Working GM_GEO_LOOKUP_TABLE

docs/README.md

@@ -10,6 +10,7 @@ Gapminder-specific custom functions for Google Spreadsheets.
 
 * [GM_AGGR](#gm_aggr)
 * [GM_DATA](#gm_data)
+* [GM_GEO_LOOKUP_TABLE](#gm_geo_lookup_table)
 * [GM_GROWTH](#gm_growth)
 * [GM_ID](#gm_id)
 * [GM_IMPORT](#gm_import)
@@ -80,6 +81,28 @@ Takes 2-4 seconds: =GM\_DATA(B7:D, "pop", "year", "countries\_etc", 'data:pop:ye
 **Returns:** `any`[][]
 A two-dimensional array containing the cell/column contents described above in the summary.
 
+___
+<a id="gm_geo_lookup_table"></a>
+
+###  GM_GEO_LOOKUP_TABLE
+
+▸ **GM_GEO_LOOKUP_TABLE**(geography: *`string`*): `string`[][]
+
+*Defined in GM_GEO_LOOKUP_TABLE.ts:18*
+
+Inserts a table with Gapminder’s geo ids together with their aliases (all spellings we have seen before), including lower cased variants without diacritics and special characters to allow for somewhat fuzzy matching.
+
+To be used as the source range for VLOOKUP where the dataset is too large for GM\_ID or GM\_NAME to be used directly.
+
+**Parameters:**
+
+| Name | Type | Description |
+| ------ | ------ | ------ |
+| geography | `string` |  Should be one of the sets listed in the gapminder geo ontology such as "countries\_etc" |
+
+**Returns:** `string`[][]
+A two-dimensional array containing the cell/column contents described above in the summary.
+
 ___
 <a id="gm_growth"></a>
 

src/GM_GEO_LOOKUP_TABLE.spec.ts

@@ -0,0 +1,36 @@
+import test, { ExecutionContext, Macro } from "ava";
+import { GM_GEO_LOOKUP_TABLE } from "./GM_GEO_LOOKUP_TABLE";
+import { MinimalUrlFetchApp } from "./lib/MinimalUrlFetchApp";
+import { MinimalUtilities } from "./lib/MinimalUtilities";
+(global as any).UrlFetchApp = MinimalUrlFetchApp;
+(global as any).Utilities = MinimalUtilities;
+
+/**
+ * @hidden
+ */
+const testGmGeoLookupTable: Macro<any> = (
+  t: ExecutionContext,
+  { geography, expectedTopFiveRowsOfOutput }
+) => {
+  const output = GM_GEO_LOOKUP_TABLE(geography);
+  const topFiveRowsOfOutput = output.slice(0, 5);
+  // t.log({ topFiveRowsOfOutput, expectedTopFiveRowsOfOutput });
+  t.deepEqual(topFiveRowsOfOutput, expectedTopFiveRowsOfOutput);
+};
+
+[
+  /* tslint:disable:object-literal-sort-keys */
+  {
+    geography: "world_4region",
+    expectedTopFiveRowsOfOutput: [
+      ["alias", "geo", "name"],
+      ["Africa", "africa", "Africa"],
+      ["africa", "africa", "Africa"],
+      ["africa (total)", "americas", "The Americas"],
+      ["Africa (total)", "americas", "The Americas"]
+    ]
+  }
+  /* tslint:enable:object-literal-sort-keys */
+].forEach((testData, index) => {
+  test("testGmGeoLookupTable - " + index, testGmGeoLookupTable, testData);
+});

src/GM_GEO_LOOKUP_TABLE.ts

@@ -0,0 +1,57 @@
+import {
+  GeoAliasesAndSynonymsDataRow,
+  GeoAliasesAndSynonymsLookupTable,
+  GeoAliasesAndSynonymsWorksheetData,
+  geoAliasesAndSynonymsWorksheetDataToGeoLookupTable,
+  getGeoAliasesAndSynonymsWorksheetData
+} from "./gsheetsData/geoAliasesAndSynonyms";
+
+/**
+ * Inserts a table with Gapminder’s geo ids together with their aliases (all spellings we have seen before), including lower cased
+ * variants without diacritics and special characters to allow for somewhat fuzzy matching.
+ *
+ * To be used as the source range for VLOOKUP where the dataset is too large for GM_ID or GM_NAME to be used directly.
+ *
+ * @param geography Should be one of the sets listed in the gapminder geo ontology such as "countries_etc"
+ * @return A two-dimensional array containing the cell/column contents described above in the summary.
+ */
+export function GM_GEO_LOOKUP_TABLE(geography: string): string[][] {
+  if (!geography) {
+    geography = "countries_etc";
+  }
+  const geoAliasesAndSynonymsWorksheetData: GeoAliasesAndSynonymsWorksheetData = getGeoAliasesAndSynonymsWorksheetData(
+    geography
+  );
+  const fuzzyMatchLookupTable: GeoAliasesAndSynonymsLookupTable = geoAliasesAndSynonymsWorksheetDataToGeoLookupTable(
+    geoAliasesAndSynonymsWorksheetData,
+    null
+  );
+  const exactMatchLookupTable: GeoAliasesAndSynonymsLookupTable = geoAliasesAndSynonymsWorksheetDataToGeoLookupTable(
+    geoAliasesAndSynonymsWorksheetData,
+    lookupKey => lookupKey
+  );
+  const lookupTable = { ...fuzzyMatchLookupTable, ...exactMatchLookupTable };
+  const lookupKeys = Object.keys(lookupTable);
+  const lookupTableOutput = lookupKeys.map(lookupKey => {
+    const geoAliasesAndSynonymsDataRow: GeoAliasesAndSynonymsDataRow =
+      lookupTable[lookupKey];
+    return [
+      lookupKey,
+      geoAliasesAndSynonymsDataRow.geo,
+      geoAliasesAndSynonymsDataRow.name
+    ];
+  });
+  const sortedLookupTableOutput = lookupTableOutput.sort((a, b) => {
+    const geoIdComparison = a[1].localeCompare(b[1], "en", {
+      sensitivity: "accent"
+    });
+    const aliasComparison = a[0].localeCompare(b[0], "en", {
+      sensitivity: "accent"
+    });
+    if (geoIdComparison === 0) {
+      return aliasComparison;
+    }
+    return geoIdComparison;
+  });
+  return [["alias", "geo", "name"]].concat(sortedLookupTableOutput);
+}

src/gsheetsData/geoAliasesAndSynonyms.ts

@@ -9,7 +9,7 @@ import { ListGeoAliasesAndSynonyms } from "./types/listGeoAliasesAndSynonyms";
 /**
  * @hidden
  */
-interface GeoAliasesAndSynonymsDataRow {
+export interface GeoAliasesAndSynonymsDataRow {
   /* tslint:disable:object-literal-sort-keys */
   alias: string;
   geo: string;
@@ -27,32 +27,39 @@ interface MissingGeoAliasDataRow {
 /**
  * @hidden
  */
-interface GeoAliasesAndSynonymsWorksheetData {
+export interface GeoAliasesAndSynonymsWorksheetData {
   rows: GeoAliasesAndSynonymsDataRow[];
 }
 
 /**
  * @hidden
  */
-interface GeoAliasesAndSynonymsLookupTable {
+export interface GeoAliasesAndSynonymsLookupTable {
   [alias: string]: GeoAliasesAndSynonymsDataRow;
 }
 
 /**
  * @hidden
  */
 export function getGeoAliasesAndSynonymsLookupTable(geography) {
+  const data = getGeoAliasesAndSynonymsWorksheetData(geography);
+  return geoAliasesAndSynonymsWorksheetDataToGeoLookupTable(data, null);
+}
+
+/**
+ * @hidden
+ */
+export function getGeoAliasesAndSynonymsWorksheetData(geography) {
   if (!geoAliasesAndSynonymsDocWorksheetReferencesByGeopgraphy[geography]) {
     throw new Error(`Unknown Gapminder geography: "${geography}"`);
   }
   const worksheetDataResponse: ListGeoAliasesAndSynonyms.Response = fetchWorksheetData(
     geoAliasesAndSynonymsDocSpreadsheetId,
     geoAliasesAndSynonymsDocWorksheetReferencesByGeopgraphy[geography]
   );
-  const data = gsheetsDataApiFeedsListGeoAliasesAndSynonymsResponseToWorksheetData(
+  return gsheetsDataApiFeedsListGeoAliasesAndSynonymsResponseToWorksheetData(
     worksheetDataResponse
   );
-  return geoAliasesAndSynonymsWorksheetDataToGeoLookupTable(data);
 }
 
 /**
@@ -79,7 +86,9 @@ function gsheetsDataApiFeedsListGeoAliasesAndSynonymsResponseToWorksheetData(
  * By trimming the lookup keys, we allow slightly fuzzy matching, such as "Foo " == "foo" and "Fóo*" == "Foo"
  * @hidden
  */
-export function keyNormalizerForSlightlySmarterLookups(lookupKey) {
+export function keyNormalizerForSlightlySmarterLookups(
+  lookupKey: string
+): string {
   const trimmedLowerCasedWithoutDiacritics = removeDiacritics(
     lookupKey.trim().toLowerCase()
   );
@@ -89,16 +98,16 @@ export function keyNormalizerForSlightlySmarterLookups(lookupKey) {
 /**
  * @hidden
  */
-function geoAliasesAndSynonymsWorksheetDataToGeoLookupTable(
-  data: GeoAliasesAndSynonymsWorksheetData
+export function geoAliasesAndSynonymsWorksheetDataToGeoLookupTable(
+  data: GeoAliasesAndSynonymsWorksheetData,
+  normalizer: (lookupKey: string) => string
 ): GeoAliasesAndSynonymsLookupTable {
+  if (!normalizer) {
+    normalizer = keyNormalizerForSlightlySmarterLookups;
+  }
   return data.rows.reduce((lookupTableAccumulator, currentValue) => {
-    lookupTableAccumulator[
-      keyNormalizerForSlightlySmarterLookups(currentValue.geo)
-    ] = currentValue;
-    lookupTableAccumulator[
-      keyNormalizerForSlightlySmarterLookups(currentValue.alias)
-    ] = currentValue;
+    lookupTableAccumulator[normalizer(currentValue.geo)] = currentValue;
+    lookupTableAccumulator[normalizer(currentValue.alias)] = currentValue;
     return lookupTableAccumulator;
   }, {});
 }
@@ -110,6 +119,9 @@ export function matchColumnValuesUsingGeoAliasesAndSynonyms(
   columnValues,
   geography
 ) {
+  if (!geography) {
+    geography = "countries_etc";
+  }
   const lookupTable = getGeoAliasesAndSynonymsLookupTable(geography);
   return columnValues.map(
     (inputRow): GeoAliasesAndSynonymsDataRow | MissingGeoAliasDataRow => {

src/index.ts

@@ -5,6 +5,7 @@
 
 import { GM_AGGR } from "./GM_AGGR";
 import { GM_DATA } from "./GM_DATA";
+import { GM_GEO_LOOKUP_TABLE } from "./GM_GEO_LOOKUP_TABLE";
 import { GM_GROWTH } from "./GM_GROWTH";
 import { GM_ID } from "./GM_ID";
 import { GM_IMPORT } from "./GM_IMPORT";
@@ -96,6 +97,19 @@ import { menuRefreshDataDependencies } from "./menuRefreshDataDependencies";
   );
 };
 
+/**
+ * Inserts a table with Gapminder’s geo ids together with their aliases (all spellings we have seen before), including lower cased
+ * variants without diacritics and special characters to allow for somewhat fuzzy matching.
+ *
+ * To be used as the source range for VLOOKUP where the dataset is too large for GM_ID or GM_NAME to be used directly.
+ *
+ * @param {"countries_etc"} geography (Optional with default "countries_etc") Should be one of the sets listed in the gapminder geo ontology such as "countries_etc"
+ * @customfunction
+ */
+(global as any).GM_GEO_LOOKUP_TABLE = function(geography: string) {
+  return GM_GEO_LOOKUP_TABLE(geography);
+};
+
 /**
  * Inserts the growth per time unit of a common Gapminder concept column, including a header row, matched against the input table range.
  *