mirror of
https://github.com/explosion/spaCy.git
synced 2025-07-18 04:02:20 +03:00
Merge branch 'explosion:master' into feature/extend_fo_tokenizer_exceptions
This commit is contained in:
commit
a92540b9a3
|
@ -1,5 +1,5 @@
|
||||||
# fmt: off
|
# fmt: off
|
||||||
__title__ = "spacy"
|
__title__ = "spacy"
|
||||||
__version__ = "3.7.3"
|
__version__ = "3.7.4"
|
||||||
__download_url__ = "https://github.com/explosion/spacy-models/releases/download"
|
__download_url__ = "https://github.com/explosion/spacy-models/releases/download"
|
||||||
__compatibility__ = "https://raw.githubusercontent.com/explosion/spacy-models/master/compatibility.json"
|
__compatibility__ = "https://raw.githubusercontent.com/explosion/spacy-models/master/compatibility.json"
|
||||||
|
|
|
@ -1,5 +1,7 @@
|
||||||
from wasabi import msg
|
from wasabi import msg
|
||||||
|
|
||||||
|
# Needed for testing
|
||||||
|
from . import download as download_module # noqa: F401
|
||||||
from ._util import app, setup_cli # noqa: F401
|
from ._util import app, setup_cli # noqa: F401
|
||||||
from .apply import apply # noqa: F401
|
from .apply import apply # noqa: F401
|
||||||
from .assemble import assemble_cli # noqa: F401
|
from .assemble import assemble_cli # noqa: F401
|
||||||
|
|
|
@ -1,5 +1,6 @@
|
||||||
import sys
|
import sys
|
||||||
from typing import Optional, Sequence
|
from typing import Optional, Sequence
|
||||||
|
from urllib.parse import urljoin
|
||||||
|
|
||||||
import requests
|
import requests
|
||||||
import typer
|
import typer
|
||||||
|
@ -63,6 +64,13 @@ def download(
|
||||||
)
|
)
|
||||||
pip_args = pip_args + ("--no-deps",)
|
pip_args = pip_args + ("--no-deps",)
|
||||||
if direct:
|
if direct:
|
||||||
|
# Reject model names with '/', in order to prevent shenanigans.
|
||||||
|
if "/" in model:
|
||||||
|
msg.fail(
|
||||||
|
title="Model download rejected",
|
||||||
|
text=f"Cannot download model '{model}'. Models are expected to be file names, not URLs or fragments",
|
||||||
|
exits=True,
|
||||||
|
)
|
||||||
components = model.split("-")
|
components = model.split("-")
|
||||||
model_name = "".join(components[:-1])
|
model_name = "".join(components[:-1])
|
||||||
version = components[-1]
|
version = components[-1]
|
||||||
|
@ -153,7 +161,16 @@ def get_latest_version(model: str) -> str:
|
||||||
def download_model(
|
def download_model(
|
||||||
filename: str, user_pip_args: Optional[Sequence[str]] = None
|
filename: str, user_pip_args: Optional[Sequence[str]] = None
|
||||||
) -> None:
|
) -> None:
|
||||||
download_url = about.__download_url__ + "/" + filename
|
# Construct the download URL carefully. We need to make sure we don't
|
||||||
|
# allow relative paths or other shenanigans to trick us into download
|
||||||
|
# from outside our own repo.
|
||||||
|
base_url = about.__download_url__
|
||||||
|
# urljoin requires that the path ends with /, or the last path part will be dropped
|
||||||
|
if not base_url.endswith("/"):
|
||||||
|
base_url = about.__download_url__ + "/"
|
||||||
|
download_url = urljoin(base_url, filename)
|
||||||
|
if not download_url.startswith(about.__download_url__):
|
||||||
|
raise ValueError(f"Download from {filename} rejected. Was it a relative path?")
|
||||||
pip_args = list(user_pip_args) if user_pip_args is not None else []
|
pip_args = list(user_pip_args) if user_pip_args is not None else []
|
||||||
cmd = [sys.executable, "-m", "pip", "install"] + pip_args + [download_url]
|
cmd = [sys.executable, "-m", "pip", "install"] + pip_args + [download_url]
|
||||||
run_command(cmd)
|
run_command(cmd)
|
||||||
|
|
|
@ -220,6 +220,7 @@ class Warnings(metaclass=ErrorsWithCodes):
|
||||||
"key attribute for vectors, configure it through Vectors(attr=) or "
|
"key attribute for vectors, configure it through Vectors(attr=) or "
|
||||||
"'spacy init vectors --attr'")
|
"'spacy init vectors --attr'")
|
||||||
W126 = ("These keys are unsupported: {unsupported}")
|
W126 = ("These keys are unsupported: {unsupported}")
|
||||||
|
W127 = ("Not all `Language.pipe` worker processes completed successfully")
|
||||||
|
|
||||||
|
|
||||||
class Errors(metaclass=ErrorsWithCodes):
|
class Errors(metaclass=ErrorsWithCodes):
|
||||||
|
|
|
@ -1730,6 +1730,9 @@ class Language:
|
||||||
for proc in procs:
|
for proc in procs:
|
||||||
proc.join()
|
proc.join()
|
||||||
|
|
||||||
|
if not all(proc.exitcode == 0 for proc in procs):
|
||||||
|
warnings.warn(Warnings.W127)
|
||||||
|
|
||||||
def _link_components(self) -> None:
|
def _link_components(self) -> None:
|
||||||
"""Register 'listeners' within pipeline components, to allow them to
|
"""Register 'listeners' within pipeline components, to allow them to
|
||||||
effectively share weights.
|
effectively share weights.
|
||||||
|
@ -2350,6 +2353,7 @@ def _apply_pipes(
|
||||||
if isinstance(texts_with_ctx, _WorkDoneSentinel):
|
if isinstance(texts_with_ctx, _WorkDoneSentinel):
|
||||||
sender.close()
|
sender.close()
|
||||||
receiver.close()
|
receiver.close()
|
||||||
|
return
|
||||||
|
|
||||||
docs = (
|
docs = (
|
||||||
ensure_doc(doc_like, context) for doc_like, context in texts_with_ctx
|
ensure_doc(doc_like, context) for doc_like, context in texts_with_ctx
|
||||||
|
@ -2375,6 +2379,7 @@ def _apply_pipes(
|
||||||
# stop processing.
|
# stop processing.
|
||||||
sender.close()
|
sender.close()
|
||||||
receiver.close()
|
receiver.close()
|
||||||
|
return
|
||||||
|
|
||||||
|
|
||||||
class _Sender:
|
class _Sender:
|
||||||
|
|
|
@ -12,7 +12,7 @@ from thinc.api import Config
|
||||||
|
|
||||||
import spacy
|
import spacy
|
||||||
from spacy import about
|
from spacy import about
|
||||||
from spacy.cli import info
|
from spacy.cli import download_module, info
|
||||||
from spacy.cli._util import parse_config_overrides, string_to_list, walk_directory
|
from spacy.cli._util import parse_config_overrides, string_to_list, walk_directory
|
||||||
from spacy.cli.apply import apply
|
from spacy.cli.apply import apply
|
||||||
from spacy.cli.debug_data import (
|
from spacy.cli.debug_data import (
|
||||||
|
@ -1066,3 +1066,15 @@ def test_debug_data_trainable_lemmatizer_not_annotated():
|
||||||
def test_project_api_imports():
|
def test_project_api_imports():
|
||||||
from spacy.cli import project_run
|
from spacy.cli import project_run
|
||||||
from spacy.cli.project.run import project_run # noqa: F401, F811
|
from spacy.cli.project.run import project_run # noqa: F401, F811
|
||||||
|
|
||||||
|
|
||||||
|
def test_download_rejects_relative_urls(monkeypatch):
|
||||||
|
"""Test that we can't tell spacy download to get an arbitrary model by using a
|
||||||
|
relative path in the filename"""
|
||||||
|
|
||||||
|
monkeypatch.setattr(download_module, "run_command", lambda cmd: None)
|
||||||
|
|
||||||
|
# Check that normal download works
|
||||||
|
download_module.download("en_core_web_sm-3.7.1", direct=True)
|
||||||
|
with pytest.raises(SystemExit):
|
||||||
|
download_module.download("../en_core_web_sm-3.7.1", direct=True)
|
||||||
|
|
|
@ -1,5 +1,6 @@
|
||||||
import itertools
|
import itertools
|
||||||
import logging
|
import logging
|
||||||
|
import warnings
|
||||||
from unittest import mock
|
from unittest import mock
|
||||||
|
|
||||||
import pytest
|
import pytest
|
||||||
|
@ -738,6 +739,10 @@ def test_pass_doc_to_pipeline(nlp, n_process):
|
||||||
assert doc.text == texts[0]
|
assert doc.text == texts[0]
|
||||||
assert len(doc.cats) > 0
|
assert len(doc.cats) > 0
|
||||||
if isinstance(get_current_ops(), NumpyOps) or n_process < 2:
|
if isinstance(get_current_ops(), NumpyOps) or n_process < 2:
|
||||||
|
# Catch warnings to ensure that all worker processes exited
|
||||||
|
# succesfully.
|
||||||
|
with warnings.catch_warnings():
|
||||||
|
warnings.simplefilter("error")
|
||||||
docs = nlp.pipe(docs, n_process=n_process)
|
docs = nlp.pipe(docs, n_process=n_process)
|
||||||
assert [doc.text for doc in docs] == texts
|
assert [doc.text for doc in docs] == texts
|
||||||
assert all(len(doc.cats) for doc in docs)
|
assert all(len(doc.cats) for doc in docs)
|
||||||
|
|
|
@ -526,13 +526,17 @@ application's `requirements.txt`. If you're running your own internal PyPi
|
||||||
installation, you can upload the pipeline packages there. pip's
|
installation, you can upload the pipeline packages there. pip's
|
||||||
[requirements file format](https://pip.pypa.io/en/latest/reference/requirements-file-format/)
|
[requirements file format](https://pip.pypa.io/en/latest/reference/requirements-file-format/)
|
||||||
supports both package names to download via a PyPi server, as well as
|
supports both package names to download via a PyPi server, as well as
|
||||||
[direct URLs](#pipeline-urls).
|
[direct URLs](#pipeline-urls). For instance, you can specify the
|
||||||
|
`en_core_web_sm` model for spaCy 3.7.x as follows:
|
||||||
|
|
||||||
```text {title="requirements.txt"}
|
```text {title="requirements.txt"}
|
||||||
spacy>=3.0.0,<4.0.0
|
spacy>=3.0.0,<4.0.0
|
||||||
en_core_web_sm @ https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.4.0/en_core_web_sm-3.4.0-py3-none-any.whl
|
en_core_web_sm @ https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.7.1/en_core_web_sm-3.7.1-py3-none-any.whl
|
||||||
```
|
```
|
||||||
|
|
||||||
|
See the [list of models](https://spacy.io/models) for model download links for
|
||||||
|
the current spaCy version.
|
||||||
|
|
||||||
All pipeline packages are versioned and specify their spaCy dependency. This
|
All pipeline packages are versioned and specify their spaCy dependency. This
|
||||||
ensures cross-compatibility and lets you specify exact version requirements for
|
ensures cross-compatibility and lets you specify exact version requirements for
|
||||||
each pipeline. If you've [trained](/usage/training) your own pipeline, you can
|
each pipeline. If you've [trained](/usage/training) your own pipeline, you can
|
||||||
|
|
Loading…
Reference in New Issue
Block a user