RO. Catalog cloud-native cu date selectate din geoportalul național geodata.gov.md. Româna este limba de referință. Textele în rusă și engleză sunt traduceri. Catalogul nu este o publicație oficială a AGCC.
RU. Облачный каталог избранных данных национального геопортала geodata.gov.md. Румынский текст является основным; русский и английский тексты — переводы. Каталог не является официальной публикацией AGCC.
EN. A cloud-native catalog of selected data from Moldova’s national geoportal, geodata.gov.md. Romanian is authoritative; Russian and English are translations. This is not an official AGCC publication.
RO. Strazi (segmentele strazilor). Conține 35 104 linii extrase din
spațiul de lucru cadastru_data al serverului geodata.gov.md prin WFS 2.0.0
la data de 2026-08-16. Coordonatele sunt păstrate în sistemul sursă MOLDREF99
/ Moldova TM (EPSG:4026).
RU. Улицы (сегменты улиц). Содержит 35 104 линий, извлечённых из рабочего
пространства cadastru_data сервера geodata.gov.md через WFS 2.0.0 от
2026-08-16. Координаты сохранены в исходной системе MOLDREF99 / Moldova TM
(EPSG:4026).
EN. Streets (street segments). Holds 35 104 lines extracted from the
cadastru_data workspace of geodata.gov.md over WFS 2.0.0 on 2026-08-16.
Coordinates are kept in the source system, MOLDREF99 / Moldova TM (EPSG:4026).
35,104 features extracted from cadastru_data:Strazi_RM on the geodata.gov.md GeoServer over
WFS 2.0.0 on 2026-08-16, and verified against the server's own feature count.
Coordinates are MOLDREF99 / Moldova TM (EPSG:4026): projected, in metres, and
the source's own system. They are not reprojected. ST_Area returns square metres
directly, and a spatial predicate written against a WGS84 envelope will match
nothing without raising an error.
resultType=hits exactly, so nothing was
lost to a capped WFS response.Row count:
Rows carrying geometry. Some layers hold null geometries, so filter before a spatial predicate:
This file is Hilbert-ordered and carries a bbox struct with GeoParquet 1.1
covering metadata. That combination makes row groups spatially coherent, so a
bounding-box predicate lets the reader skip most of the file unread.
DuckDB does not apply it for you. Writing it yourself is worth 3-7x on a spatial query, measured on this catalog:
Coordinates are metres, so those bounds are a 10 km box.
cadastru_data_uat2 holds the 35 second-level administrative units and is the join
target for anything aggregated by raion or municipality. Both sides are EPSG:4026,
so a spatial join needs no transform. Prefilter on each raion's bounds so the
expensive predicate runs on a fraction of the rows:
The catalog's own AGENTS.md covers the join keys and the quirks that apply across every collection here.