Logbook of code changes

git · February 24, 2026, 7:14pm

git · February 24, 2026, 7:14pm

git · February 24, 2026, 7:17pm

git · February 25, 2026, 5:40pm

git · February 25, 2026, 7:03pm

git · February 25, 2026, 7:03pm

git · February 25, 2026, 7:04pm

git · February 25, 2026, 8:22pm

github.com/earthobservations/wetterdienst

provider/eccc: migrate observation values to api.weather.gc.ca OGC API (#1616)

committed 08:21PM - 25 Feb 26 UTC

gutzbenj

+420 -642

* provider/eccc: migrate observation values to api.weather.gc.ca OGC API Replac…e the legacy CSV bulk download endpoint with the new api.weather.gc.ca OGC API for ECCC observation data retrieval. Changes: - Switch from multi-file CSV downloads to single GeoJSON API calls per resolution (climate-hourly, climate-daily, climate-monthly) - Rewrite _tidy_up_df to handle new _flag suffix quality columns using LazyFrame unpivot instead of manual column pairing - Update parameter metadata (name_original) to match new API column names (e.g. MEAN_TEMPERATURE instead of "temp (°c)") - Remove quality parameters as standalone metadata entries; quality is now derived automatically from *_flag columns - Expand timezone mapping to include daylight saving time variants (ADT, CDT, EDT, MDT, NDT, PDT, YDT, YST) Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> --------- Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

git · February 25, 2026, 9:17pm

github.com/earthobservations/wetterdienst

fix(dwd): update describe_fields for new PDF location and format

committed 09:17PM - 25 Feb 26 UTC

gutzbenj

+61 -46

DWD moved description PDFs from the period subdirectory (e.g. daily/kl/recent/) …to the dataset directory (daily/kl/) and updated the PDF content to a structured table format. Changes: - Strip the period path component from the URL when searching for description PDFs so the file index scans the correct directory - Fix get_column("filename") -> get_column("url") to use the full download URL instead of the bare filename - Rewrite parse_parameters() to handle the new table format where each row starts with an all-caps column name followed by description, unit, type and format on the same line - Rewrite read_description() to use position-based section extraction (csv content description -> Quality Information) instead of empty-line-stop logic - Update German section header from "Parameter" to "CSV Inhaltsbeschreibung" to match the new PDF structure - Update __main__ example URLs to current paths - Skip SQL type keywords (NUMBER, VARCHAR2) and the German section header artifact (CSV) that were misidentified as parameter names Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

git · February 25, 2026, 9:17pm

git · February 26, 2026, 1:15am

git · February 26, 2026, 1:15am

git · February 26, 2026, 1:15am

git · February 26, 2026, 1:15am

git · February 26, 2026, 1:15am

git · February 26, 2026, 1:15am

git · February 26, 2026, 1:15am

git · February 26, 2026, 3:51am

git · February 26, 2026, 3:51am

git · February 26, 2026, 3:51am