names
5 rows where place_id = 675925
This data as json, CSV (advanced)
Suggested facets: role, surname, forename, occupation, seq
| id ▼ | case_id | role | surname | forename | occupation | place_id | raw_segment | seq |
|---|---|---|---|---|---|---|---|---|
| 4020665 | 2032119 2032119 | defendant | Barwell | George | gent | 675925 675925 | Barwell, George, of Saperton, Lincs, gent | 0 |
| 4020668 | 2032120 2032120 | defendant | Barwen or Barwell | George | gent | 675925 675925 | Barwen or Barwell, George, of Saperton, Lincs, gent | 0 |
| 4534402 | 2211032 2211032 | plaintiff | Saperton | John | 675925 675925 | Saperton, John, of Saperton, Lincs | 0 | |
| 4678000 | 2258734 2258734 | defendant | Saperton | John | gent | 675925 675925 | Saperton, John, of Saperton, Lincs, gent | 2 |
| 4842042 | 2309048 2309048 | defendant | Clerk | John | yeoman | 675925 675925 | Clerk, John, of Saperton, Lincs, yeoman | 0 |
Advanced export
JSON shape: default, array, newline-delimited, object
CREATE TABLE names (
id INTEGER PRIMARY KEY AUTOINCREMENT,
case_id INTEGER NOT NULL REFERENCES cases(id) ON DELETE CASCADE,
role TEXT NOT NULL CHECK (role IN ('plaintiff', 'defendant')),
surname TEXT NOT NULL,
forename TEXT,
occupation TEXT,
place_id INTEGER REFERENCES places(id),
raw_segment TEXT NOT NULL,
seq INTEGER NOT NULL DEFAULT 0
);
CREATE INDEX idx_names_case_id ON names(case_id);
CREATE INDEX idx_names_surname ON names(surname);
CREATE INDEX idx_names_place_id ON names(place_id);
CREATE INDEX idx_names_role_surname ON names(role, surname);