Merge pull request #11 from letsbuilda/gh10

Retrieve package metadata from the JSON API

docs/source/changelog.rst

@@ -4,6 +4,9 @@
 Changelog
 =========
 
+- :release:`3.0.0 <21th April 2023>`
+- :feature:`10` Retrieve package metadata from the JSON API
+
 - :release:`2.0.0 <14th April 2023>`
 - :feature:`4` Use single method for all RSS feeds
 

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "letsbuilda-pypi"
-version = "2.0.0"
+version = "3.0.0"
 description = "A wrapper for PyPI's API and RSS feed"
 authors = [
     { name = "Bradley Reynolds", email = "[email protected]" },
@@ -11,6 +11,7 @@ requires-python = ">=3.10"
 dependencies = [
     "aiohttp",
     "xmltodict",
+    "pendulum",
 ]
 
 [project.urls]

src/letsbuilda/pypi/__init__.py

@@ -2,63 +2,11 @@
 A wrapper for PyPI's API and RSS feed
 """
 
-from dataclasses import dataclass
-from datetime import datetime
-from typing import Final
-
-import xmltodict
-from aiohttp import ClientSession
+from .client import PyPIServices
+from .models import PackageMetadata, RSSPackageMetadata
 
 __all__: list[str] = [
     "PyPIServices",
     "PackageMetadata",
+    "RSSPackageMetadata",
 ]
-
-
-def _parse_publication_date(publication_date: str) -> datetime:
-    return datetime.strptime(publication_date, "%a, %d %b %Y %H:%M:%S %Z")
-
-
-@dataclass(frozen=True, slots=True)
-class PackageMetadata:
-    """Package metadata"""
-
-    title: str
-    package_link: str
-    guid: str
-    description: str | None
-    author: str | None
-    publication_date: datetime
-
-    @classmethod
-    def build_from(cls, data: dict[str, str]) -> "PackageMetadata":
-        """Build an instance from raw data"""
-        publication_date: str | None = data.get("pubDate")
-        if publication_date is not None:
-            publication_date: datetime = _parse_publication_date(publication_date)
-
-        return cls(
-            title=data.get("title").split()[0],
-            package_link=data.get("link"),
-            guid=data.get("guid"),
-            description=data.get("description"),
-            author=data.get("author"),
-            publication_date=publication_date,
-        )
-
-
-class PyPIServices:
-    """A class for interacting with PyPI"""
-
-    NEWEST_PACKAGES_FEED_URL: Final[str] = "https://pypi.org/rss/packages.xml"
-    PACKAGE_UPDATES_FEED_URL: Final[str] = "https://pypi.org/rss/updates.xml"
-
-    def __init__(self, http_session: ClientSession) -> None:
-        self.http_session = http_session
-
-    async def get_rss_feed(self, feed_url: str) -> list[PackageMetadata]:
-        """Get the new packages RSS feed"""
-        async with self.http_session.get(feed_url) as response:
-            response_text = await response.text()
-            rss_data = xmltodict.parse(response_text)["rss"]["channel"]["item"]
-            return [PackageMetadata.build_from(package_data) for package_data in rss_data]

src/letsbuilda/pypi/client.py

@@ -0,0 +1,35 @@
+"""Service wrapper"""
+
+from typing import Final
+
+import xmltodict
+from aiohttp import ClientSession
+
+from .models import PackageMetadata, RSSPackageMetadata
+
+
+class PyPIServices:
+    """A class for interacting with PyPI"""
+
+    NEWEST_PACKAGES_FEED_URL: Final[str] = "https://pypi.org/rss/packages.xml"
+    PACKAGE_UPDATES_FEED_URL: Final[str] = "https://pypi.org/rss/updates.xml"
+
+    def __init__(self, http_session: ClientSession) -> None:
+        self.http_session = http_session
+
+    async def get_rss_feed(self, feed_url: str) -> list[RSSPackageMetadata]:
+        """Get the new packages RSS feed"""
+        async with self.http_session.get(feed_url) as response:
+            response_text = await response.text()
+            rss_data = xmltodict.parse(response_text)["rss"]["channel"]["item"]
+            return [RSSPackageMetadata.build_from(package_data) for package_data in rss_data]
+
+    async def get_package_metadata(self, package_name: str) -> PackageMetadata:
+        """Get the new packages RSS feed"""
+        async with self.http_session.get(f"https://pypi.org/pypi/{package_name}/json") as response:
+            return PackageMetadata.from_dict(await response.json())
+
+    async def get_package_metadata_for_release(self, package_name: str, package_version: str) -> PackageMetadata:
+        """Get the new packages RSS feed"""
+        async with self.http_session.get(f"https://pypi.org/pypi/{package_name}/{package_version}/json") as response:
+            return PackageMetadata.from_dict(await response.json())

src/letsbuilda/pypi/models.py

@@ -0,0 +1,168 @@
+"""Models"""
+
+from dataclasses import dataclass
+from datetime import datetime
+
+import pendulum
+from pendulum import DateTime
+
+
+def _parse_publication_date(publication_date: str) -> datetime:
+    return datetime.strptime(publication_date, "%a, %d %b %Y %H:%M:%S %Z")
+
+
+@dataclass(frozen=True, slots=True)
+class Vulnerability:
+    """Security vulnerability"""
+
+    id: str
+    aliases: list[str]
+    link: str
+    source: str
+    withdrawn: DateTime | None
+    summary: str
+    details: str
+    fixed_in: list[str]
+
+    @classmethod
+    def from_dict(cls, data: dict) -> "Vulnerability":
+        if data["withdrawn"] is not None:
+            data["withdrawn"]: DateTime = pendulum.parse(data["withdrawn"])
+
+        return cls(**data)
+
+
+@dataclass(frozen=True, slots=True)
+class Downloads:
+    """Release download counts"""
+
+    last_day: int
+    last_month: int
+    last_week: int
+
+    @classmethod
+    def from_dict(cls, data: dict) -> "Downloads":
+        return cls(**data)
+
+
+@dataclass(frozen=True, slots=True)
+class Digests:
+    """URL file digests"""
+
+    blake2_b_256: str
+    md5: str
+    sha256: str
+
+    @classmethod
+    def from_dict(cls, data: dict) -> "Digests":
+        return cls(**data)
+
+
+@dataclass(frozen=True, slots=True)
+class URL:
+    """Package release URL"""
+
+    comment_text: str
+    digests: Digests
+    downloads: int
+    filename: str
+    has_sig: bool
+    md5_digest: str
+    packagetype: str
+    python_version: str
+    requires_python: str | None
+    size: int
+    upload_time: DateTime
+    upload_time_iso_8601: DateTime
+    url: str
+    yanked: bool
+    yanked_reason: None
+
+    @classmethod
+    def from_dict(cls, data: dict) -> "URL":
+        data["upload_time"]: DateTime = pendulum.parse(data["upload_time"])
+        data["upload_time_iso_8601"]: DateTime = pendulum.parse(data["upload_time_iso_8601"])
+        return cls(**data)
+
+
+@dataclass(frozen=True, slots=True)
+class Info:
+    """Package metadata internal info block"""
+
+    author: str
+    author_email: str
+    bugtrack_url: None
+    classifiers: list[str]
+    description: str
+    description_content_type: str
+    docs_url: None
+    download_url: str
+    downloads: Downloads
+    home_page: str
+    keywords: str
+    license: str
+    maintainer: str
+    maintainer_email: str
+    name: str
+    package_url: str
+    platform: str | None
+    project_url: str
+    project_urls: dict[str, str]
+    release_url: str
+    requires_dist: list[str]
+    requires_python: str
+    summary: str
+    version: str
+    yanked: bool
+    yanked_reason: str | None
+
+    @classmethod
+    def from_dict(cls, data: dict) -> "Info":
+        return cls(**data)
+
+
+@dataclass(frozen=True, slots=True)
+class PackageMetadata:
+    """Package metadata"""
+
+    info: Info
+    last_serial: int
+    urls: list[URL]
+    vulnerabilities: list["Vulnerability"]
+
+    @classmethod
+    def from_dict(cls, data: dict) -> "PackageMetadata":
+        return cls(
+            info=Info.from_dict(data["info"]),
+            last_serial=data["last_serial"],
+            urls=[URL.from_dict(url_data) for url_data in data["urls"]],
+            vulnerabilities=[Vulnerability.from_dict(vuln_data) for vuln_data in data["vulnerabilities"]],
+        )
+
+
+@dataclass(frozen=True, slots=True)
+class RSSPackageMetadata:
+    """RSS Package metadata"""
+
+    title: str
+    package_link: str
+    guid: str
+    description: str | None
+    author: str | None
+    publication_date: datetime | None
+
+    @classmethod
+    def build_from(cls, data: dict[str, str]) -> "RSSPackageMetadata":
+        """Build an instance from raw data"""
+        publication_date: str | None = data.get("pubDate")
+        if publication_date is not None:
+            publication_date: datetime = _parse_publication_date(publication_date)
+
+        return cls(
+            title=data.get("title").split()[0],
+            package_link=data.get("link"),
+            guid=data.get("guid"),
+            description=data.get("description"),
+            author=data.get("author"),
+            publication_date=publication_date,
+        )

tests/test_new_packages_feed.py

@@ -2,7 +2,7 @@
 
 from datetime import datetime
 
-from letsbuilda.pypi import PackageMetadata
+from letsbuilda.pypi import RSSPackageMetadata
 
 
 def test_parse_publication_date() -> None:
@@ -15,7 +15,7 @@ def test_parse_publication_date() -> None:
         "author": "[email protected]",
         "pubDate": "Wed, 29 Mar 2023 21:30:05 GMT",
     }
-    parsed_data = PackageMetadata.build_from(data)
+    parsed_data = RSSPackageMetadata.build_from(data)
     assert parsed_data.publication_date == datetime(2023, 3, 29, 21, 30, 5)
 
 
@@ -28,7 +28,7 @@ def test_author_missing() -> None:
         "description": "a test package",
         "pubDate": "Wed, 29 Mar 2023 21:30:05 GMT",
     }
-    parsed_data = PackageMetadata.build_from(data)
+    parsed_data = RSSPackageMetadata.build_from(data)
     assert parsed_data.author is None
 
 
@@ -41,5 +41,5 @@ def test_description_missing() -> None:
         "author": "[email protected]",
         "pubDate": "Wed, 29 Mar 2023 21:30:05 GMT",
     }
-    parsed_data = PackageMetadata.build_from(data)
+    parsed_data = RSSPackageMetadata.build_from(data)
     assert parsed_data.description is None