dravoice 0.1.1 → 0.1.3

package/src/v2/text-utils.js

@@ -1,123 +1,123 @@
-const WORD_RE = /[a-z][a-z0-9'-]*/gi;
-const ABBREVIATIONS = ["Dr.", "Mr.", "Mrs.", "Ms.", "Prof.", "Sr.", "Jr.", "e.g.", "i.e."];
-
-export const FUNCTION_WORDS = [
-  "a", "an", "and", "as", "at", "because", "but", "by", "for", "from",
-  "if", "in", "into", "is", "it", "its", "of", "on", "or", "that",
-  "the", "their", "then", "there", "this", "to", "was", "we", "with",
-  "you", "your", "i", "my", "not", "so", "before", "after", "while",
-];
-
-export const STOP_WORDS = new Set([
-  ...FUNCTION_WORDS,
-  "about", "also", "between", "can", "could", "each", "had", "has",
-  "have", "more", "one", "only", "our", "out", "over", "should",
-  "than", "they", "were", "when", "where", "which", "who", "will",
-  "would",
-]);
-
-export function normalizeText(text) {
-  return String(text ?? "").replace(/\s+/g, " ").trim().toLowerCase();
-}
-
-export function tokenizeWords(text) {
-  return (String(text ?? "").match(WORD_RE) ?? [])
-    .map((word) => word.toLowerCase().replace(/'s$/, ""));
-}
-
-export function contentWords(text) {
-  return tokenizeWords(text).filter((word) => word.length > 2 && !STOP_WORDS.has(word));
-}
-
-export function splitSentences(text) {
-  let protectedText = String(text ?? "");
-  const replacements = new Map();
-  const protect = (pattern, prefix) => {
-    protectedText = protectedText.replace(pattern, (match) => {
-      const key = `__${prefix}_${replacements.size}__`;
-      replacements.set(key, match);
-      return key;
-    });
-  };
-
-  for (const abbreviation of ABBREVIATIONS) {
-    protectedText = protectedText.replaceAll(abbreviation, abbreviation.replaceAll(".", "§"));
-  }
-  protect(/https?:\/\/\S+/g, "URL");
-  protect(/\b[\w/-]+\.(?:js|ts|py|md|mdx|txt|html|json|yml|yaml|toml)\b/g, "PATH");
-  protect(/\b\d+(?:\.\d+)+\b/g, "VERSION");
-
-  const parts = protectedText
-    .split(/(?<=[.!?])\s+(?=["'(\[]?[A-Z0-9])/)
-    .map((part) => part.trim())
-    .filter(Boolean);
-
-  return parts.map((part) => {
-    let restored = part.replaceAll("§", ".");
-    for (const [key, value] of replacements.entries()) {
-      restored = restored.replaceAll(key, value);
-    }
-    return restored;
-  });
-}
-
-export function characterNgrams(text, size = 3) {
-  const compact = normalizeText(text).replace(/[^a-z0-9]+/g, " ").trim();
-  const grams = [];
-  for (let index = 0; index <= compact.length - size; index += 1) {
-    const gram = compact.slice(index, index + size);
-    if (!gram.includes(" ")) {
-      grams.push(gram);
-    }
-  }
-  return grams;
-}
-
-export function distribution(values) {
-  if (!values.length) {
-    return { count: 0, min: 0, max: 0, mean: 0, median: 0, p25: 0, p75: 0 };
-  }
-  const sorted = [...values].sort((a, b) => a - b);
-  return {
-    count: sorted.length,
-    min: sorted[0],
-    max: sorted[sorted.length - 1],
-    mean: round(sorted.reduce((sum, value) => sum + value, 0) / sorted.length, 2),
-    median: percentile(sorted, 0.5),
-    p25: percentile(sorted, 0.25),
-    p75: percentile(sorted, 0.75),
-  };
-}
-
-export function topItems(values, limit = 10) {
-  const counts = new Map();
-  for (const value of values.filter(Boolean)) {
-    counts.set(value, (counts.get(value) ?? 0) + 1);
-  }
-  return Array.from(counts.entries())
-    .sort((left, right) => right[1] - left[1] || left[0].localeCompare(right[0]))
-    .slice(0, limit)
-    .map(([value, count]) => ({ value, count }));
-}
-
-export function rate(count, total, digits = 3) {
-  return round(count / Math.max(1, total), digits);
-}
-
-export function round(value, digits = 0) {
-  const multiplier = 10 ** digits;
-  const rounded = Math.floor(value * multiplier + 0.5) / multiplier;
-  return digits === 0 ? Math.trunc(rounded) : rounded;
-}
-
-export function clampScore(value) {
-  return Math.max(0, Math.min(100, Math.round(value)));
-}
-
-function percentile(sortedValues, ratio) {
-  const index = Math.min(
-    sortedValues.length - 1,
-    Math.max(0, Math.round((sortedValues.length - 1) * ratio)),
-  );
-  return sortedValues[index];
-}
+const WORD_RE = /[a-z][a-z0-9'-]*/gi;
+const ABBREVIATIONS = ["Dr.", "Mr.", "Mrs.", "Ms.", "Prof.", "Sr.", "Jr.", "e.g.", "i.e."];
+
+export const FUNCTION_WORDS = [
+  "a", "an", "and", "as", "at", "because", "but", "by", "for", "from",
+  "if", "in", "into", "is", "it", "its", "of", "on", "or", "that",
+  "the", "their", "then", "there", "this", "to", "was", "we", "with",
+  "you", "your", "i", "my", "not", "so", "before", "after", "while",
+];
+
+export const STOP_WORDS = new Set([
+  ...FUNCTION_WORDS,
+  "about", "also", "between", "can", "could", "each", "had", "has",
+  "have", "more", "one", "only", "our", "out", "over", "should",
+  "than", "they", "were", "when", "where", "which", "who", "will",
+  "would",
+]);
+
+export function normalizeText(text) {
+  return String(text ?? "").replace(/\s+/g, " ").trim().toLowerCase();
+}
+
+export function tokenizeWords(text) {
+  return (String(text ?? "").match(WORD_RE) ?? [])
+    .map((word) => word.toLowerCase().replace(/'s$/, ""));
+}
+
+export function contentWords(text) {
+  return tokenizeWords(text).filter((word) => word.length > 2 && !STOP_WORDS.has(word));
+}
+
+export function splitSentences(text) {
+  let protectedText = String(text ?? "");
+  const replacements = new Map();
+  const protect = (pattern, prefix) => {
+    protectedText = protectedText.replace(pattern, (match) => {
+      const key = `__${prefix}_${replacements.size}__`;
+      replacements.set(key, match);
+      return key;
+    });
+  };
+
+  for (const abbreviation of ABBREVIATIONS) {
+    protectedText = protectedText.replaceAll(abbreviation, abbreviation.replaceAll(".", "§"));
+  }
+  protect(/https?:\/\/\S+/g, "URL");
+  protect(/\b[\w/-]+\.(?:js|ts|py|md|mdx|txt|html|json|yml|yaml|toml)\b/g, "PATH");
+  protect(/\b\d+(?:\.\d+)+\b/g, "VERSION");
+
+  const parts = protectedText
+    .split(/(?<=[.!?])\s+(?=["'(\[]?[A-Z0-9])/)
+    .map((part) => part.trim())
+    .filter(Boolean);
+
+  return parts.map((part) => {
+    let restored = part.replaceAll("§", ".");
+    for (const [key, value] of replacements.entries()) {
+      restored = restored.replaceAll(key, value);
+    }
+    return restored;
+  });
+}
+
+export function characterNgrams(text, size = 3) {
+  const compact = normalizeText(text).replace(/[^a-z0-9]+/g, " ").trim();
+  const grams = [];
+  for (let index = 0; index <= compact.length - size; index += 1) {
+    const gram = compact.slice(index, index + size);
+    if (!gram.includes(" ")) {
+      grams.push(gram);
+    }
+  }
+  return grams;
+}
+
+export function distribution(values) {
+  if (!values.length) {
+    return { count: 0, min: 0, max: 0, mean: 0, median: 0, p25: 0, p75: 0 };
+  }
+  const sorted = [...values].sort((a, b) => a - b);
+  return {
+    count: sorted.length,
+    min: sorted[0],
+    max: sorted[sorted.length - 1],
+    mean: round(sorted.reduce((sum, value) => sum + value, 0) / sorted.length, 2),
+    median: percentile(sorted, 0.5),
+    p25: percentile(sorted, 0.25),
+    p75: percentile(sorted, 0.75),
+  };
+}
+
+export function topItems(values, limit = 10) {
+  const counts = new Map();
+  for (const value of values.filter(Boolean)) {
+    counts.set(value, (counts.get(value) ?? 0) + 1);
+  }
+  return Array.from(counts.entries())
+    .sort((left, right) => right[1] - left[1] || left[0].localeCompare(right[0]))
+    .slice(0, limit)
+    .map(([value, count]) => ({ value, count }));
+}
+
+export function rate(count, total, digits = 3) {
+  return round(count / Math.max(1, total), digits);
+}
+
+export function round(value, digits = 0) {
+  const multiplier = 10 ** digits;
+  const rounded = Math.floor(value * multiplier + 0.5) / multiplier;
+  return digits === 0 ? Math.trunc(rounded) : rounded;
+}
+
+export function clampScore(value) {
+  return Math.max(0, Math.min(100, Math.round(value)));
+}
+
+function percentile(sortedValues, ratio) {
+  const index = Math.min(
+    sortedValues.length - 1,
+    Math.max(0, Math.round((sortedValues.length - 1) * ratio)),
+  );
+  return sortedValues[index];
+}