Merge pull request #21 from jepler/document-data-sources

Update & document data sources

.github/workflows/test.yml

@@ -67,7 +67,7 @@ jobs:
       run: make mypy
 
     - name: Test
-      run: python -mcoverage run --branch -m unittest testleapseconddata.py && python -mcoverage report --fail-under=100 && python -mcoverage xml
+      run: python -X tracemalloc=3 -mcoverage run --branch -m unittest testleapseconddata.py && python -mcoverage report --fail-under=100 && python -mcoverage xml
 
   pre-commit:
     runs-on: ubuntu-latest

docs/conf.py

@@ -51,7 +51,7 @@ templates_path = ['_templates']
 # List of patterns, relative to source directory, that match files and
 # directories to ignore when looking for source files.
 # This pattern also affects html_static_path and html_extra_path.
-exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store']
+exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store', '_env']
 
 
 # -- Options for HTML output -------------------------------------------------

leapseconddata/__init__.py

@@ -30,7 +30,7 @@ import pathlib
 import re
 import urllib.request
 from dataclasses import dataclass, field
-from typing import BinaryIO
+from typing import BinaryIO, ClassVar
 
 tai = datetime.timezone(datetime.timedelta(0), "TAI")
 
@@ -76,6 +76,49 @@ class LeapSecondData:
     :param Optional[datetime.datetime] updated: The last update time of the data
     """
 
+    standard_file_sources: ClassVar[list[str]] = [
+        "file:///usr/share/zoneinfo/leap-seconds.list",  # Debian Linux
+        "file:///var/db/ntpd.leap-seconds.list",  # FreeBSD
+    ]
+    """When using `LeapSecondData.from_standard_source`, these local sources are checked first.
+
+    Locations for Debian Linux & FreeBSD are supported."""
+
+    standard_network_sources: ClassVar[list[str]] = [
+        "https://hpiers.obspm.fr/iers/bul/bulc/ntp/leap-seconds.list",
+        "https://data.iana.org/time-zones/tzdb/leap-seconds.list",
+        "https://raw.githubusercontent.com/eggert/tz/main/leap-seconds.list",
+        "ftp://ftp.boulder.nist.gov/pub/time/leap-seconds.list",
+        "https://www.meinberg.de/download/ntp/leap-seconds.list",
+    ]
+    """When using `LeapSecondData.from_standard_source`, these network sources are checked second.
+
+    Remote sources are checked in the following order until a suitable file is found:
+
+     * The `International Earth Rotation Service (IERS)
+       <https://www.iers.org/IERS/EN/Home/home_node.html>`_ is the international
+       body charged with various duties including scheduling leap seconds.
+     * The `Internet Assigned Numbers Authority (IANA)
+       <https://www.iana.org/>`_ publishes the IANA timezone database, used by
+       many major operating sytsems for handling the world's time zones. As part
+       of this activity they publish a version of the leap second list.
+     * `eggert/tz <https://github.com/eggert/tz>`_ is the canonical github home
+       of the IANA timezone database, and updated versions of the leap second
+       list can appear here before they are part of an official IANA timezone
+       database release.
+     * `The National Institute of Standards and Technology (NIST)'s Time
+       Realization and Distribution Group
+       <https://www.nist.gov/pml/time-and-frequency-division/time-distribution/internet-time-service-its>`_
+       is a US federal organization that publishes a version of the leap second
+       database.
+     * `Meinberg Funkuhren GmbH & Co. KG
+       <https://www.meinbergglobal.com/english/company/>`_ is a Germany-based
+       company that published a `helpful article in its knowledge base
+       <https://kb.meinbergglobal.com/kb/time_sync/ntp/configuration/ntp_leap_second_file>`_
+       including URLs of sites that disseminate the leap second list. They state
+       that the version they distribute is frequently more up to date than other
+       sources, including IANA, NIST, and tzdb."""
+
     leap_seconds: list[LeapSecondInfo]
     """All known and scheduled leap seconds"""
 
@@ -204,12 +247,7 @@ class LeapSecondData:
         leap-second.list data valid for the given timestamp, or the current
         time (if unspecified)
         """
-        for location in [  # pragma no branch
-            "file:///usr/share/zoneinfo/leap-seconds.list",  # Debian Linux
-            "file:///var/db/ntpd.leap-seconds.list",  # FreeBSD
-            "https://raw.githubusercontent.com/eggert/tz/main/leap-seconds.list",
-            "https://www.meinberg.de/download/ntp/leap-seconds.list",
-        ]:
+        for location in cls.standard_file_sources + cls.standard_network_sources:
             logging.debug("Trying leap second data from %s", location)
             try:
                 candidate = cls.from_url(location, check_hash=check_hash)
@@ -244,14 +282,13 @@ class LeapSecondData:
     @classmethod
     def from_url(
         cls,
-        url: str = "https://raw.githubusercontent.com/eggert/tz/main/leap-seconds.list",
+        url: str,
         *,
         check_hash: bool = True,
     ) -> LeapSecondData | None:
         """Retrieve the leap second list from a local file
 
-        :param filename: URL to read leap second data from.  The
-            default is maintained by the tzdata authors
+        :param filename: URL to read leap second data from
         :param check_hash: Whether to check the embedded hash
         """
         try:

leapseconddata/__main__.py

@@ -13,7 +13,7 @@ from dataclasses import dataclass
 
 import click
 
-from . import LeapSecondData, tai
+from . import InvalidHashError, LeapSecondData, tai
 
 utc = datetime.timezone.utc
 
@@ -165,5 +165,29 @@ def table(ctx: click.Context, *, start: datetime.datetime, end: datetime.datetim
         print(f"{leap_second.start:%Y-%m-%d}: {leap_second.tai_offset.total_seconds():.0f}")
 
 
+@cli.command
+def sources() -> None:
+    """Print information about leap-second.list data sources"""
+    first = True
+    for location in LeapSecondData.standard_file_sources + LeapSecondData.standard_network_sources:
+        if not first:
+            print()
+        first = False
+        try:
+            leap_second_data = LeapSecondData.from_url(location, check_hash=True)
+        except InvalidHashError:  # pragma no coverage
+            print(f"{location}: Invalid hash")
+            leap_second_data = LeapSecondData.from_url(location, check_hash=False)
+        except Exception as e:  # pragma no coverage # noqa: BLE001
+            print(f"{location}: {e}")
+            leap_second_data = None
+
+        if leap_second_data is not None:
+            print(f"{location}: Last updated {leap_second_data.last_updated}")
+            print(f"{location}: Valid until {leap_second_data.valid_until}")
+        else:
+            print(f"{location}: Could not be read")
+
+
 if __name__ == "__main__":  # pragma no cover
     cli()

pyproject.toml

@@ -19,7 +19,7 @@ write_to = "leapseconddata/__version__.py"
 line-length=120
 [tool.ruff.lint]
 select = ["E", "F", "D", "I", "N", "UP", "YTT", "BLE", "B", "FBT", "A", "COM", "C4", "DTZ", "FA", "ISC", "ICN", "PIE", "PYI", "Q", "RET", "SIM", "TID", "TCH", "ARG", "PTH", "C", "R", "W", "FLY", "RUF", "PL"]
-ignore = ["D203", "D213", "D400", "D415", "ISC001"]
+ignore = ["D203", "D213", "D400", "D415", "ISC001", "COM812"]
 [project]
 name = "leapseconddata"
 authors = [{name = "Jeff Epler", email = "jepler@gmail.com"}]

testleapseconddata.py

@@ -41,6 +41,7 @@ class LeapSecondDataTest(unittest.TestCase):
         self.run_main("next-leapsecond", "2100-2-2")
         self.run_main("previous-leapsecond", "2009-2-2")
         self.run_main("previous-leapsecond", "1960-2-2")
+        self.run_main("sources")
 
     def test_corrupt(self) -> None:
         self.assertRaises(