Merge pull request #114 from jepler/iersdata-json

store iersdata in json

Author: jepler

.github/workflows/cron.yml

@@ -41,4 +41,4 @@ jobs:
         git config user.name "${GITHUB_ACTOR} (github actions cron)"
         git config user.email "${GITHUB_ACTOR}@users.noreply.github.com"
         git remote set-url --push origin https://${GITHUB_ACTOR}:${{ secrets.GITHUB_TOKEN }}@github.com/${{ github.repository }}
-        if git commit -m"update iersdata" src/wwvb/iersdata_dist.py; then git push origin HEAD:main; fi
+        if git commit -m"update iersdata" src/wwvb/iersdata.json; then git push origin HEAD:main; fi

pyproject.toml

@@ -10,12 +10,11 @@ requires = [
 build-backend = "setuptools.build_meta"
 [tool.setuptool]
 package_dir = {"" = "src"}
+include-package-data = true
 
 [tool.setuptools.dynamic]
 readme = {file = ["README.md"], content-type="text/markdown"}
 dependencies = {file = "requirements.txt"}
-[tool.setuptools.package-data]
-"pkgname" = ["py.typed"]
 [tool.setuptools_scm]
 write_to = "src/wwvb/__version__.py"
 [tool.ruff.lint]
@@ -40,9 +39,8 @@ classifiers = [
 ]
 requires-python = ">=3.9"
 [project.urls]
-homepage = "https://github.com/jepler/wwvbpy"
-repository = "https://github.com/jepler/wwvbpy"
-
+Source = "https://github.com/jepler/wwvbpy"
+Documentation = "https://github.com/jepler/wwvbpy"
 [project.scripts]
 wwvbgen = "wwvb.gen:main"
 wwvbdecode = "wwvb.decode:main"

requirements-dev.txt

@@ -18,8 +18,8 @@ sphinx
 sphinx-autodoc-typehints
 sphinx-rtd-theme
 twine; implementation_name=="cpython"
-types-beautifulsoup4
-types-python-dateutil
-types-requests
+types-beautifulsoup4; implementation_name=="cpython"
+types-python-dateutil; implementation_name=="cpython"
+types-requests; implementation_name=="cpython"
 tzdata
 wheel

src/wwvb/iersdata.json

@@ -0,0 +1 @@
+{"START": "1972-01-01", "OFFSETS_GZ": "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"}

src/wwvb/iersdata.json.license

@@ -0,0 +1,2 @@
+SPDX-FileCopyrightText: Public domain
+SPDX-License-Identifier: CC0-1.0

src/wwvb/iersdata.py

@@ -5,23 +5,33 @@
 #
 # SPDX-License-Identifier: GPL-3.0-only
 
+import binascii
 import datetime
-import pathlib
+import gzip
+import importlib.resources
+import json
 
 import platformdirs
 
 __all__ = ["DUT1_DATA_START", "DUT1_OFFSETS", "start", "span", "end"]
-from .iersdata_dist import DUT1_DATA_START, DUT1_OFFSETS
 
-for location in [
-    platformdirs.user_data_dir("wwvbpy", "unpythonic.net"),
-    platformdirs.site_data_dir("wwvbpy", "unpythonic.net"),
+content: dict[str, str] = {"START": "1970-01-01", "OFFSETS_GZ": "H4sIAFNx1mYC/wMAAAAAAAAAAAA="}
+
+path = importlib.resources.files("wwvb") / "iersdata.json"
+content = json.loads(path.read_text(encoding="utf-8"))
+
+for location in [  # pragma no cover
+    platformdirs.user_data_path("wwvbpy", "unpythonic.net"),
+    platformdirs.site_data_path("wwvbpy", "unpythonic.net"),
 ]:
-    path = pathlib.Path(location) / "wwvbpy_iersdata.py"
-    if path.exists():  # pragma no cover
-        exec(path.read_text(encoding="utf-8"), globals(), globals())
+    path = location / "iersdata.json"
+    if path.exists():
+        content = json.loads(path.read_text(encoding="utf-8"))
         break
 
+DUT1_DATA_START = datetime.date.fromisoformat(content["START"])
+DUT1_OFFSETS = gzip.decompress(binascii.a2b_base64(content["OFFSETS_GZ"])).decode("ascii")
+
 start = datetime.datetime.combine(DUT1_DATA_START, datetime.time(), tzinfo=datetime.timezone.utc)
 span = datetime.timedelta(days=len(DUT1_OFFSETS))
 end = start + span

src/wwvb/iersdata_dist.py

@@ -8,10 +8,12 @@
 
 from __future__ import annotations
 
+import binascii
 import csv
 import datetime
+import gzip
 import io
-import itertools
+import json
 import pathlib
 from typing import Callable
 
@@ -20,15 +22,7 @@
 import platformdirs
 import requests
 
-DIST_PATH = pathlib.Path(__file__).parent / "iersdata_dist.py"
-
-OLD_TABLE_START: datetime.date | None = None
-OLD_TABLE_END: datetime.date | None = None
-if DIST_PATH.exists():
-    import wwvb.iersdata_dist
-
-    OLD_TABLE_START = wwvb.iersdata_dist.DUT1_DATA_START
-    OLD_TABLE_END = OLD_TABLE_START + datetime.timedelta(days=len(wwvb.iersdata_dist.DUT1_OFFSETS) - 1)
+DIST_PATH = pathlib.Path(__file__).parent / "iersdata.json"
 
 IERS_URL = "https://datacenter.iers.org/data/csv/finals2000A.all.csv"
 IERS_PATH = pathlib.Path("finals2000A.all.csv")
@@ -47,7 +41,7 @@ def _get_text(url: str) -> str:
         return pathlib.Path(url).read_text(encoding="utf-8")
 
 
-def update_iersdata(  # noqa: PLR0915, PLR0912
+def update_iersdata(  # noqa: PLR0915
     target_path: pathlib.Path,
 ) -> None:
     """Update iersdata.py"""
@@ -124,57 +118,25 @@ def patch(patch_start: datetime.date, patch_end: datetime.date, val: int) -> Non
     assert wwvb_start is not None
     patch(wwvb_start, wwvb_data_stamp + datetime.timedelta(days=1), wwvb_dut1)
 
-    with target_path.open("w", encoding="utf-8") as output:
-
-        def code(*args: str) -> None:
-            """Print to the output file"""
-            print(*args, file=output)
-
-        code("# -*- python3 -*-")
-        code("# fmt: off")
-        code('"""File generated from public data - not subject to copyright"""')
-        code("# SPDX" + "-FileCopyrightText: Public domain")
-        code("# SPDX" + "-License-Identifier: CC0-1.0")
-        code("# isort: skip_file")
-        code("import datetime")
-
-        code("__all__ = ['DUT1_DATA_START', 'DUT1_OFFSETS']")
-        code(f"DUT1_DATA_START = {table_start!r}")
-        c = sorted(chr(ord("a") + ch + 10) for ch in set(offsets))
-        code(f"{','.join(c)} = tuple({''.join(c)!r})")
-        code(f"DUT1_OFFSETS = str( # {table_start.year:04d}{table_start.month:02d}{table_start.day:02d}")
-        line = ""
-        j = 0
-
-        for val, it in itertools.groupby(offsets):
-            part = ""
-            ch = chr(ord("a") + val + 10)
-            sz = len(list(it))
-            if j:
-                part = part + "+"
-            part = part + ch if sz < 2 else part + f"{ch}*{sz}"
-            j += sz
-            if len(line + part) > 60:
-                d = table_start + datetime.timedelta(j - 1)
-                code(f"    {line:<60s} # {d.year:04d}{d.month:02d}{d.day:02d}")
-                line = part
-            else:
-                line = line + part
-        d = table_start + datetime.timedelta(j - 1)
-        code(f"    {line:<60s} # {d.year:04d}{d.month:02d}{d.day:02d}")
-        code(")")
     table_end = table_start + datetime.timedelta(len(offsets) - 1)
-    if OLD_TABLE_START:
-        print(f"old iersdata covered {OLD_TABLE_START} .. {OLD_TABLE_END}")
+    base = ord("a") + 10
+    offsets_bin = bytes(base + ch for ch in offsets)
+
+    target_path.write_text(
+        json.dumps(
+            {
+                "START": table_start.isoformat(),
+                "OFFSETS_GZ": binascii.b2a_base64(gzip.compress(offsets_bin)).decode("ascii").strip(),
+            },
+        ),
+    )
+
     print(f"iersdata covers {table_start} .. {table_end}")
 
 
 def iersdata_path(callback: Callable[[str, str], pathlib.Path]) -> pathlib.Path:
     """Find out the path for this directory"""
-    print("iersdata_path", callback)
-    r = callback("wwvbpy", "unpythonic.net") / "wwvb_iersdata.py"
-    print(f"iersdata_path {r=!r}")
-    return r
+    return callback("wwvbpy", "unpythonic.net") / "iersdata.json"
 
 
 @click.command()