From ffbceda69d3d788df7020361d3d6b5f26ad9a999 Mon Sep 17 00:00:00 2001
From: Steve McConnel <stephen_mcconnel@sil.org>
Date: Thu, 21 Nov 2024 11:57:22 -0700
Subject: [PATCH] fix: allow nonASCII page numbers for book urls (BL-13882)

---
 src/bloom-player-controls.tsx |  8 ++++++--
 src/utilities/math.test.ts    | 18 +++++++++++++++++-
 src/utilities/mathUtils.ts    | 31 +++++++++++++++++++++++++++++++
 tsconfig.json                 |  2 +-
 4 files changed, 55 insertions(+), 4 deletions(-)

diff --git a/src/bloom-player-controls.tsx b/src/bloom-player-controls.tsx
index acc5813f..7e462ccf 100644
--- a/src/bloom-player-controls.tsx
+++ b/src/bloom-player-controls.tsx
@@ -32,7 +32,7 @@ import { withStyles, createTheme } from "@material-ui/core/styles";
 import DragBar from "@material-ui/core/Slider";
 import { bloomRed } from "./bloomPlayerTheme";
 import { setDurationOfPagesWithoutNarration } from "./narration";
-import { roundToNearestK } from "./utilities/mathUtils";
+import { roundToNearestK, normalizeDigits } from "./utilities/mathUtils";
 
 // This component is designed to wrap a BloomPlayer with some controls
 // for things like pausing audio and motion, hiding and showing
@@ -1055,6 +1055,7 @@ function getExtraButtons(): IExtraButton[] {
         return [];
     }
 }
+
 // a bit goofy...we need some way to get react called when this code is loaded into an HTML
 // document (as part of bloomPlayerControlBundle.js). When that module is loaded, any
 // not-in-a-class code gets called. So we arrange in bloom-player-root.ts to call this
@@ -1068,7 +1069,10 @@ export function InitBloomPlayerControls() {
         "motion",
     )! as autoPlayType;
     const startPageString = getQueryStringParamAndUnencode("start-page");
-    const startPage = startPageString ? parseInt(startPageString) : undefined;
+    const decimalStartPageString = normalizeDigits(startPageString);
+    const startPage = decimalStartPageString
+        ? parseInt(decimalStartPageString)
+        : undefined;
     const autoplayCountString =
         getQueryStringParamAndUnencode("autoplay-count");
     const autoplayCount = startPageString
diff --git a/src/utilities/math.test.ts b/src/utilities/math.test.ts
index c65758f4..e82bb5b5 100644
--- a/src/utilities/math.test.ts
+++ b/src/utilities/math.test.ts
@@ -1,4 +1,4 @@
-import { roundToNearestK } from "./mathUtils";
+import { roundToNearestK, normalizeDigits } from "./mathUtils";
 
 const roundToNearestEvenCases = [
     [0, 0],
@@ -31,3 +31,19 @@ test.each(roundToNearest4Cases)("roundToNearestK, k=4", (input, expected) => {
     const result = roundToNearestK(input, 4);
     expect(result).toEqual(expected);
 });
+
+test("normalizeDigits from mathUtils.ts", () => {
+    expect(normalizeDigits("1234567890")).toEqual("1234567890");
+    expect(normalizeDigits("١٢٣٤٥٦٧٨٩٠")).toEqual("1234567890");
+    expect(normalizeDigits("०१२३४५६७८९")).toEqual("0123456789");
+    expect(normalizeDigits("٦٧٨٩٠١٢٣٤٥")).toEqual("6789012345");
+    expect(normalizeDigits("१२३४५६७८९०١٢٣٤٥٦٧٨٩٠")).toEqual(
+        "12345678901234567890",
+    );
+    expect(normalizeDigits("١٢٣٤٥٦٧٨٩٠cover")).toEqual("1234567890cover");
+    expect(normalizeDigits("test")).toEqual("test");
+    expect(normalizeDigits("cover")).toBeUndefined(); // special case
+    expect(normalizeDigits("")).toBeUndefined();
+    expect(normalizeDigits(undefined)).toBeUndefined();
+    expect(normalizeDigits(null)).toBeUndefined();
+});
diff --git a/src/utilities/mathUtils.ts b/src/utilities/mathUtils.ts
index b1f95e0c..3fc5156e 100644
--- a/src/utilities/mathUtils.ts
+++ b/src/utilities/mathUtils.ts
@@ -8,3 +8,34 @@ export function roundToNearestEven(x: number): number {
 export function roundToNearestK(x: number, k: number): number {
     return Math.round(x / k) * k;
 }
+
+// This may not be a math function, strictly speaking, but it's at least
+// math adjacent.
+/**
+ * Convert all Unicode digits in a string to ASCII digits.
+ */
+// adapted from https://stackoverflow.com/questions/17024985/javascript-cant-convert-hindi-arabic-numbers-to-real-numeric-variables
+export function normalizeDigits(str): string | undefined {
+    if (!str || str === "cover") {
+        return undefined; // not a number we can parse
+    }
+    if (/^\d+$/.test(str)) {
+        return str; // entire string is ASCII decimal digits
+    }
+    try {
+        // find all characters which are DecimalNumber (property Nd), except for ASCII 0-9
+        return str.replace(/(?![0-9])\p{Nd}/gu, (g) => {
+            // all Nd blocks start at 0x...0 or end at 0x...F (and starts at 0x...6)
+            // if it starts at 0x...0, the ASCII decimal number is (i & 0xf)
+            // if it ends at 0x...F, the ASCII decimal number is (i & 0xf) - 6
+            // we recognize the 2 cases by testing if code | 0xf == 0x...F is still a decimal number
+            const code = g.charCodeAt(0);
+            return (
+                (code & 0xf) -
+                6 * (/\p{Nd}/u.test(String.fromCodePoint(code | 0xf)) ? 1 : 0)
+            );
+        });
+    } catch (e) {
+        return undefined;
+    }
+}
diff --git a/tsconfig.json b/tsconfig.json
index 8774b6a3..ea124d76 100644
--- a/tsconfig.json
+++ b/tsconfig.json
@@ -1,6 +1,6 @@
 {
     "compilerOptions": {
-        "target": "es5",
+        "target": "es6",
         "module": "commonjs",
         "sourceMap": true,
         "jsx": "react",