Merge pull request #5516 from explosion/feature/improve-model-version-deps

requirements.txt

@@ -13,8 +13,11 @@ numpy>=1.15.0
 requests>=2.13.0,<3.0.0
 plac>=0.9.6,<1.2.0
 tqdm>=4.38.0,<5.0.0
-importlib_metadata>=0.20; python_version < "3.8"
 pydantic>=1.3.0,<2.0.0
+# Official Python utilities
+setuptools
+packaging
+importlib_metadata>=0.20; python_version < "3.8"
 # Development dependencies
 cython>=0.25
 pytest>=4.6.5

setup.cfg

@@ -50,11 +50,13 @@ install_requires =
     ml_datasets>=0.1.1
     # Third-party dependencies
     tqdm>=4.38.0,<5.0.0
-    setuptools
     numpy>=1.15.0
     plac>=0.9.6,<1.2.0
     requests>=2.13.0,<3.0.0
     pydantic>=1.3.0,<2.0.0
+    # Official Python utilities
+    setuptools
+    packaging
     importlib_metadata>=0.20; python_version < "3.8"
 
 [options.extras_require]

spacy/cli/download.py

@@ -5,7 +5,7 @@ import sys
 from wasabi import msg
 
 from .. import about
-from ..util import is_package
+from ..util import is_package, get_base_version
 
 
 def download(
@@ -63,8 +63,7 @@ def get_json(url, desc):
 
 
 def get_compatibility():
-    version = about.__version__
-    version = version.rsplit(".dev", 1)[0]
+    version = get_base_version(about.__version__)
     comp_table = get_json(about.__compatibility__, "compatibility table")
     comp = comp_table["spacy"]
     if version not in comp:
@@ -73,7 +72,7 @@ def get_compatibility():
 
 
 def get_version(model, comp):
-    model = model.rsplit(".dev", 1)[0]
+    model = get_base_version(model)
     if model not in comp:
         msg.fail(
             f"No compatible model found for '{model}' (spaCy v{about.__version__})",

spacy/cli/package.py

@@ -90,7 +90,7 @@ def generate_meta(model_path, existing_meta, msg):
         ("license", "License", meta.get("license", "MIT")),
     ]
     nlp = util.load_model_from_path(Path(model_path))
-    meta["spacy_version"] = about.__version__
+    meta["spacy_version"] = util.get_model_version_range(about.__version__)
     meta["pipeline"] = nlp.pipe_names
     meta["vectors"] = {
         "width": nlp.vocab.vectors_length,

spacy/cli/train.py

@@ -467,7 +467,6 @@ def train(
                     # Update model meta.json
                     meta["lang"] = nlp.lang
                     meta["pipeline"] = nlp.pipe_names
-                    meta["spacy_version"] = about.__version__
                     if beam_width == 1:
                         meta["speed"] = {
                             "nwords": nwords,

spacy/cli/validate.py

@@ -4,8 +4,8 @@ import requests
 from wasabi import msg
 
 from .. import about
-from ..util import get_package_version, get_installed_models, split_version
-from ..util import get_package_path, get_model_meta, is_compatible_model
+from ..util import get_package_version, get_installed_models, get_base_version
+from ..util import get_package_path, get_model_meta, is_compatible_version
 
 
 def validate():
@@ -14,7 +14,7 @@ def validate():
     with the installed models. Should be run after `pip install -U spacy`.
     """
     model_pkgs, compat = get_model_pkgs()
-    spacy_version = about.__version__.rsplit(".dev", 1)[0]
+    spacy_version = get_base_version(about.__version__)
     current_compat = compat.get(spacy_version, {})
     if not current_compat:
         msg.warn(f"No compatible models found for v{spacy_version} of spaCy")
@@ -78,13 +78,12 @@ def get_model_pkgs():
         version = get_package_version(pkg_name)
         if package in compat:
             is_compat = version in compat[package]
-            v_maj, v_min = split_version(about.__version__)
-            spacy_version = f"{v_maj}.{v_min}"
+            spacy_version = about.__version__
         else:
             model_path = get_package_path(package)
             model_meta = get_model_meta(model_path)
-            is_compat = is_compatible_model(model_meta)
             spacy_version = model_meta.get("spacy_version", "n/a")
+            is_compat = is_compatible_version(about.__version__, spacy_version)
         pkgs[pkg_name] = {
             "name": package,
             "version": version,

spacy/errors.py

@@ -104,6 +104,12 @@ class Warnings(object):
             "string \"Field1=Value1,Value2|Field2=Value3\".")
 
     # TODO: fix numbering after merging develop into master
+    W095 = ("Model '{model}' ({model_version}) requires spaCy {version} and is "
+            "incompatible with the current version ({current}). This may lead "
+            "to unexpected results or runtime errors. To resolve this, "
+            "download a newer compatible model or retrain your custom model "
+            "with the current spaCy version. For more details and available "
+            "updates, run: python -m spacy validate")
     W096 = ("The method 'disable_pipes' has become deprecated - use 'select_pipes' "
             "instead.")
     W097 = ("No Model config was provided to create the '{name}' component, "

spacy/language.py

@@ -191,13 +191,14 @@ class Language(object):
 
     @property
     def meta(self):
+        spacy_version = util.get_model_version_range(about.__version__)
         if self.vocab.lang:
             self._meta.setdefault("lang", self.vocab.lang)
         else:
             self._meta.setdefault("lang", self.lang)
         self._meta.setdefault("name", "model")
         self._meta.setdefault("version", "0.0.0")
-        self._meta.setdefault("spacy_version", about.__version__)
+        self._meta.setdefault("spacy_version", spacy_version)
         self._meta.setdefault("description", "")
         self._meta.setdefault("author", "")
         self._meta.setdefault("email", "")

spacy/tests/test_misc.py

@@ -94,8 +94,18 @@ def test_ascii_filenames():
 
 
 @pytest.mark.parametrize(
-    "version,compatible",
-    [(spacy_version, True), ("2.0.0", False), (">=1.2.3,<4.5.6", False)],
+    "version,constraint,compatible",
+    [
+        (spacy_version, spacy_version, True),
+        (spacy_version, f">={spacy_version}", True),
+        ("3.0.0", "2.0.0", False),
+        ("3.2.1", ">=2.0.0", True),
+        ("2.2.10a1", ">=1.0.0,<2.1.1", False),
+        ("3.0.0.dev3", ">=1.2.3,<4.5.6", True),
+        ("n/a", ">=1.2.3,<4.5.6", None),
+        ("1.2.3", "n/a", None),
+        ("n/a", "n/a", None),
+    ],
 )
-def test_is_compatible_model(version, compatible):
-    assert util.is_compatible_model({"spacy_version": version}) is compatible
+def test_is_compatible_version(version, constraint, compatible):
+    assert util.is_compatible_version(version, constraint) is compatible

spacy/util.py

@@ -14,6 +14,8 @@ import srsly
 import catalogue
 import sys
 import warnings
+from packaging.specifiers import SpecifierSet, InvalidSpecifier
+from packaging.version import Version, InvalidVersion
 
 
 try:
@@ -236,33 +238,46 @@ def get_package_version(name):
         return None
 
 
-def split_version(version):
-    """RETURNS (tuple): Two integers, the major and minor spaCy version."""
-    pieces = version.split(".", 3)
-    return int(pieces[0]), int(pieces[1])
+def is_compatible_version(version, constraint, prereleases=True):
+    """Check if a version (e.g. "2.0.0") is compatible given a version
+    constraint (e.g. ">=1.9.0,<2.2.1"). If the constraint is a specific version,
+    it's interpreted as =={version}.
 
-
-def is_compatible_model(meta):
-    """Check if a model is compatible with the current version of spaCy, based
-    on its meta.json. We compare the version of spaCy the model was created with
-    with the current version. If the minor version is different, it's considered
-    incompatible.
-
-    meta (dict): The model's meta.
-    RETURNS (bool / None): Whether the model is compatible with the current
-        spaCy or None if we don't have enough info.
+    version (str): The version to check.
+    constraint (str): The constraint string.
+    prereleases (bool): Whether to allow prereleases. If set to False,
+        prerelease versions will be considered incompatible.
+    RETURNS (bool / None): Whether the version is compatible, or None if the
+        version or constraint are invalid.
     """
-    cur_v = about.__version__
-    pkg_v = meta.get("spacy_version")
-    if not pkg_v or not isinstance(pkg_v, str):
+    # Handle cases where exact version is provided as constraint
+    if constraint[0].isdigit():
+        constraint = f"=={constraint}"
+    try:
+        spec = SpecifierSet(constraint)
+        version = Version(version)
+    except (InvalidSpecifier, InvalidVersion):
         return None
-    # Handle spacy_version values like >=x,<y, just in case
-    pkg_v = re.sub(r"[^0-9.]", "", pkg_v.split(",")[0])
-    cur_major, cur_minor = split_version(cur_v)
-    pkg_major, pkg_minor = split_version(pkg_v)
-    if cur_major != pkg_major or cur_minor != pkg_minor:
-        return False
-    return True
+    spec.prereleases = prereleases
+    return version in spec
+
+
+def get_model_version_range(spacy_version):
+    """Generate a version range like >=1.2.3,<1.3.0 based on a given spaCy
+    version. Models are always compatible across patch versions but not
+    across minor or major versions.
+    """
+    release = Version(spacy_version).release
+    return f">={spacy_version},<{release[0]}.{release[1] + 1}.0"
+
+
+def get_base_version(version):
+    """Generate the base version without any prerelease identifiers.
+
+    version (str): The version, e.g. "3.0.0.dev1".
+    RETURNS (str): The base version, e.g. "3.0.0".
+    """
+    return Version(version).base_version
 
 
 def load_config(path, create_objects=False):
@@ -315,6 +330,16 @@ def get_model_meta(path):
     for setting in ["lang", "name", "version"]:
         if setting not in meta or not meta[setting]:
             raise ValueError(Errors.E054.format(setting=setting))
+    if "spacy_version" in meta:
+        if not is_compatible_version(about.__version__, meta["spacy_version"]):
+            warnings.warn(
+                Warnings.W095.format(
+                    model=f"{meta['lang']}_{meta['name']}",
+                    model_version=meta["version"],
+                    version=meta["spacy_version"],
+                    current=about.__version__,
+                )
+            )
     return meta