Refactor get-words.ts to streamline data parsing

The changes in get-words.ts attempt to make the data parsing process more efficient by consolidating two separate actions into one. We specifically eliminated the separate 'word' parsing operation and directly supplied the parsed 'jsonObject' to 'zodParse'. Moreover, this refactor ensures all yielded words are in lower case.

apps/learnbefore-bff/src/get-words.ts

@@ -28,14 +28,13 @@ export async function* getWords(
         const jsonObject = data.slice(startIndex, endIndex + 1)
         data = data.slice(endIndex + 1)
         try {
-          const word = parse(jsonObject)
-          const zodParse = wordSchema.safeParse(word)
+          const zodParse = wordSchema.safeParse(parse(jsonObject))
           if (zodParse.success) {
-            yield word
+            yield { ...zodParse.data, word: zodParse.data.word.toLowerCase() }
           } else {
             console.error(`Zod validation error.`)
             console.error("Word validation error. " + zodParse.error.message)
-            console.error(`Parsed content: `, word)
+            console.error(`Parsed content: `, parse(jsonObject))
           }
         } catch (err) {
           console.error("Error while parsing JSON:", err)