@nshiab/simple-data-analysis-core 0.0.20 → 0.0.22

package/script/class/SimpleTable.js

@@ -52,6 +52,7 @@ const normalizeQuery_js_1 = __importDefault(require("../methods/normalizeQuery.j
 const rollingQuery_js_1 = __importDefault(require("../methods/rollingQuery.js"));
 const distanceQuery_js_1 = __importDefault(require("../methods/distanceQuery.js"));
 const getGeoData_js_1 = __importDefault(require("../methods/getGeoData.js"));
+const writeGeoData_js_1 = __importDefault(require("../helpers/writeGeoData.js"));
 const splitSpread_js_1 = __importDefault(require("../methods/splitSpread.js"));
 const node_fs_1 = require("node:fs");
 const stringToArray_js_1 = __importDefault(require("../helpers/stringToArray.js"));
@@ -59,24 +60,20 @@ const loadDataQuery_js_1 = __importDefault(require("../methods/loadDataQuery.js"
 const mergeOptions_js_1 = __importDefault(require("../helpers/mergeOptions.js"));
 const queryDB_js_1 = __importDefault(require("../helpers/queryDB.js"));
 const writeDataQuery_js_1 = __importDefault(require("../methods/writeDataQuery.js"));
-const writeGeoDataQuery_js_1 = __importDefault(require("../methods/writeGeoDataQuery.js"));
 const runQuery_js_1 = __importDefault(require("../helpers/runQuery.js"));
 const aggregateGeoQuery_js_1 = __importDefault(require("../methods/aggregateGeoQuery.js"));
 const summarize_js_1 = __importDefault(require("../methods/summarize.js"));
 const correlations_js_1 = __importDefault(require("../methods/correlations.js"));
 const linearRegressions_js_1 = __importDefault(require("../methods/linearRegressions.js"));
 const joinGeo_js_1 = __importDefault(require("../methods/joinGeo.js"));
-const shouldFlipBeforeExport_js_1 = __importDefault(require("../helpers/shouldFlipBeforeExport.js"));
 const getProjection_js_1 = __importDefault(require("../helpers/getProjection.js"));
 const cache_js_1 = __importDefault(require("../methods/cache.js"));
 const camelCase_js_1 = __importDefault(require("../helpers/camelCase.js"));
 const formatNumber_js_1 = __importDefault(require("../helpers/formatNumber.js"));
 const createDirectory_js_1 = __importDefault(require("../helpers/createDirectory.js"));
-const rewind_js_1 = __importDefault(require("../helpers/rewind.js"));
 const writeDataAsArrays_js_1 = __importDefault(require("../helpers/writeDataAsArrays.js"));
 const logData_js_1 = __importDefault(require("../helpers/logData.js"));
 const fill_js_1 = __importDefault(require("../methods/fill.js"));
-const node_fs_2 = require("node:fs");
 const loadArray_js_1 = __importDefault(require("../methods/loadArray.js"));
 const cleanPath_js_1 = __importDefault(require("../helpers/cleanPath.js"));
 const Simple_js_1 = __importDefault(require("./Simple.js"));
@@ -93,16 +90,16 @@ const capitalizeQuery_js_1 = __importDefault(require("../methods/capitalizeQuery
 const truncateQuery_js_1 = __importDefault(require("../methods/truncateQuery.js"));
 const padQuery_js_1 = __importDefault(require("../methods/padQuery.js"));
 const getProjectionParquet_js_1 = __importDefault(require("../helpers/getProjectionParquet.js"));
+const hasGeometryColumn_js_1 = __importDefault(require("../helpers/hasGeometryColumn.js"));
 const unifyColumns_js_1 = __importDefault(require("../helpers/unifyColumns.js"));
 const accumulateQuery_js_1 = __importDefault(require("../helpers/accumulateQuery.js"));
-const stringifyDates_js_1 = __importDefault(require("../helpers/stringifyDates.js"));
-const stringifyDatesInvert_js_1 = __importDefault(require("../helpers/stringifyDatesInvert.js"));
 const unnestQuery_js_1 = __importDefault(require("../helpers/unnestQuery.js"));
 const nestQuery_js_1 = __importDefault(require("../helpers/nestQuery.js"));
 const concatenateRowQuery_js_1 = __importDefault(require("../helpers/concatenateRowQuery.js"));
 const createFtsIndex_js_1 = __importDefault(require("../methods/createFtsIndex.js"));
 const createVssIndex_js_1 = __importDefault(require("../methods/createVssIndex.js"));
 const bm25_js_1 = __importDefault(require("../methods/bm25.js"));
+const loadSample_js_1 = __importDefault(require("../methods/loadSample.js"));
 const normalizeString_js_1 = __importDefault(require("../methods/normalizeString.js"));
 /**
  * IMPORTANT: When extending this class, always use `this.sdb.newTable()` to
@@ -427,7 +424,13 @@ class SimpleTable extends Simple_js_1.default {
      *
      * @example
      * ```ts
-     * // Load geospatial data from a shapefile and reproject to WGS84
+     * // Load geospatial data from a shapefile (with relevant files in the same folder) and reproject to WGS84
+     * await table.loadGeoData("./some-data/some-data.shp", { toWGS84: true });
+     * ```
+     *
+     * @example
+     * ```ts
+     * // Load geospatial data from a zipped shapefile and reproject to WGS84
      * await table.loadGeoData("./some-data.shp.zip", { toWGS84: true });
      * ```
      */
@@ -811,6 +814,32 @@ class SimpleTable extends Simple_js_1.default {
             }
         }
     }
+    /**
+     * Fetches sample data from the simple-data-analysis-core GitHub repository.
+     *
+     * @param sample - The name of the sample to load.
+     *
+     * Tabular data:
+     * - "fires": [firesCanada2023.csv](https://raw.githubusercontent.com/nshiab/simple-data-analysis-core/refs/heads/main/test/geodata/files/firesCanada2023.csv)
+     * - "recipes": [recipes.parquet](https://github.com/nshiab/simple-data-analysis-core/raw/refs/heads/main/test/data/files/recipes.parquet)
+     * - "temperatures": [dailyTemperatures.csv](https://raw.githubusercontent.com/nshiab/simple-data-analysis-core/refs/heads/main/test/data/files/dailyTemperatures.csv)
+     * - "temperaturesCities": [cities.csv](https://raw.githubusercontent.com/nshiab/simple-data-analysis-core/refs/heads/main/test/data/files/cities.csv)
+     *
+     * Geospatial data:
+     * - "canada": [CanadianProvincesAndTerritories.json](https://raw.githubusercontent.com/nshiab/simple-data-analysis-core/refs/heads/main/test/geodata/files/CanadianProvincesAndTerritories.json)
+     * - "firesGeo": [firesCanada2023.geojson](https://raw.githubusercontent.com/nshiab/simple-data-analysis-core/refs/heads/main/test/geodata/files/firesCanada2023.geojson)
+     *
+     * @category Importing Data
+     *
+     * @example
+     * ```ts
+     * // Load the fires sample data
+     * await table.loadSample("fires");
+     * ```
+     */
+    async loadSample(sample) {
+        return (await (0, loadSample_js_1.default)(this, sample));
+    }
     /**
      * Returns a new table with the same structure and data as this table. The data can be optionally filtered.
      * Note that cloning large tables can be a slow operation.
@@ -1908,48 +1937,57 @@ class SimpleTable extends Simple_js_1.default {
      * @param rightTable - The SimpleTable instance to be joined with this table.
      * @param leftColumn - The name of the column in this (left) table containing the text to compare.
      * @param rightColumn - The name of the column in the right table containing the text to compare.
+     * @param threshold - The minimum similarity score (0–100) required for two rows to be joined. For `method: "ratio"`, a length-based pre-filter is automatically applied based on the threshold to improve performance without losing accuracy.
      * @param options - An optional object with configuration options:
      * @param options.method - The rapidfuzz similarity algorithm to use. Defaults to `"ratio"`.
      *   - `"ratio"`: Overall similarity (Levenshtein-based).
      *   - `"partial_ratio"`: Best partial/substring similarity.
      *   - `"token_sort_ratio"`: Similarity after sorting tokens (words), useful for reordered words.
      *   - `"token_set_ratio"`: Similarity based on sets of tokens, ignoring duplicates and word order.
-     * @param options.threshold - The minimum similarity score (0–100) required for two rows to be joined. Defaults to `80`.
      * @param options.similarityColumn - If provided, a column with this name is added to the result containing the similarity score (0–100). If omitted, the score is not included in the output.
      * @param options.outputTable - If `true`, the results will be stored in a new table with a generated name. If a string, it will be used as the name for the new table. If `false` or omitted, the current table will be overwritten. Defaults to `false`.
+     * @param options.preFilterPrefixLen - An optional prefix length. Only strings sharing the same first N characters are compared. Note that prefix filtering is lossy (e.g. "John" vs. "Phon" will not match despite high similarity).
      * @returns A promise that resolves to a table instance containing the fuzzy-joined data (either the modified current table or a new table).
      * @category Table Operations
      *
      * @example
      * ```ts
-     * // Fuzzy left join tableA with tableB on 'name' (left) and 'standardName' (right) (ratio >= 80)
-     * await tableA.fuzzyJoin(tableB, "name", "standardName");
+     * // Fuzzy left join tableA with tableB on 'name' (left) and 'standardName' (right) with a threshold of 80
+     * // A length-based pre-filter is automatically applied.
+     * await tableA.fuzzyJoin(tableB, "name", "standardName", 80);
+     * ```
+     *
+     * @example
+     * ```ts
+     * // Fuzzy join with a prefix-based pre-filter and a threshold of 80
+     * await tableA.fuzzyJoin(tableB, "name", "standardName", 80, {
+     *   preFilterPrefixLen: 3, // Must share the same first 3 characters
+     * });
      * ```
      *
      * @example
      * ```ts
      * // Fuzzy join with a custom threshold and method, storing results in a new table
-     * const tableC = await tableA.fuzzyJoin(tableB, "name", "standardName", {
+     * const tableC = await tableA.fuzzyJoin(tableB, "name", "standardName", 90, {
      *   method: "token_sort_ratio",
-     *   threshold: 90,
      *   outputTable: "tableC",
      * });
      * ```
      *
      * @example
      * ```ts
-     * // Fuzzy join with a custom similarity column name
-     * await tableA.fuzzyJoin(tableB, "name", "standardName", {
+     * // Fuzzy join with a custom similarity column name and a threshold of 80
+     * await tableA.fuzzyJoin(tableB, "name", "standardName", 80, {
      *   similarityColumn: "matchScore",
      * });
      * ```
      */
-    async fuzzyJoin(rightTable, leftColumn, rightColumn, options = {}) {
+    async fuzzyJoin(rightTable, leftColumn, rightColumn, threshold, options = {}) {
         if (options.outputTable === true) {
             options.outputTable = `table${this.sdb.tableIncrement}`;
             this.sdb.tableIncrement += 1;
         }
-        return await (0, fuzzyJoin_js_1.default)(this, rightTable, leftColumn, rightColumn, options);
+        return await (0, fuzzyJoin_js_1.default)(this, rightTable, leftColumn, rightColumn, threshold, options);
     }
     /**
      * Normalizes string values in a column by detecting fuzzy duplicates and replacing them with a single canonical value.
@@ -1963,42 +2001,52 @@ class SimpleTable extends Simple_js_1.default {
      *
      * @param column - The name of the column containing the strings to normalize.
      * @param newColumn - The name of the column to write the normalized values to. Use the same name as `column` to normalize in-place.
+     * @param threshold - The minimum similarity score (0–100) for two strings to be considered duplicates. For `method: "ratio"`, a length-based pre-filter is automatically applied based on the threshold to improve performance without losing accuracy.
      * @param options - An optional object with configuration options:
      * @param options.method - The rapidfuzz similarity algorithm to use. Defaults to `"ratio"`.
      *   - `"ratio"`: Overall similarity.
      *   - `"partial_ratio"`: Best partial/substring similarity.
      *   - `"token_sort_ratio"`: Similarity after sorting tokens (words), useful for reordered words.
      *   - `"token_set_ratio"`: Similarity based on sets of tokens, ignoring duplicates and word order.
-     * @param options.threshold - The minimum similarity score (0–100) for two strings to be considered duplicates. Defaults to `80`.
      * @param options.keep - The strategy for choosing the canonical value within each cluster of similar strings. Defaults to `"mostCommon"`.
      *   - `"mostCommon"`: Keep the value that appears most frequently in the original column.
      *   - `"longestString"`: Keep the longest string in the cluster.
      *   - `"shortestString"`: Keep the shortest string in the cluster.
      *   - `"mostCentral"`: Keep the string with the highest total similarity score to all other cluster members (the most "central" string).
      *   - `"maxScore"`: Keep the string that participates in the single highest-scoring pairwise match within the cluster.
+     * @param options.preFilterPrefixLen - An optional prefix length. Only strings sharing the same first N characters are compared. Note that prefix filtering is lossy (e.g. "John" vs. "Phon" will not match despite high similarity).
      * @returns A promise that resolves when the column has been normalized.
      * @category Updating Data
      *
      * @example
      * ```ts
-     * // Normalize 'city' into a new 'cityClean' column, keeping the most common string per cluster
-     * await table.fuzzyClean("city", "cityClean");
+     * // Normalize 'city' into a new 'cityClean' column, keeping the most common string per cluster with a threshold of 80
+     * // A length-based pre-filter is automatically applied.
+     * await table.fuzzyClean("city", "cityClean", 80);
+     * ```
+     *
+     * @example
+     * ```ts
+     * // Normalize with a prefix-based pre-filter and a threshold of 80
+     * await table.fuzzyClean("city", "cityClean", 80, {
+     *   preFilterPrefixLen: 5, // Must share the same first 5 characters
+     * });
      * ```
      *
      * @example
      * ```ts
-     * // Normalize 'companyName' into a new column using token_sort_ratio and a stricter threshold
-     * await table.fuzzyClean("companyName", "companyNameClean", { method: "token_sort_ratio", threshold: 90 });
+     * // Normalize 'companyName' into a new column using token_sort_ratio and a threshold of 90
+     * await table.fuzzyClean("companyName", "companyNameClean", 90, { method: "token_sort_ratio" });
      * ```
      *
      * @example
      * ```ts
-     * // Normalize 'category' in-place, keeping the longest string in each cluster
-     * await table.fuzzyClean("category", "category", { keep: "longestString" });
+     * // Normalize 'category' in-place, keeping the longest string in each cluster and a threshold of 80
+     * await table.fuzzyClean("category", "category", 80, { keep: "longestString" });
      * ```
      */
-    async fuzzyClean(column, newColumn, options = {}) {
-        await (0, fuzzyClean_js_1.default)(this, column, newColumn, options);
+    async fuzzyClean(column, newColumn, threshold, options = {}) {
+        await (0, fuzzyClean_js_1.default)(this, column, newColumn, threshold, options);
     }
     /**
      * Replaces specified strings in the selected columns.
@@ -3930,6 +3978,9 @@ class SimpleTable extends Simple_js_1.default {
      * ```
      */
     async getData(options = {}) {
+        if (await (0, hasGeometryColumn_js_1.default)(this)) {
+            throw new Error("Table contains geometry columns. Use getGeoData() instead.");
+        }
         const columns = options.columns
             ? (typeof options.columns === "string"
                 ? [options.columns]
@@ -3994,8 +4045,9 @@ class SimpleTable extends Simple_js_1.default {
      * ```
      */
     async points(columnLat, columnLon, newColumn) {
-        await (0, queryDB_js_1.default)(this, `INSTALL spatial; LOAD spatial;
-            ALTER TABLE "${this.name}" ADD COLUMN "${newColumn}" GEOMETRY; UPDATE "${this.name}" SET "${newColumn}" = ST_Point2D("${columnLat}", "${columnLon}")`, (0, mergeOptions_js_1.default)(this, {
+        await (0, queryDB_js_1.default)(this, (await this.getColumns()).includes(newColumn)
+            ? `INSTALL spatial; LOAD spatial; UPDATE "${this.name}" SET "${newColumn}" = ST_Point2D("${columnLat}", "${columnLon}")`
+            : `INSTALL spatial; LOAD spatial; ALTER TABLE "${this.name}" ADD COLUMN "${newColumn}" GEOMETRY; UPDATE "${this.name}" SET "${newColumn}" = ST_Point2D("${columnLat}", "${columnLon}")`, (0, mergeOptions_js_1.default)(this, {
             table: this.name,
             method: "points()",
             parameters: { columnLat, columnLon, newColumn },
@@ -4411,7 +4463,9 @@ class SimpleTable extends Simple_js_1.default {
         const column = typeof options.column === "string"
             ? options.column
             : await (0, findGeoColumn_js_1.default)(this);
-        await (0, queryDB_js_1.default)(this, `ALTER TABLE "${this.name}" ADD "${newColumn}" GEOMETRY; UPDATE "${this.name}" SET "${newColumn}" =  ST_Buffer("${column}", ${distance});`, (0, mergeOptions_js_1.default)(this, {
+        await (0, queryDB_js_1.default)(this, (await this.getColumns()).includes(newColumn)
+            ? `INSTALL spatial; LOAD spatial; UPDATE "${this.name}" SET "${newColumn}" = ST_Buffer("${column}", ${distance})`
+            : `INSTALL spatial; LOAD spatial; ALTER TABLE "${this.name}" ADD "${newColumn}" GEOMETRY; UPDATE "${this.name}" SET "${newColumn}" = ST_Buffer("${column}", ${distance})`, (0, mergeOptions_js_1.default)(this, {
             table: this.name,
             method: "buffer()",
             parameters: { column, newColumn, distance },
@@ -4497,7 +4551,9 @@ class SimpleTable extends Simple_js_1.default {
         if (this.projections[column1] !== this.projections[column2]) {
             throw new Error(`${column1} and ${column2} don't have the same projection.\n${column1}: ${this.projections[column1]}\n${column2}: ${this.projections[column2]}`);
         }
-        await (0, queryDB_js_1.default)(this, `ALTER TABLE "${this.name}" ADD "${newColumn}" GEOMETRY; UPDATE "${this.name}" SET "${newColumn}" = ST_Intersection("${column1}", "${column2}")`, (0, mergeOptions_js_1.default)(this, {
+        await (0, queryDB_js_1.default)(this, (await this.getColumns()).includes(newColumn)
+            ? `INSTALL spatial; LOAD spatial; UPDATE "${this.name}" SET "${newColumn}" = ST_Intersection("${column1}", "${column2}")`
+            : `INSTALL spatial; LOAD spatial; ALTER TABLE "${this.name}" ADD "${newColumn}" GEOMETRY; UPDATE "${this.name}" SET "${newColumn}" = ST_Intersection("${column1}", "${column2}")`, (0, mergeOptions_js_1.default)(this, {
             table: this.name,
             method: "intersection()",
             parameters: { column1, column2, newColumn },
@@ -4622,7 +4678,9 @@ class SimpleTable extends Simple_js_1.default {
         if (this.projections[column1] !== this.projections[column2]) {
             throw new Error(`${column1} and ${column2} don't have the same projection.\n${column1}: ${this.projections[column1]}\n${column2}: ${this.projections[column2]}`);
         }
-        await (0, queryDB_js_1.default)(this, `ALTER TABLE "${this.name}" ADD "${newColumn}" GEOMETRY; UPDATE "${this.name}" SET "${newColumn}" = ST_Union("${column1}", "${column2}")`, (0, mergeOptions_js_1.default)(this, {
+        await (0, queryDB_js_1.default)(this, (await this.getColumns()).includes(newColumn)
+            ? `INSTALL spatial; LOAD spatial; UPDATE "${this.name}" SET "${newColumn}" = ST_Union("${column1}", "${column2}")`
+            : `INSTALL spatial; LOAD spatial; ALTER TABLE "${this.name}" ADD "${newColumn}" GEOMETRY; UPDATE "${this.name}" SET "${newColumn}" = ST_Union("${column1}", "${column2}")`, (0, mergeOptions_js_1.default)(this, {
             table: this.name,
             method: "union()",
             parameters: { column1, column2, newColumn },
@@ -4713,7 +4771,9 @@ class SimpleTable extends Simple_js_1.default {
         const column = typeof options.column === "string"
             ? options.column
             : await (0, findGeoColumn_js_1.default)(this);
-        await (0, queryDB_js_1.default)(this, `ALTER TABLE "${this.name}" ADD "${newColumn}" GEOMETRY; UPDATE "${this.name}" SET "${newColumn}" =  ST_Centroid("${column}");`, (0, mergeOptions_js_1.default)(this, {
+        await (0, queryDB_js_1.default)(this, (await this.getColumns()).includes(newColumn)
+            ? `INSTALL spatial; LOAD spatial; UPDATE "${this.name}" SET "${newColumn}" = ST_Centroid("${column}")`
+            : `INSTALL spatial; LOAD spatial; ALTER TABLE "${this.name}" ADD "${newColumn}" GEOMETRY; UPDATE "${this.name}" SET "${newColumn}" = ST_Centroid("${column}")`, (0, mergeOptions_js_1.default)(this, {
             table: this.name,
             method: "centroid()",
             parameters: { column, newColumn },
@@ -4998,6 +5058,9 @@ class SimpleTable extends Simple_js_1.default {
      * ```
      */
     async writeData(file, options = {}) {
+        if (await (0, hasGeometryColumn_js_1.default)(this)) {
+            throw new Error("Table contains geometry columns. Use writeGeoData() instead.");
+        }
         (0, createDirectory_js_1.default)(file);
         const extension = (0, getExtension_js_1.default)(file);
         if (options.dataAsArrays) {
@@ -5012,12 +5075,12 @@ class SimpleTable extends Simple_js_1.default {
         }
     }
     /**
-     * Writes the table's geospatial data to a file in GeoJSON or GeoParquet format.
+     * Writes the table's geospatial data to a file in GeoJSON, GeoParquet, or Shapefile format.
      * If the specified path does not exist, it will be created.
      *
      * For GeoJSON files (`.geojson` or `.json`), if the projection is WGS84 or EPSG:4326 (`[latitude, longitude]` axis order), the coordinates will be flipped to follow the RFC7946 standard (`[longitude, latitude]` axis order) in the output.
      *
-     * @param file - The absolute path to the output file (e.g., `"./output.geojson"`, `"./output.geoparquet"`).
+     * @param file - The absolute path to the output file (e.g., `"./output.geojson"`, `"./output.geoparquet"`, `"./shapefile-folder/output.shp"`).
      * @param options - An optional object with configuration options:
      * @param options.precision - For GeoJSON, the maximum number of figures after the decimal separator to write in coordinates. Defaults to `undefined` (full precision).
      * @param options.compression - For GeoParquet, if `true`, the output will be ZSTD compressed. Defaults to `false`.
@@ -5041,6 +5104,12 @@ class SimpleTable extends Simple_js_1.default {
      *
      * @example
      * ```ts
+     * // Write geospatial data to a Shapefile with all relevant files  in the same folder
+     * await table.writeGeoData("./shapefile-folder/output.shp");
+     * ```
+     *
+     * @example
+     * ```ts
      * // Write GeoJSON with specific precision and metadata
      * await table.writeGeoData("./output_high_precision.geojson", {
      *   precision: 6,
@@ -5049,71 +5118,7 @@ class SimpleTable extends Simple_js_1.default {
      * ```
      */
     async writeGeoData(file, options = {}) {
-        (0, createDirectory_js_1.default)(file);
-        const fileExtension = (0, getExtension_js_1.default)(file);
-        if (fileExtension === "geojson" || fileExtension === "json") {
-            let types;
-            if (options.formatDates === true) {
-                types = await this.getTypes();
-                if (Object.values(types).includes("DATE") ||
-                    Object.values(types).includes("TIMESTAMP")) {
-                    await (0, stringifyDates_js_1.default)(this, types);
-                }
-            }
-            if (typeof options.compression === "boolean") {
-                throw new Error("The compression option is not supported for writing GeoJSON files.");
-            }
-            const geoColumn = await (0, findGeoColumn_js_1.default)(this);
-            const flip = (0, shouldFlipBeforeExport_js_1.default)(this.projections[geoColumn]);
-            if (flip) {
-                await this.flipCoordinates(geoColumn);
-                await (0, queryDB_js_1.default)(this, (0, writeGeoDataQuery_js_1.default)(this.name, file, fileExtension, options), (0, mergeOptions_js_1.default)(this, {
-                    table: this.name,
-                    method: "writeGeoData()",
-                    parameters: { file, options },
-                }));
-                await this.flipCoordinates(geoColumn);
-            }
-            else {
-                await (0, queryDB_js_1.default)(this, (0, writeGeoDataQuery_js_1.default)(this.name, file, fileExtension, options), (0, mergeOptions_js_1.default)(this, {
-                    table: this.name,
-                    method: "writeGeoData()",
-                    parameters: { file, options },
-                }));
-            }
-            if (options.metadata) {
-                const fileData = JSON.parse((0, node_fs_2.readFileSync)(file, "utf-8"));
-                fileData.metadata = options.metadata;
-                (0, node_fs_2.writeFileSync)(file, JSON.stringify(fileData));
-            }
-            if (options.rewind) {
-                const fileData = JSON.parse((0, node_fs_2.readFileSync)(file, "utf-8"));
-                const fileRewinded = (0, rewind_js_1.default)(fileData);
-                (0, node_fs_2.writeFileSync)(file, JSON.stringify(fileRewinded));
-            }
-            if (types && (Object.values(types).includes("DATE") ||
-                Object.values(types).includes("TIMESTAMP"))) {
-                await (0, stringifyDatesInvert_js_1.default)(this, types);
-            }
-        }
-        else if (fileExtension === "geoparquet") {
-            if (typeof options.precision === "number") {
-                throw new Error("The precision option is not supported for writing PARQUET files. Use the .reducePrecision() method.");
-            }
-            if (typeof options.rewind === "boolean") {
-                throw new Error("The rewind option is not supported for writing PARQUET files.");
-            }
-            await (0, queryDB_js_1.default)(this, `COPY "${this.name}" TO '${(0, cleanPath_js_1.default)(file)}' WITH (FORMAT PARQUET${options.compression === true ? ", COMPRESSION 'zstd'" : ""}, KV_METADATA {
-             projections: '${JSON.stringify(this.projections)}'
-        });`, (0, mergeOptions_js_1.default)(this, {
-                table: this.name,
-                method: "writeGeoData()",
-                parameters: { file, options },
-            }));
-        }
-        else {
-            throw new Error(`Unknown extension ${fileExtension}`);
-        }
+        await (0, writeGeoData_js_1.default)(this, file, options);
     }
     /**
      * Caches the results of computations in `./.sda-cache`.

package/script/helpers/findGeoColumn.js

@@ -6,7 +6,7 @@ async function findGeoColumn(SimpleTable) {
     const types = await SimpleTable.getTypes();
     const geometries = Object.values(types).filter((d) => d.toLowerCase() === "geometry");
     if (geometries.length === 0) {
-        throw new Error("No column storing geometries");
+        throw new Error("Table contains no geometry columns.");
     }
     else if (geometries.length > 1) {
         throw new Error("More than one column storing geometries. If the method allows to specify one, do it. Otherwise, use the selectColumns methods beforehand.");

package/script/helpers/getProjection.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"getProjection.d.ts","sourceRoot":"","sources":["../../src/helpers/getProjection.ts"],"names":[],"mappings":"AAEA,OAAO,KAAK,QAAQ,MAAM,sBAAsB,CAAC;AAGjD,wBAA8B,aAAa,CACzC,QAAQ,EAAE,QAAQ,EAClB,IAAI,EAAE,MAAM,mBA8Bb"}
+{"version":3,"file":"getProjection.d.ts","sourceRoot":"","sources":["../../src/helpers/getProjection.ts"],"names":[],"mappings":"AAEA,OAAO,KAAK,QAAQ,MAAM,sBAAsB,CAAC;AAGjD,wBAA8B,aAAa,CACzC,QAAQ,EAAE,QAAQ,EAClB,IAAI,EAAE,MAAM,mBAiCb"}

package/script/helpers/getProjection.js

@@ -21,6 +21,9 @@ async function getProjection(simpleDB, file) {
         throw new Error("No queryResults");
     }
     const proj4 = queryResult[0].proj4;
+    if (proj4 === null) {
+        return "UNKNOWN";
+    }
     if (typeof proj4 !== "string") {
         throw new Error(`Expected proj4 to be a string, got ${typeof proj4}`);
     }

package/script/helpers/hasGeometryColumn.d.ts

@@ -0,0 +1,20 @@
+import type SimpleTable from "../class/SimpleTable.js";
+/**
+ * Returns `true` if the table has one or more columns of type geometry.
+ *
+ * Uses `getTypes()` to inspect column types and checks for any column
+ * whose type normalizes to `"geometry"`.
+ *
+ * @param table - The SimpleTable instance to inspect.
+ * @returns `true` if at least one geometry column exists, `false` otherwise.
+ *
+ * @example
+ * ```ts
+ * const hasGeo = await hasGeometryColumn(table);
+ * if (hasGeo) {
+ *   console.log("This table contains geometry columns");
+ * }
+ * ```
+ */
+export default function hasGeometryColumn(table: SimpleTable): Promise<boolean>;
+//# sourceMappingURL=hasGeometryColumn.d.ts.map

package/script/helpers/hasGeometryColumn.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"hasGeometryColumn.d.ts","sourceRoot":"","sources":["../../src/helpers/hasGeometryColumn.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,WAAW,MAAM,yBAAyB,CAAC;AAEvD;;;;;;;;;;;;;;;;GAgBG;AACH,wBAA8B,iBAAiB,CAC7C,KAAK,EAAE,WAAW,GACjB,OAAO,CAAC,OAAO,CAAC,CAGlB"}

package/script/helpers/hasGeometryColumn.js

@@ -0,0 +1,24 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.default = hasGeometryColumn;
+/**
+ * Returns `true` if the table has one or more columns of type geometry.
+ *
+ * Uses `getTypes()` to inspect column types and checks for any column
+ * whose type normalizes to `"geometry"`.
+ *
+ * @param table - The SimpleTable instance to inspect.
+ * @returns `true` if at least one geometry column exists, `false` otherwise.
+ *
+ * @example
+ * ```ts
+ * const hasGeo = await hasGeometryColumn(table);
+ * if (hasGeo) {
+ *   console.log("This table contains geometry columns");
+ * }
+ * ```
+ */
+async function hasGeometryColumn(table) {
+    const types = await table.getTypes();
+    return Object.values(types).some((t) => t.toLowerCase() === "geometry");
+}

package/script/helpers/writeGeoData.d.ts

@@ -0,0 +1,9 @@
+import type SimpleTable from "../class/SimpleTable.js";
+export default function writeGeoData(table: SimpleTable, file: string, options?: {
+    precision?: number;
+    compression?: boolean;
+    rewind?: boolean;
+    metadata?: unknown;
+    formatDates?: boolean;
+}): Promise<void>;
+//# sourceMappingURL=writeGeoData.d.ts.map

package/script/helpers/writeGeoData.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"writeGeoData.d.ts","sourceRoot":"","sources":["../../src/helpers/writeGeoData.ts"],"names":[],"mappings":"AAaA,OAAO,KAAK,WAAW,MAAM,yBAAyB,CAAC;AAEvD,wBAA8B,YAAY,CACxC,KAAK,EAAE,WAAW,EAClB,IAAI,EAAE,MAAM,EACZ,OAAO,GAAE;IACP,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,WAAW,CAAC,EAAE,OAAO,CAAC;IACtB,MAAM,CAAC,EAAE,OAAO,CAAC;IACjB,QAAQ,CAAC,EAAE,OAAO,CAAC;IACnB,WAAW,CAAC,EAAE,OAAO,CAAC;CAClB,GACL,OAAO,CAAC,IAAI,CAAC,CAqIf"}

package/script/helpers/writeGeoData.js

@@ -0,0 +1,116 @@
+"use strict";
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.default = writeGeoData;
+const node_fs_1 = require("node:fs");
+const createDirectory_js_1 = __importDefault(require("./createDirectory.js"));
+const getExtension_js_1 = __importDefault(require("./getExtension.js"));
+const findGeoColumn_js_1 = __importDefault(require("./findGeoColumn.js"));
+const hasGeometryColumn_js_1 = __importDefault(require("./hasGeometryColumn.js"));
+const shouldFlipBeforeExport_js_1 = __importDefault(require("./shouldFlipBeforeExport.js"));
+const queryDB_js_1 = __importDefault(require("./queryDB.js"));
+const writeGeoDataQuery_js_1 = __importDefault(require("../methods/writeGeoDataQuery.js"));
+const mergeOptions_js_1 = __importDefault(require("./mergeOptions.js"));
+const rewind_js_1 = __importDefault(require("./rewind.js"));
+const stringifyDates_js_1 = __importDefault(require("./stringifyDates.js"));
+const stringifyDatesInvert_js_1 = __importDefault(require("./stringifyDatesInvert.js"));
+const cleanPath_js_1 = __importDefault(require("./cleanPath.js"));
+async function writeGeoData(table, file, options = {}) {
+    (0, createDirectory_js_1.default)(file);
+    if (!(await (0, hasGeometryColumn_js_1.default)(table))) {
+        throw new Error("Table contains no geometry columns. Use writeData() instead.");
+    }
+    const fileExtension = (0, getExtension_js_1.default)(file);
+    if (fileExtension === "geojson" || fileExtension === "json") {
+        let types;
+        if (options.formatDates === true) {
+            types = await table.getTypes();
+            if (Object.values(types).includes("DATE") ||
+                Object.values(types).includes("TIMESTAMP")) {
+                await (0, stringifyDates_js_1.default)(table, types);
+            }
+        }
+        if (typeof options.compression === "boolean") {
+            throw new Error("The compression option is not supported for writing GeoJSON files.");
+        }
+        const geoColumn = await (0, findGeoColumn_js_1.default)(table);
+        const flip = (0, shouldFlipBeforeExport_js_1.default)(table.projections[geoColumn]);
+        if (flip) {
+            await table.flipCoordinates(geoColumn);
+            await (0, queryDB_js_1.default)(table, (0, writeGeoDataQuery_js_1.default)(table.name, file, fileExtension, options), (0, mergeOptions_js_1.default)(table, {
+                table: table.name,
+                method: "writeGeoData()",
+                parameters: { file, options },
+            }));
+            await table.flipCoordinates(geoColumn);
+        }
+        else {
+            await (0, queryDB_js_1.default)(table, (0, writeGeoDataQuery_js_1.default)(table.name, file, fileExtension, options), (0, mergeOptions_js_1.default)(table, {
+                table: table.name,
+                method: "writeGeoData()",
+                parameters: { file, options },
+            }));
+        }
+        if (options.metadata) {
+            const fileData = JSON.parse((0, node_fs_1.readFileSync)(file, "utf-8"));
+            fileData.metadata = options.metadata;
+            (0, node_fs_1.writeFileSync)(file, JSON.stringify(fileData));
+        }
+        if (options.rewind) {
+            const fileData = JSON.parse((0, node_fs_1.readFileSync)(file, "utf-8"));
+            const fileRewinded = (0, rewind_js_1.default)(fileData);
+            (0, node_fs_1.writeFileSync)(file, JSON.stringify(fileRewinded));
+        }
+        if (types && (Object.values(types).includes("DATE") ||
+            Object.values(types).includes("TIMESTAMP"))) {
+            await (0, stringifyDatesInvert_js_1.default)(table, types);
+        }
+    }
+    else if (fileExtension === "shp") {
+        if (typeof options.precision === "number" ||
+            typeof options.compression === "boolean" ||
+            typeof options.rewind === "boolean" ||
+            options.metadata ||
+            options.formatDates === true) {
+            throw new Error("The following options are not supported for writing SHAPEFILE files: precision, compression, rewind, metadata, and formatDates.");
+        }
+        const geoColumn = await (0, findGeoColumn_js_1.default)(table);
+        const flip = (0, shouldFlipBeforeExport_js_1.default)(table.projections[geoColumn]);
+        if (flip) {
+            await table.flipCoordinates(geoColumn);
+            await (0, queryDB_js_1.default)(table, (0, writeGeoDataQuery_js_1.default)(table.name, file, fileExtension, options), (0, mergeOptions_js_1.default)(table, {
+                table: table.name,
+                method: "writeGeoData()",
+                parameters: { file, options },
+            }));
+            await table.flipCoordinates(geoColumn);
+        }
+        else {
+            await (0, queryDB_js_1.default)(table, (0, writeGeoDataQuery_js_1.default)(table.name, file, fileExtension, options), (0, mergeOptions_js_1.default)(table, {
+                table: table.name,
+                method: "writeGeoData()",
+                parameters: { file, options },
+            }));
+        }
+    }
+    else if (fileExtension === "geoparquet") {
+        if (typeof options.precision === "number") {
+            throw new Error("The precision option is not supported for writing PARQUET files. Use the .reducePrecision() method.");
+        }
+        if (typeof options.rewind === "boolean") {
+            throw new Error("The rewind option is not supported for writing PARQUET files.");
+        }
+        await (0, queryDB_js_1.default)(table, `COPY "${table.name}" TO '${(0, cleanPath_js_1.default)(file)}' WITH (FORMAT PARQUET${options.compression === true ? ", COMPRESSION 'zstd'" : ""}, KV_METADATA {
+             projections: '${JSON.stringify(table.projections)}'
+        });`, (0, mergeOptions_js_1.default)(table, {
+            table: table.name,
+            method: "writeGeoData()",
+            parameters: { file, options },
+        }));
+    }
+    else {
+        throw new Error(`Unknown extension ${fileExtension}`);
+    }
+}

package/script/methods/fuzzyClean.d.ts

@@ -1,7 +1,7 @@
 import type SimpleTable from "../class/SimpleTable.js";
-export default function fuzzyClean(table: SimpleTable, column: string, newColumn: string, options?: {
+export default function fuzzyClean(table: SimpleTable, column: string, newColumn: string, threshold: number, options?: {
     method?: "ratio" | "partial_ratio" | "token_sort_ratio" | "token_set_ratio";
-    threshold?: number;
     keep?: "mostCommon" | "longestString" | "shortestString" | "mostCentral" | "maxScore";
+    preFilterPrefixLen?: number;
 }): Promise<void>;
 //# sourceMappingURL=fuzzyClean.d.ts.map

package/script/methods/fuzzyClean.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"fuzzyClean.d.ts","sourceRoot":"","sources":["../../src/methods/fuzzyClean.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,WAAW,MAAM,yBAAyB,CAAC;AAIvD,wBAA8B,UAAU,CACtC,KAAK,EAAE,WAAW,EAClB,MAAM,EAAE,MAAM,EACd,SAAS,EAAE,MAAM,EACjB,OAAO,GAAE;IACP,MAAM,CAAC,EACH,OAAO,GACP,eAAe,GACf,kBAAkB,GAClB,iBAAiB,CAAC;IACtB,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,IAAI,CAAC,EACD,YAAY,GACZ,eAAe,GACf,gBAAgB,GAChB,aAAa,GACb,UAAU,CAAC;CACX,GACL,OAAO,CAAC,IAAI,CAAC,CAyNf"}
+{"version":3,"file":"fuzzyClean.d.ts","sourceRoot":"","sources":["../../src/methods/fuzzyClean.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,WAAW,MAAM,yBAAyB,CAAC;AAIvD,wBAA8B,UAAU,CACtC,KAAK,EAAE,WAAW,EAClB,MAAM,EAAE,MAAM,EACd,SAAS,EAAE,MAAM,EACjB,SAAS,EAAE,MAAM,EACjB,OAAO,GAAE;IACP,MAAM,CAAC,EACH,OAAO,GACP,eAAe,GACf,kBAAkB,GAClB,iBAAiB,CAAC;IACtB,IAAI,CAAC,EACD,YAAY,GACZ,eAAe,GACf,gBAAgB,GAChB,aAAa,GACb,UAAU,CAAC;IACf,kBAAkB,CAAC,EAAE,MAAM,CAAC;CACxB,GACL,OAAO,CAAC,IAAI,CAAC,CAmPf"}