be replaced with Doc.to_json, which will produce a unified format * Add Doc.to_json() method (see #2928) Converts Doc objects to JSON using the same unified format as the training data. Method also supports serializing selected custom attributes in the doc._. space. * Remove outdated test * Add write_json and write_jsonl helpers * WIP: Update spacy train * Tidy up spacy train * WIP: Use wasabi for formatting * Add GoldParse helpers for JSON format * WIP: add debug-data command * Fix typo * Add missing import * Update wasabi pin * Add missing import * 💫 Refactor CLI (#2943) To be merged into #2932. ## Description - [x] refactor CLI To use [`wasabi`](https://github.com/ines/wasabi) - [x] use [`black`](https://github.com/ambv/black) for auto-formatting - [x] add `flake8` config - [x] move all messy UD-related scripts to `cli.ud` - [x] make converters function that take the opened file and return the converted data (instead of having them handle the IO) ### Types of change enhancement ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [x] My changes don't require a change to the documentation, or if they do, I've added all required information. * Update wasabi pin * Delete old test * Update errors * Fix typo * Tidy up and format remaining code * Fix formatting * Improve formatting of messages * Auto-format remaining code * Add tok2vec stuff to spacy.train * Fix typo * Update wasabi pin * Fix path checks for when train() is called as function * Reformat and tidy up pretrain script * Update argument annotations * Raise error if model language doesn't match lang * Document new train command
2018-11-30 19:16:14 +00:00
exits=1,
)
Path.mkdir(package_path, parents=True)
2020-06-21 19:35:01 +00:00
shutil.copytree(str(input_dir), str(package_path / model_name_v))
create_file(main_path / "meta.json", srsly.json_dumps(meta, indent=2))
💫 New JSON helpers, training data internals & CLI rewrite (#2932) * Support nowrap setting in util.prints * Tidy up and fix whitespace * Simplify script and use read_jsonl helper * Add JSON schemas (see #2928) * Deprecate Doc.print_tree Will be replaced with Doc.to_json, which will produce a unified format * Add Doc.to_json() method (see #2928) Converts Doc objects to JSON using the same unified format as the training data. Method also supports serializing selected custom attributes in the doc._. space. * Remove outdated test * Add write_json and write_jsonl helpers * WIP: Update spacy train * Tidy up spacy train * WIP: Use wasabi for formatting * Add GoldParse helpers for JSON format * WIP: add debug-data command * Fix typo * Add missing import * Update wasabi pin * Add missing import * 💫 Refactor CLI (#2943) To be merged into #2932. ## Description - [x] refactor CLI To use [`wasabi`](https://github.com/ines/wasabi) - [x] use [`black`](https://github.com/ambv/black) for auto-formatting - [x] add `flake8` config - [x] move all messy UD-related scripts to `cli.ud` - [x] make converters function that take the opened file and return the converted data (instead of having them handle the IO) ### Types of change enhancement ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [x] My changes don't require a change to the documentation, or if they do, I've added all required information. * Update wasabi pin * Delete old test * Update errors * Fix typo * Tidy up and format remaining code * Fix formatting * Improve formatting of messages * Auto-format remaining code * Add tok2vec stuff to spacy.train * Fix typo * Update wasabi pin * Fix path checks for when train() is called as function * Reformat and tidy up pretrain script * Update argument annotations * Raise error if model language doesn't match lang * Document new train command
2018-11-30 19:16:14 +00:00
create_file(main_path / "setup.py", TEMPLATE_SETUP)
create_file(main_path / "MANIFEST.in", TEMPLATE_MANIFEST)
create_file(package_path / "__init__.py", TEMPLATE_INIT)
msg.good(f"Successfully created package '{model_name_v}'", main_path)
if create_sdist:
with util.working_dir(main_path):
util.run_command([sys.executable, "setup.py", "sdist"])
zip_file = main_path / "dist" / f"{model_name_v}.tar.gz"
msg.good(f"Successfully created zipped Python package", zip_file)
2020-06-21 19:35:01 +00:00
def create_file(file_path: Path, contents: str) -> None:
file_path.touch()
💫 New JSON helpers, training data internals & CLI rewrite (#2932) * Support nowrap setting in util.prints * Tidy up and fix whitespace * Simplify script and use read_jsonl helper * Add JSON schemas (see #2928) * Deprecate Doc.print_tree Will be replaced with Doc.to_json, which will produce a unified format * Add Doc.to_json() method (see #2928) Converts Doc objects to JSON using the same unified format as the training data. Method also supports serializing selected custom attributes in the doc._. space. * Remove outdated test * Add write_json and write_jsonl helpers * WIP: Update spacy train * Tidy up spacy train * WIP: Use wasabi for formatting * Add GoldParse helpers for JSON format * WIP: add debug-data command * Fix typo * Add missing import * Update wasabi pin * Add missing import * 💫 Refactor CLI (#2943) To be merged into #2932. ## Description - [x] refactor CLI To use [`wasabi`](https://github.com/ines/wasabi) - [x] use [`black`](https://github.com/ambv/black) for auto-formatting - [x] add `flake8` config - [x] move all messy UD-related scripts to `cli.ud` - [x] make converters function that take the opened file and return the converted data (instead of having them handle the IO) ### Types of change enhancement ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [x] My changes don't require a change to the documentation, or if they do, I've added all required information. * Update wasabi pin * Delete old test * Update errors * Fix typo * Tidy up and format remaining code * Fix formatting * Improve formatting of messages * Auto-format remaining code * Add tok2vec stuff to spacy.train * Fix typo * Update wasabi pin * Fix path checks for when train() is called as function * Reformat and tidy up pretrain script * Update argument annotations * Raise error if model language doesn't match lang * Document new train command
2018-11-30 19:16:14 +00:00
file_path.open("w", encoding="utf-8").write(contents)
def get_meta(
model_path: Union[str, Path], existing_meta: Dict[str, Any]
2020-06-21 19:35:01 +00:00
) -> Dict[str, Any]:
meta = {
"lang": "en",
"name": "pipeline",
"version": "0.0.0",
"description": "",
"author": "",
"email": "",
"url": "",
"license": "MIT",
}
meta.update(existing_meta)
nlp = util.load_model_from_path(Path(model_path))
meta["spacy_version"] = util.get_model_version_range(about.__version__)
💫 New JSON helpers, training data internals & CLI rewrite (#2932) * Support nowrap setting in util.prints * Tidy up and fix whitespace * Simplify script and use read_jsonl helper * Add JSON schemas (see #2928) * Deprecate Doc.print_tree Will be replaced with Doc.to_json, which will produce a unified format * Add Doc.to_json() method (see #2928) Converts Doc objects to JSON using the same unified format as the training data. Method also supports serializing selected custom attributes in the doc._. space. * Remove outdated test * Add write_json and write_jsonl helpers * WIP: Update spacy train * Tidy up spacy train * WIP: Use wasabi for formatting * Add GoldParse helpers for JSON format * WIP: add debug-data command * Fix typo * Add missing import * Update wasabi pin * Add missing import * 💫 Refactor CLI (#2943) To be merged into #2932. ## Description - [x] refactor CLI To use [`wasabi`](https://github.com/ines/wasabi) - [x] use [`black`](https://github.com/ambv/black) for auto-formatting - [x] add `flake8` config - [x] move all messy UD-related scripts to `cli.ud` - [x] make converters function that take the opened file and return the converted data (instead of having them handle the IO) ### Types of change enhancement ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [x] My changes don't require a change to the documentation, or if they do, I've added all required information. * Update wasabi pin * Delete old test * Update errors * Fix typo * Tidy up and format remaining code * Fix formatting * Improve formatting of messages * Auto-format remaining code * Add tok2vec stuff to spacy.train * Fix typo * Update wasabi pin * Fix path checks for when train() is called as function * Reformat and tidy up pretrain script * Update argument annotations * Raise error if model language doesn't match lang * Document new train command
2018-11-30 19:16:14 +00:00
meta["vectors"] = {
"width": nlp.vocab.vectors_length,
"vectors": len(nlp.vocab.vectors),
"keys": nlp.vocab.vectors.n_keys,
"name": nlp.vocab.vectors.name,
💫 New JSON helpers, training data internals & CLI rewrite (#2932) * Support nowrap setting in util.prints * Tidy up and fix whitespace * Simplify script and use read_jsonl helper * Add JSON schemas (see #2928) * Deprecate Doc.print_tree Will be replaced with Doc.to_json, which will produce a unified format * Add Doc.to_json() method (see #2928) Converts Doc objects to JSON using the same unified format as the training data. Method also supports serializing selected custom attributes in the doc._. space. * Remove outdated test * Add write_json and write_jsonl helpers * WIP: Update spacy train * Tidy up spacy train * WIP: Use wasabi for formatting * Add GoldParse helpers for JSON format * WIP: add debug-data command * Fix typo * Add missing import * Update wasabi pin * Add missing import * 💫 Refactor CLI (#2943) To be merged into #2932. ## Description - [x] refactor CLI To use [`wasabi`](https://github.com/ines/wasabi) - [x] use [`black`](https://github.com/ambv/black) for auto-formatting - [x] add `flake8` config - [x] move all messy UD-related scripts to `cli.ud` - [x] make converters function that take the opened file and return the converted data (instead of having them handle the IO) ### Types of change enhancement ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [x] My changes don't require a change to the documentation, or if they do, I've added all required information. * Update wasabi pin * Delete old test * Update errors * Fix typo * Tidy up and format remaining code * Fix formatting * Improve formatting of messages * Auto-format remaining code * Add tok2vec stuff to spacy.train * Fix typo * Update wasabi pin * Fix path checks for when train() is called as function * Reformat and tidy up pretrain script * Update argument annotations * Raise error if model language doesn't match lang * Document new train command
2018-11-30 19:16:14 +00:00
}
if about.__title__ != "spacy":
meta["parent_package"] = about.__title__
return meta
def generate_meta(existing_meta: Dict[str, Any], msg: Printer) -> Dict[str, Any]:
meta = existing_meta or {}
settings = [
("lang", "Pipeline language", meta.get("lang", "en")),
("name", "Pipeline name", meta.get("name", "pipeline")),
("version", "Package version", meta.get("version", "0.0.0")),
("description", "Package description", meta.get("description", None)),
("author", "Author", meta.get("author", None)),
("email", "Author email", meta.get("email", None)),
("url", "Author website", meta.get("url", None)),
("license", "License", meta.get("license", "MIT")),
]
msg.divider("Generating meta.json")
msg.text(
"Enter the package settings for your pipeline. The following information "
"will be read from your pipeline data: pipeline, vectors."
)
for setting, desc, default in settings:
💫 New JSON helpers, training data internals & CLI rewrite (#2932) * Support nowrap setting in util.prints * Tidy up and fix whitespace * Simplify script and use read_jsonl helper * Add JSON schemas (see #2928) * Deprecate Doc.print_tree Will be replaced with Doc.to_json, which will produce a unified format * Add Doc.to_json() method (see #2928) Converts Doc objects to JSON using the same unified format as the training data. Method also supports serializing selected custom attributes in the doc._. space. * Remove outdated test * Add write_json and write_jsonl helpers * WIP: Update spacy train * Tidy up spacy train * WIP: Use wasabi for formatting * Add GoldParse helpers for JSON format * WIP: add debug-data command * Fix typo * Add missing import * Update wasabi pin * Add missing import * 💫 Refactor CLI (#2943) To be merged into #2932. ## Description - [x] refactor CLI To use [`wasabi`](https://github.com/ines/wasabi) - [x] use [`black`](https://github.com/ambv/black) for auto-formatting - [x] add `flake8` config - [x] move all messy UD-related scripts to `cli.ud` - [x] make converters function that take the opened file and return the converted data (instead of having them handle the IO) ### Types of change enhancement ## Checklist <!--- Before you submit the PR, go over this checklist and make sure you can tick off all the boxes. [] -> [x] --> - [x] I have submitted the spaCy Contributor Agreement. - [x] I ran the tests, and all new and existing tests passed. - [x] My changes don't require a change to the documentation, or if they do, I've added all required information. * Update wasabi pin * Delete old test * Update errors * Fix typo * Tidy up and format remaining code * Fix formatting * Improve formatting of messages * Auto-format remaining code * Add tok2vec stuff to spacy.train * Fix typo * Update wasabi pin * Fix path checks for when train() is called as function * Reformat and tidy up pretrain script * Update argument annotations * Raise error if model language doesn't match lang * Document new train command
2018-11-30 19:16:14 +00:00
response = get_raw_input(desc, default)
meta[setting] = default if response == "" and default else response
return meta
TEMPLATE_SETUP = """
#!/usr/bin/env python
import io
import json
from os import path, walk
from shutil import copy
from setuptools import setup
def load_meta(fp):
with io.open(fp, encoding='utf8') as f:
return json.load(f)
def list_files(data_dir):
output = []
for root, _, filenames in walk(data_dir):
for filename in filenames:
if not filename.startswith('.'):
output.append(path.join(root, filename))
output = [path.relpath(p, path.dirname(data_dir)) for p in output]
output.append('meta.json')
return output
def list_requirements(meta):
parent_package = meta.get('parent_package', 'spacy')
requirements = [parent_package + meta['spacy_version']]
if 'setup_requires' in meta:
requirements += meta['setup_requires']
if 'requirements' in meta:
requirements += meta['requirements']
return requirements
def setup_package():
root = path.abspath(path.dirname(__file__))
meta_path = path.join(root, 'meta.json')
meta = load_meta(meta_path)
model_name = str(meta['lang'] + '_' + meta['name'])
model_dir = path.join(model_name, model_name + '-' + meta['version'])
copy(meta_path, path.join(model_name))
copy(meta_path, model_dir)
setup(
name=model_name,
description=meta.get('description'),
author=meta.get('author'),
author_email=meta.get('email'),
url=meta.get('url'),
version=meta['version'],
license=meta.get('license'),
packages=[model_name],
package_data={model_name: list_files(model_dir)},
install_requires=list_requirements(meta),
zip_safe=False,
entry_points={'spacy_models': ['{m} = {m}'.format(m=model_name)]}
)
if __name__ == '__main__':
setup_package()
""".strip()
TEMPLATE_MANIFEST = """
include meta.json
2020-08-18 12:39:40 +00:00
include config.cfg
""".strip()
TEMPLATE_INIT = """
from pathlib import Path
from spacy.util import load_model_from_init_py, get_model_meta
__version__ = get_model_meta(Path(__file__).parent)['version']
def load(**overrides):
return load_model_from_init_py(__file__, **overrides)
""".strip()