redashify 0.0.0 → 0.2.0

package/LICENSE

@@ -0,0 +1,190 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to the Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by the Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding any notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   Copyright 2025 Haihang Wang
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

package/README.md

@@ -0,0 +1,169 @@
+# redashify
+
+[![npm](https://img.shields.io/npm/v/redashify)](https://www.npmjs.com/package/redashify)
+[![license](https://img.shields.io/npm/l/redashify)](https://github.com/hangingahaw/redashify/blob/main/LICENSE)
+
+Context-aware dash correction powered by LLMs.
+
+Converts hyphens (`-`), double-hyphens (`--`), and triple-hyphens (`---`) to the typographically correct character — hyphen, en dash (`–`), or em dash (`—`) — based on surrounding context.
+
+Unlike regex, redashify uses an LLM to determine the correct dash because the rules are too context-dependent for pattern matching alone. Only the dash contexts (not the full document) are sent to the LLM, making it token-efficient and privacy-conscious.
+
+## Install
+
+```sh
+npm install redashify
+```
+
+## Quick start
+
+```ts
+import { redashify } from 'redashify'
+
+const result = await redashify(
+  'The court held--in a 5-4 decision--that pages 10-20 applied.',
+  { apiKey: process.env.OPENAI_API_KEY, provider: 'openai' }
+)
+
+result.text
+// → 'The court held—in a 5-4 decision—that pages 10–20 applied.'
+
+result.corrections
+// → [{ position: 14, original: '--', replacement: '—', context: '...' }, ...]
+
+result.unchanged
+// → false
+```
+
+## Providers
+
+Built-in support for any OpenAI-compatible API, plus a native Anthropic adapter.
+
+| Provider | Default model | Notes |
+|---|---|---|
+| `openai` | `gpt-4o-mini` | |
+| `anthropic` | `claude-haiku-4-5-20251001` | Native adapter (different API format) |
+| `gemini` | `gemini-2.0-flash` | OpenAI-compatible endpoint |
+| `groq` | `llama-3.3-70b-versatile` | |
+| `together` | `meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo` | |
+| `mistral` | `mistral-small-latest` | |
+| `xai` | `grok-3-mini-fast` | |
+| `deepseek` | `deepseek-chat` | |
+| `openrouter` | *(none — must specify `model`)* | |
+
+Override the default model:
+
+```ts
+const result = await redashify(text, {
+  apiKey: process.env.OPENAI_API_KEY,
+  provider: 'openai',
+  model: 'gpt-4o',
+})
+```
+
+For unlisted providers that support the OpenAI chat completions format:
+
+```ts
+const result = await redashify(text, {
+  apiKey: '...',
+  baseURL: 'https://my-provider.com/v1',
+  model: 'my-model',
+})
+```
+
+### Custom LLM function
+
+Bypass the built-in client entirely:
+
+```ts
+const result = await redashify(text, {
+  llm: async (messages) => {
+    const res = await myLlmCall(messages)
+    return res.text
+  },
+})
+```
+
+## Options
+
+| Option | Type | Default | Description |
+|---|---|---|---|
+| `apiKey` | `string` | — | API key for the LLM provider |
+| `provider` | `Provider` | — | Provider name (maps to base URL + default model) |
+| `model` | `string` | *(per provider)* | Model name. Required if no provider default. |
+| `baseURL` | `string` | — | Custom endpoint URL. Overrides provider mapping. |
+| `llm` | `(messages) => Promise<string>` | — | Custom LLM function. Overrides apiKey/provider/model. |
+| `rules` | `string` | `“”` | Custom rules prepended to the system prompt |
+| `contextSize` | `number` | `50` | Characters of context on each side of a dash |
+| `batchSize` | `number` | `50` | Maximum dashes per LLM call |
+
+You must provide either `apiKey` (with `provider` or `model`) or `llm`.
+
+## Result
+
+```ts
+interface RedashifyResult {
+  text: string          // The corrected text
+  corrections: Array<{  // Only dashes that were changed
+    position: number    // Index in original text
+    original: string    // What was there (e.g. "--")
+    replacement: string // What it became (e.g. "—")
+    context: string     // Surrounding snippet for audit
+  }>
+  unchanged: boolean    // true if nothing was modified
+}
+```
+
+**No dashes in text:** LLM is not called. Returns immediately with `unchanged: true`.
+
+**All dashes already correct:** LLM is called (correctness can’t be pre-judged), but `corrections` is empty and `unchanged` is `true`.
+
+## Custom rules
+
+Pass domain-specific rules via the `rules` option. Works with [lexstyle](https://github.com/hangingahaw/lexstyle) for structured rule management:
+
+```ts
+import { rules, serialize } from 'lexstyle'
+import { redashify } from 'redashify'
+
+const result = await redashify(text, {
+  apiKey: process.env.OPENAI_API_KEY,
+  provider: 'openai',
+  rules: serialize(rules, 'dashes'),
+})
+```
+
+Or pass rules as a plain string:
+
+```ts
+const result = await redashify(text, {
+  apiKey: '...',
+  provider: 'openai',
+  rules: `Use en dashes for page ranges (e.g. pp. 45–67).
+Use em dashes for parenthetical asides with no spaces.
+Keep hyphens in compound modifiers and vote tallies (5-4).`,
+})
+```
+
+## Design decisions
+
+**LLM over regex.** Dash correction depends on semantic context — is “10-20” a range (en dash) or a compound modifier (hyphen)? Is “--” an em dash or a typo? Regex can’t answer these questions. An LLM can, given a few words of surrounding context.
+
+**Privacy by design.** Only short context windows around each dash are sent to the LLM — never the full document. A 10,000-word document with 5 dashes sends ~5 small context snippets, regardless of document length.
+
+**Batch validation.** Each batch response is validated against its expected IDs before merging. Cross-batch ID leakage or missing corrections are caught immediately, not silently swallowed.
+
+**Robust response parsing.** LLM output is parsed via strict JSON first, with a hardened bracket-extraction fallback that skips stray brackets in preamble text. Every correction is validated for known dash characters and duplicate IDs.
+
+## Development
+
+```sh
+npm install
+npm test
+npm run typecheck
+npm run build     # ESM + CJS + .d.ts
+```
+
+## License
+
+Apache-2.0

package/dist/index.cjs

@@ -0,0 +1,244 @@
+"use strict";
+var __defProp = Object.defineProperty;
+var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
+var __getOwnPropNames = Object.getOwnPropertyNames;
+var __hasOwnProp = Object.prototype.hasOwnProperty;
+var __export = (target, all) => {
+  for (var name in all)
+    __defProp(target, name, { get: all[name], enumerable: true });
+};
+var __copyProps = (to, from, except, desc) => {
+  if (from && typeof from === "object" || typeof from === "function") {
+    for (let key of __getOwnPropNames(from))
+      if (!__hasOwnProp.call(to, key) && key !== except)
+        __defProp(to, key, { get: () => from[key], enumerable: !(desc = __getOwnPropDesc(from, key)) || desc.enumerable });
+  }
+  return to;
+};
+var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: true }), mod);
+
+// src/index.ts
+var index_exports = {};
+__export(index_exports, {
+  VALID_DASHES: () => VALID_DASHES,
+  buildMessages: () => buildMessages,
+  extractDashes: () => extractDashes,
+  redashify: () => redashify
+});
+module.exports = __toCommonJS(index_exports);
+
+// src/core.ts
+var import_llm_client = require("@lexstyle/llm-client");
+
+// src/extractor.ts
+var DASH_RE = /---+|--|\u2014|\u2013|-/g;
+function extractDashes(text, contextSize = 50) {
+  const results = [];
+  let match;
+  let id = 0;
+  DASH_RE.lastIndex = 0;
+  while ((match = DASH_RE.exec(text)) !== null) {
+    const start = match.index;
+    const end = start + match[0].length;
+    const bStart = Math.max(0, start - contextSize);
+    let before = text.slice(bStart, start);
+    if (bStart > 0) {
+      const idx = before.search(/\s/);
+      if (idx !== -1) {
+        before = before.slice(idx).trimStart();
+      }
+    }
+    const aEnd = Math.min(text.length, end + contextSize);
+    let after = text.slice(end, aEnd);
+    if (aEnd < text.length) {
+      const idx = after.search(/\s[^\s]*$/);
+      if (idx > 0) {
+        after = after.slice(0, idx);
+      }
+    }
+    results.push({ id: id++, original: match[0], before, after, start, end });
+  }
+  return results;
+}
+
+// src/prompt.ts
+function buildMessages(contexts, rules) {
+  const ruleBlock = rules ? `
+${rules}
+` : "";
+  const system = `You are a typography expert. Your task is to determine the correct dash character for each occurrence.
+${ruleBlock}
+
+For each dash below, return the correct Unicode character:
+  - (U+002D) hyphen
+  \u2013 (U+2013) en dash
+  \u2014 (U+2014) em dash
+
+IMPORTANT: You must return exactly one entry for every id provided. Do not skip any.
+Respond with ONLY a JSON array. No explanation, no markdown fences.
+Format: [{"id":0,"dash":"-"},{"id":1,"dash":"\u2013"}]`;
+  const user = contexts.map((ctx) => `[${ctx.id}] \u201C${ctx.before}\u201D [${ctx.original}] \u201C${ctx.after}\u201D`).join("\n");
+  return [
+    { role: "system", content: system },
+    { role: "user", content: user }
+  ];
+}
+
+// src/types.ts
+var VALID_DASHES = Object.freeze(["-", "\u2013", "\u2014"]);
+var VALID_DASH_SET = new Set(VALID_DASHES);
+function isValidDash(s) {
+  return VALID_DASH_SET.has(s);
+}
+
+// src/replacer.ts
+function parseResponse(response) {
+  let cleaned = response.trim();
+  cleaned = cleaned.replace(/^```(?:json)?\s*\n?/i, "").replace(/\n?```\s*$/, "");
+  cleaned = cleaned.trim();
+  let parsed;
+  try {
+    const strict = JSON.parse(cleaned);
+    if (Array.isArray(strict)) {
+      parsed = strict;
+    }
+  } catch {
+  }
+  if (parsed === void 0) {
+    const end = cleaned.lastIndexOf("]");
+    if (end === -1) {
+      throw new Error(`Invalid LLM response: no JSON array found. Response: ${cleaned.slice(0, 200)}`);
+    }
+    for (let i = 0; i < end; i++) {
+      if (cleaned[i] !== "[") continue;
+      try {
+        const candidate = JSON.parse(cleaned.slice(i, end + 1));
+        if (Array.isArray(candidate)) {
+          parsed = candidate;
+          break;
+        }
+      } catch {
+        continue;
+      }
+    }
+  }
+  if (!Array.isArray(parsed)) {
+    throw new Error(
+      parsed === void 0 ? `Invalid LLM response: no JSON array found. Response: ${cleaned.slice(0, 200)}` : "LLM response is not a JSON array"
+    );
+  }
+  const seenIds = /* @__PURE__ */ new Set();
+  return parsed.map((item, idx) => {
+    const rec = item;
+    if (typeof rec !== "object" || rec === null || typeof rec.id !== "number" || typeof rec.dash !== "string") {
+      throw new Error(`Invalid correction at index ${idx}: ${JSON.stringify(item)}`);
+    }
+    const correction = { id: rec.id, dash: rec.dash };
+    if (!isValidDash(correction.dash)) {
+      throw new Error(
+        `Invalid dash character at index ${idx} (id ${correction.id}): got ${JSON.stringify(correction.dash)}, expected "-", "\\u2013", or "\\u2014"`
+      );
+    }
+    if (seenIds.has(correction.id)) {
+      throw new Error(`Duplicate correction id ${correction.id} at index ${idx}`);
+    }
+    seenIds.add(correction.id);
+    return { id: correction.id, dash: correction.dash };
+  });
+}
+function applyCorrections(text, contexts, corrections) {
+  const correctionMap = /* @__PURE__ */ new Map();
+  for (const c of corrections) {
+    correctionMap.set(c.id, c.dash);
+  }
+  const contextIds = new Set(contexts.map((c) => c.id));
+  for (const id of contextIds) {
+    if (!correctionMap.has(id)) {
+      throw new Error(`Missing correction for dash id ${id}`);
+    }
+  }
+  for (const id of correctionMap.keys()) {
+    if (!contextIds.has(id)) {
+      throw new Error(`Unknown correction id ${id}: no matching dash context`);
+    }
+  }
+  const changes = [];
+  for (const ctx of contexts) {
+    const replacement = correctionMap.get(ctx.id);
+    if (replacement !== ctx.original) {
+      changes.push({ context: ctx, replacement });
+    }
+  }
+  changes.sort((a, b) => b.context.start - a.context.start);
+  let result = text;
+  const appliedCorrections = [];
+  for (const { context: ctx, replacement } of changes) {
+    const snippetBefore = ctx.before.slice(-20);
+    const snippetAfter = ctx.after.slice(0, 20);
+    const contextSnippet = `${snippetBefore}[${ctx.original}\u2192${replacement}]${snippetAfter}`;
+    result = result.slice(0, ctx.start) + replacement + result.slice(ctx.end);
+    appliedCorrections.push({
+      position: ctx.start,
+      original: ctx.original,
+      replacement,
+      context: contextSnippet
+    });
+  }
+  appliedCorrections.reverse();
+  return { text: result, appliedCorrections };
+}
+
+// src/core.ts
+function requireInt(value, name, min) {
+  if (!Number.isFinite(value) || value < min || Math.floor(value) !== value) {
+    throw new Error(`Invalid ${name}: ${value}. Must be a finite integer >= ${min}.`);
+  }
+}
+function validateBatchIds(corrections, expectedIds) {
+  for (const { id } of corrections) {
+    if (!expectedIds.has(id)) {
+      throw new Error(`LLM returned unexpected id ${id} (not in this batch)`);
+    }
+  }
+  for (const id of expectedIds) {
+    if (!corrections.some((c) => c.id === id)) {
+      throw new Error(`LLM missing correction for id ${id} in batch`);
+    }
+  }
+}
+async function redashify(text, options) {
+  if (!options || typeof options !== "object") {
+    throw new Error("redashify requires an options object with `apiKey` + `model`, `apiKey` + `provider`, or `llm`");
+  }
+  const batchSize = options.batchSize ?? 50;
+  requireInt(batchSize, "batchSize", 1);
+  const contextSize = options.contextSize ?? 50;
+  requireInt(contextSize, "contextSize", 0);
+  const llmFn = (0, import_llm_client.resolveLlm)(options, "redashify");
+  const contexts = extractDashes(text, contextSize);
+  if (contexts.length === 0) {
+    return { text, corrections: [], unchanged: true };
+  }
+  const allCorrections = [];
+  for (let i = 0; i < contexts.length; i += batchSize) {
+    const batch = contexts.slice(i, i + batchSize);
+    const messages = buildMessages(batch, options.rules);
+    const response = await llmFn(messages);
+    const corrections = parseResponse(response);
+    validateBatchIds(corrections, new Set(batch.map((ctx) => ctx.id)));
+    allCorrections.push(...corrections);
+  }
+  const { text: correctedText, appliedCorrections } = applyCorrections(text, contexts, allCorrections);
+  return {
+    text: correctedText,
+    corrections: appliedCorrections,
+    unchanged: appliedCorrections.length === 0
+  };
+}
+// Annotate the CommonJS export names for ESM import in node:
+0 && (module.exports = {
+  VALID_DASHES,
+  buildMessages,
+  extractDashes,
+  redashify
+});

package/dist/index.d.cts

@@ -0,0 +1,82 @@
+import { LlmOptions, Message } from '@lexstyle/llm-client';
+export { Message, Provider } from '@lexstyle/llm-client';
+
+/** The valid dash characters, exported for documentation/reference */
+declare const VALID_DASHES: readonly ["-", "–", "—"];
+/** A single valid dash character: hyphen, en dash, or em dash */
+type DashChar = (typeof VALID_DASHES)[number];
+/** Result of applying corrections to text */
+interface ApplyResult {
+    text: string;
+    appliedCorrections: Correction[];
+}
+/** A dash occurrence extracted from the input text */
+interface DashContext {
+    /** Unique identifier for this dash occurrence */
+    id: number;
+    /** The original dash string found (e.g. "-", "--", "---", "\u2013", "\u2014") */
+    original: string;
+    /** Text before the dash for context */
+    before: string;
+    /** Text after the dash for context */
+    after: string;
+    /** Start index in the original text */
+    start: number;
+    /** End index in the original text (exclusive) */
+    end: number;
+}
+/** A single correction applied to the text, for audit/review */
+interface Correction {
+    /** Position in original text */
+    position: number;
+    /** The original dash string */
+    original: string;
+    /** The replacement dash character */
+    replacement: DashChar;
+    /** Surrounding context snippet */
+    context: string;
+}
+/** Result returned by redashify */
+interface RedashifyResult {
+    /** The corrected text */
+    text: string;
+    /** List of corrections that were applied */
+    corrections: Correction[];
+    /** True if no changes were needed */
+    unchanged: boolean;
+}
+/** Options for redashify */
+interface RedashifyOptions extends LlmOptions {
+    /** Characters of context on each side of a dash (default: 50) */
+    contextSize?: number;
+    /** Custom rules to prepend to the system prompt */
+    rules?: string;
+    /** Maximum dashes per LLM call (default: 50). Must be >= 1. */
+    batchSize?: number;
+}
+
+/**
+ * Correct dashes in text using an LLM.
+ *
+ * Architecture: extract dashes -> batch LLM calls -> replace.
+ * Only the dash contexts (not the full document) are sent to the LLM.
+ */
+declare function redashify(text: string, options?: RedashifyOptions): Promise<RedashifyResult>;
+
+/**
+ * Extract all dashes from text with surrounding context.
+ *
+ * Returns a DashContext for every dash occurrence, giving the LLM
+ * enough surrounding text to determine the correct replacement.
+ */
+declare function extractDashes(text: string, contextSize?: number): DashContext[];
+
+/**
+ * Build the messages array for a single LLM call.
+ *
+ * System prompt contains the rules (sent once).
+ * User prompt contains only the extracted dash contexts (compact).
+ */
+declare function buildMessages(contexts: readonly DashContext[], rules?: string): Message[];
+
+export { type ApplyResult, type Correction, type DashChar, type DashContext, type RedashifyOptions, type RedashifyResult, VALID_DASHES, buildMessages, extractDashes, redashify };

package/dist/index.d.ts

@@ -0,0 +1,82 @@
+import { LlmOptions, Message } from '@lexstyle/llm-client';
+export { Message, Provider } from '@lexstyle/llm-client';
+
+/** The valid dash characters, exported for documentation/reference */
+declare const VALID_DASHES: readonly ["-", "–", "—"];
+/** A single valid dash character: hyphen, en dash, or em dash */
+type DashChar = (typeof VALID_DASHES)[number];
+/** Result of applying corrections to text */
+interface ApplyResult {
+    text: string;
+    appliedCorrections: Correction[];
+}
+/** A dash occurrence extracted from the input text */
+interface DashContext {
+    /** Unique identifier for this dash occurrence */
+    id: number;
+    /** The original dash string found (e.g. "-", "--", "---", "\u2013", "\u2014") */
+    original: string;
+    /** Text before the dash for context */
+    before: string;
+    /** Text after the dash for context */
+    after: string;
+    /** Start index in the original text */
+    start: number;
+    /** End index in the original text (exclusive) */
+    end: number;
+}
+/** A single correction applied to the text, for audit/review */
+interface Correction {
+    /** Position in original text */
+    position: number;
+    /** The original dash string */
+    original: string;
+    /** The replacement dash character */
+    replacement: DashChar;
+    /** Surrounding context snippet */
+    context: string;
+}
+/** Result returned by redashify */
+interface RedashifyResult {
+    /** The corrected text */
+    text: string;
+    /** List of corrections that were applied */
+    corrections: Correction[];
+    /** True if no changes were needed */
+    unchanged: boolean;
+}
+/** Options for redashify */
+interface RedashifyOptions extends LlmOptions {
+    /** Characters of context on each side of a dash (default: 50) */
+    contextSize?: number;
+    /** Custom rules to prepend to the system prompt */
+    rules?: string;
+    /** Maximum dashes per LLM call (default: 50). Must be >= 1. */
+    batchSize?: number;
+}
+
+/**
+ * Correct dashes in text using an LLM.
+ *
+ * Architecture: extract dashes -> batch LLM calls -> replace.
+ * Only the dash contexts (not the full document) are sent to the LLM.
+ */
+declare function redashify(text: string, options?: RedashifyOptions): Promise<RedashifyResult>;
+
+/**
+ * Extract all dashes from text with surrounding context.
+ *
+ * Returns a DashContext for every dash occurrence, giving the LLM
+ * enough surrounding text to determine the correct replacement.
+ */
+declare function extractDashes(text: string, contextSize?: number): DashContext[];
+
+/**
+ * Build the messages array for a single LLM call.
+ *
+ * System prompt contains the rules (sent once).
+ * User prompt contains only the extracted dash contexts (compact).
+ */
+declare function buildMessages(contexts: readonly DashContext[], rules?: string): Message[];
+
+export { type ApplyResult, type Correction, type DashChar, type DashContext, type RedashifyOptions, type RedashifyResult, VALID_DASHES, buildMessages, extractDashes, redashify };

package/dist/index.js

@@ -0,0 +1,214 @@
+// src/core.ts
+import { resolveLlm } from "@lexstyle/llm-client";
+
+// src/extractor.ts
+var DASH_RE = /---+|--|\u2014|\u2013|-/g;
+function extractDashes(text, contextSize = 50) {
+  const results = [];
+  let match;
+  let id = 0;
+  DASH_RE.lastIndex = 0;
+  while ((match = DASH_RE.exec(text)) !== null) {
+    const start = match.index;
+    const end = start + match[0].length;
+    const bStart = Math.max(0, start - contextSize);
+    let before = text.slice(bStart, start);
+    if (bStart > 0) {
+      const idx = before.search(/\s/);
+      if (idx !== -1) {
+        before = before.slice(idx).trimStart();
+      }
+    }
+    const aEnd = Math.min(text.length, end + contextSize);
+    let after = text.slice(end, aEnd);
+    if (aEnd < text.length) {
+      const idx = after.search(/\s[^\s]*$/);
+      if (idx > 0) {
+        after = after.slice(0, idx);
+      }
+    }
+    results.push({ id: id++, original: match[0], before, after, start, end });
+  }
+  return results;
+}
+
+// src/prompt.ts
+function buildMessages(contexts, rules) {
+  const ruleBlock = rules ? `
+${rules}
+` : "";
+  const system = `You are a typography expert. Your task is to determine the correct dash character for each occurrence.
+${ruleBlock}
+
+For each dash below, return the correct Unicode character:
+  - (U+002D) hyphen
+  \u2013 (U+2013) en dash
+  \u2014 (U+2014) em dash
+
+IMPORTANT: You must return exactly one entry for every id provided. Do not skip any.
+Respond with ONLY a JSON array. No explanation, no markdown fences.
+Format: [{"id":0,"dash":"-"},{"id":1,"dash":"\u2013"}]`;
+  const user = contexts.map((ctx) => `[${ctx.id}] \u201C${ctx.before}\u201D [${ctx.original}] \u201C${ctx.after}\u201D`).join("\n");
+  return [
+    { role: "system", content: system },
+    { role: "user", content: user }
+  ];
+}
+
+// src/types.ts
+var VALID_DASHES = Object.freeze(["-", "\u2013", "\u2014"]);
+var VALID_DASH_SET = new Set(VALID_DASHES);
+function isValidDash(s) {
+  return VALID_DASH_SET.has(s);
+}
+
+// src/replacer.ts
+function parseResponse(response) {
+  let cleaned = response.trim();
+  cleaned = cleaned.replace(/^```(?:json)?\s*\n?/i, "").replace(/\n?```\s*$/, "");
+  cleaned = cleaned.trim();
+  let parsed;
+  try {
+    const strict = JSON.parse(cleaned);
+    if (Array.isArray(strict)) {
+      parsed = strict;
+    }
+  } catch {
+  }
+  if (parsed === void 0) {
+    const end = cleaned.lastIndexOf("]");
+    if (end === -1) {
+      throw new Error(`Invalid LLM response: no JSON array found. Response: ${cleaned.slice(0, 200)}`);
+    }
+    for (let i = 0; i < end; i++) {
+      if (cleaned[i] !== "[") continue;
+      try {
+        const candidate = JSON.parse(cleaned.slice(i, end + 1));
+        if (Array.isArray(candidate)) {
+          parsed = candidate;
+          break;
+        }
+      } catch {
+        continue;
+      }
+    }
+  }
+  if (!Array.isArray(parsed)) {
+    throw new Error(
+      parsed === void 0 ? `Invalid LLM response: no JSON array found. Response: ${cleaned.slice(0, 200)}` : "LLM response is not a JSON array"
+    );
+  }
+  const seenIds = /* @__PURE__ */ new Set();
+  return parsed.map((item, idx) => {
+    const rec = item;
+    if (typeof rec !== "object" || rec === null || typeof rec.id !== "number" || typeof rec.dash !== "string") {
+      throw new Error(`Invalid correction at index ${idx}: ${JSON.stringify(item)}`);
+    }
+    const correction = { id: rec.id, dash: rec.dash };
+    if (!isValidDash(correction.dash)) {
+      throw new Error(
+        `Invalid dash character at index ${idx} (id ${correction.id}): got ${JSON.stringify(correction.dash)}, expected "-", "\\u2013", or "\\u2014"`
+      );
+    }
+    if (seenIds.has(correction.id)) {
+      throw new Error(`Duplicate correction id ${correction.id} at index ${idx}`);
+    }
+    seenIds.add(correction.id);
+    return { id: correction.id, dash: correction.dash };
+  });
+}
+function applyCorrections(text, contexts, corrections) {
+  const correctionMap = /* @__PURE__ */ new Map();
+  for (const c of corrections) {
+    correctionMap.set(c.id, c.dash);
+  }
+  const contextIds = new Set(contexts.map((c) => c.id));
+  for (const id of contextIds) {
+    if (!correctionMap.has(id)) {
+      throw new Error(`Missing correction for dash id ${id}`);
+    }
+  }
+  for (const id of correctionMap.keys()) {
+    if (!contextIds.has(id)) {
+      throw new Error(`Unknown correction id ${id}: no matching dash context`);
+    }
+  }
+  const changes = [];
+  for (const ctx of contexts) {
+    const replacement = correctionMap.get(ctx.id);
+    if (replacement !== ctx.original) {
+      changes.push({ context: ctx, replacement });
+    }
+  }
+  changes.sort((a, b) => b.context.start - a.context.start);
+  let result = text;
+  const appliedCorrections = [];
+  for (const { context: ctx, replacement } of changes) {
+    const snippetBefore = ctx.before.slice(-20);
+    const snippetAfter = ctx.after.slice(0, 20);
+    const contextSnippet = `${snippetBefore}[${ctx.original}\u2192${replacement}]${snippetAfter}`;
+    result = result.slice(0, ctx.start) + replacement + result.slice(ctx.end);
+    appliedCorrections.push({
+      position: ctx.start,
+      original: ctx.original,
+      replacement,
+      context: contextSnippet
+    });
+  }
+  appliedCorrections.reverse();
+  return { text: result, appliedCorrections };
+}
+
+// src/core.ts
+function requireInt(value, name, min) {
+  if (!Number.isFinite(value) || value < min || Math.floor(value) !== value) {
+    throw new Error(`Invalid ${name}: ${value}. Must be a finite integer >= ${min}.`);
+  }
+}
+function validateBatchIds(corrections, expectedIds) {
+  for (const { id } of corrections) {
+    if (!expectedIds.has(id)) {
+      throw new Error(`LLM returned unexpected id ${id} (not in this batch)`);
+    }
+  }
+  for (const id of expectedIds) {
+    if (!corrections.some((c) => c.id === id)) {
+      throw new Error(`LLM missing correction for id ${id} in batch`);
+    }
+  }
+}
+async function redashify(text, options) {
+  if (!options || typeof options !== "object") {
+    throw new Error("redashify requires an options object with `apiKey` + `model`, `apiKey` + `provider`, or `llm`");
+  }
+  const batchSize = options.batchSize ?? 50;
+  requireInt(batchSize, "batchSize", 1);
+  const contextSize = options.contextSize ?? 50;
+  requireInt(contextSize, "contextSize", 0);
+  const llmFn = resolveLlm(options, "redashify");
+  const contexts = extractDashes(text, contextSize);
+  if (contexts.length === 0) {
+    return { text, corrections: [], unchanged: true };
+  }
+  const allCorrections = [];
+  for (let i = 0; i < contexts.length; i += batchSize) {
+    const batch = contexts.slice(i, i + batchSize);
+    const messages = buildMessages(batch, options.rules);
+    const response = await llmFn(messages);
+    const corrections = parseResponse(response);
+    validateBatchIds(corrections, new Set(batch.map((ctx) => ctx.id)));
+    allCorrections.push(...corrections);
+  }
+  const { text: correctedText, appliedCorrections } = applyCorrections(text, contexts, allCorrections);
+  return {
+    text: correctedText,
+    corrections: appliedCorrections,
+    unchanged: appliedCorrections.length === 0
+  };
+}
+export {
+  VALID_DASHES,
+  buildMessages,
+  extractDashes,
+  redashify
+};

package/package.json

@@ -1,7 +1,64 @@
 {
   "name": "redashify",
-  "version": "0.0.0",
+  "version": "0.2.0",
   "description": "Context-aware dash correction powered by LLMs",
-  "main": "index.js",
-  "license": "MIT"
+  "author": "Haihang Wang",
+  "type": "module",
+  "main": "./dist/index.cjs",
+  "module": "./dist/index.js",
+  "types": "./dist/index.d.ts",
+  "sideEffects": false,
+  "exports": {
+    ".": {
+      "import": {
+        "types": "./dist/index.d.ts",
+        "default": "./dist/index.js"
+      },
+      "require": {
+        "types": "./dist/index.d.cts",
+        "default": "./dist/index.cjs"
+      }
+    }
+  },
+  "files": [
+    "dist"
+  ],
+  "scripts": {
+    "build": "tsup",
+    "test": "vitest run",
+    "test:watch": "vitest",
+    "typecheck": "tsc --noEmit",
+    "typecheck:test": "tsc --noEmit -p tsconfig.test.json",
+    "prepublishOnly": "npm run typecheck && npm test && npm run build"
+  },
+  "keywords": [
+    "typography",
+    "dash",
+    "em-dash",
+    "en-dash",
+    "hyphen",
+    "legal",
+    "smartypants",
+    "llm"
+  ],
+  "engines": {
+    "node": ">=18"
+  },
+  "license": "Apache-2.0",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/hangingahaw/redashify.git"
+  },
+  "homepage": "https://github.com/hangingahaw/redashify#readme",
+  "bugs": {
+    "url": "https://github.com/hangingahaw/redashify/issues"
+  },
+  "dependencies": {
+    "@lexstyle/llm-client": "^0.1.0"
+  },
+  "devDependencies": {
+    "tsup": "^8.0.0",
+    "typescript": "^5.4.0",
+    "vitest": "^2.0.0"
+  }
 }

package/index.js

@@ -1 +0,0 @@
-module.exports = {};