codedx 0.1.0__tar.gz → 0.2.0__tar.gz

{codedx-0.1.0 → codedx-0.2.0}/.gitignore

@@ -10,8 +10,10 @@ wheels/
 # Virtual environments
 .venv
 
-# local config. while I do use claude to code this
-# I dont consider the agent config part of the project.
+# environment file. may contain secrets.
+.env
+
+# agent config is not part of the project.
 .claude
 
 # Downloaded at build time — not committed

{codedx-0.1.0 → codedx-0.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: codedx
-Version: 0.1.0
+Version: 0.2.0
 Summary: Swedish medical coding used for diagnosis codes
 Author-email: Ludvig Hult <ludvig.hult@gmail.com>
 Requires-Python: >=3.12
@@ -46,6 +46,13 @@ Dots are dropped throughout, matching what is typically found in databases: `A01
 pip install codedx
 ```
 
-Downloads data at install time from Socialstyrelsen, WHO, and CDC. Raw files are not redistributed due to licence restrictions.
+Downloads data at install time from Ehälsomyndigheten, WHO, and CDC. Raw files are not redistributed due to licence restrictions.
 
-> **Note:** Socialstyrelsen is migrating data to Ehälsomyndigheten — download URLs may stop working in the future.
+## Changelog
+
+### 0.2.0
+- ICD-10-SE and history file now downloaded from Ehälsomyndigheten (replacing Socialstyrelsen URLs)
+- Retired code lookup now uses the comprehensive history file (`historik-icd-10-se-1997-2026.xlsx`) covering all codes 1997–2026, instead of the partial inactivation list
+
+### 0.1.0
+- Initial release

{codedx-0.1.0 → codedx-0.2.0}/README.md

@@ -36,6 +36,13 @@ Dots are dropped throughout, matching what is typically found in databases: `A01
 pip install codedx
 ```
 
-Downloads data at install time from Socialstyrelsen, WHO, and CDC. Raw files are not redistributed due to licence restrictions.
+Downloads data at install time from Ehälsomyndigheten, WHO, and CDC. Raw files are not redistributed due to licence restrictions.
 
-> **Note:** Socialstyrelsen is migrating data to Ehälsomyndigheten — download URLs may stop working in the future.
+## Changelog
+
+### 0.2.0
+- ICD-10-SE and history file now downloaded from Ehälsomyndigheten (replacing Socialstyrelsen URLs)
+- Retired code lookup now uses the comprehensive history file (`historik-icd-10-se-1997-2026.xlsx`) covering all codes 1997–2026, instead of the partial inactivation list
+
+### 0.1.0
+- Initial release

{codedx-0.1.0 → codedx-0.2.0}/hatch_build.py

@@ -40,8 +40,8 @@ _MULTI_ZIPS = [
         "icd10se_latest.zip",
         "ICD-10-SE latest release (~15 MB)",
         [
-            ("icd-10-se.tsv", "https://www.socialstyrelsen.se/globalassets/sharepoint-dokument/dokument-webb/klassifikationer-och-koder/icd-10-se.tsv"),
-            ("andringar-icd-10-se.xlsx", "https://www.socialstyrelsen.se/globalassets/sharepoint-dokument/dokument-webb/klassifikationer-och-koder/andringar-icd-10-se.xlsx"),
+            ("icd-10-se.tsv", "https://samarbetsyta.ehalsomyndigheten.se/download/attachments/451267009/icd-10-se.tsv?api=v2"),
+            ("historik-icd-10-se-1997-2026.xlsx", "https://samarbetsyta.ehalsomyndigheten.se/download/attachments/451267009/historik-icd-10-se-1997-2026%20%281%29.xlsx?api=v2"),
         ],
     ),
 ]

{codedx-0.1.0 → codedx-0.2.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "codedx"
-version = "0.1.0"
+version = "0.2.0"
 description = "Swedish medical coding used for diagnosis codes"
 readme = "README.md"
 authors = [

{codedx-0.1.0 → codedx-0.2.0}/src/codedx/__init__.py

@@ -1,6 +1,6 @@
 import pathlib
 
-__version__ = "0.1.0"
+__version__ = "0.2.0"
 _CACHE_DIR = pathlib.Path.home() / ".cache" / "codedx" / f"v{__version__}"
 
 from codedx._core import (  # noqa: E402

{codedx-0.1.0 → codedx-0.2.0}/src/codedx/icd10se.py

@@ -52,19 +52,19 @@ def _build_retired(work_dir: pathlib.Path) -> pl.DataFrame:
         "Code": pl.Series([], dtype=pl.String),
         "Titel": pl.Series([], dtype=pl.String),
     })
-    xlsx = work_dir / "andringar-icd-10-se.xlsx"
-    if not xlsx.exists():
+    historik = work_dir / "historik-icd-10-se-1997-2026.xlsx"
+    if not historik.exists():
         return _empty
     try:
-        df = pl.read_excel(xlsx, sheet_name="Inaktiverade koder")
+        df = pl.read_excel(historik)
     except Exception:
         return _empty
-    if "Inaktiverad kod" not in df.columns:
+    if "Giltig till" not in df.columns:
         return _empty
     return (
-        df.filter(pl.col("Inaktiverad kod").is_not_null())
+        df.filter(pl.col("Giltig till").is_not_null() & (pl.col("Giltig till") != ""))
         .select(
-            pl.col("Inaktiverad kod").str.replace_all("\\.", "").alias("Code"),
+            pl.col("Kod").str.replace_all("\\.", "").alias("Code"),
             pl.col("Titel"),
         )
         .unique("Code")