commit 57682ed72bb14dea16c460fc1cf1287de383c6e4
Author: Andre Henriques <andr3h3nriqu3s@gmail.com>
Date:   Mon Dec 25 22:44:16 2023 +0000

    Inital work on the lsp

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..6db8392
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,178 @@
+# Based on https://raw.githubusercontent.com/github/gitignore/main/Node.gitignore
+
+# Logs
+
+logs
+_.log
+npm-debug.log_
+yarn-debug.log*
+yarn-error.log*
+lerna-debug.log*
+.pnpm-debug.log*
+
+# Caches 
+
+.cache
+
+# Diagnostic reports (https://nodejs.org/api/report.html)
+
+report.[0-9]_.[0-9]_.[0-9]_.[0-9]_.json
+
+# Runtime data
+
+pids
+_.pid
+_.seed
+*.pid.lock
+
+# Directory for instrumented libs generated by jscoverage/JSCover
+
+lib-cov
+
+# Coverage directory used by tools like istanbul
+
+coverage
+*.lcov
+
+# nyc test coverage
+
+.nyc_output
+
+# Grunt intermediate storage (https://gruntjs.com/creating-plugins#storing-task-files)
+
+.grunt
+
+# Bower dependency directory (https://bower.io/)
+
+bower_components
+
+# node-waf configuration
+
+.lock-wscript
+
+# Compiled binary addons (https://nodejs.org/api/addons.html)
+
+build/Release
+
+# Dependency directories
+
+node_modules/
+jspm_packages/
+
+# Snowpack dependency directory (https://snowpack.dev/)
+
+web_modules/
+
+# TypeScript cache
+
+*.tsbuildinfo
+
+# Optional npm cache directory
+
+.npm
+
+# Optional eslint cache
+
+.eslintcache
+
+# Optional stylelint cache
+
+.stylelintcache
+
+# Microbundle cache
+
+.rpt2_cache/
+.rts2_cache_cjs/
+.rts2_cache_es/
+.rts2_cache_umd/
+
+# Optional REPL history
+
+.node_repl_history
+
+# Output of 'npm pack'
+
+*.tgz
+
+# Yarn Integrity file
+
+.yarn-integrity
+
+# dotenv environment variable files
+
+.env
+.env.development.local
+.env.test.local
+.env.production.local
+.env.local
+
+# parcel-bundler cache (https://parceljs.org/)
+
+.parcel-cache
+
+# Next.js build output
+
+.next
+out
+
+# Nuxt.js build / generate output
+
+.nuxt
+dist
+
+# Gatsby files
+
+# Comment in the public line in if your project uses Gatsby and not Next.js
+
+# https://nextjs.org/blog/next-9-1#public-directory-support
+
+# public
+
+# vuepress build output
+
+.vuepress/dist
+
+# vuepress v2.x temp and cache directory
+
+.temp
+
+# Docusaurus cache and generated files
+
+.docusaurus
+
+# Serverless directories
+
+.serverless/
+
+# FuseBox cache
+
+.fusebox/
+
+# DynamoDB Local files
+
+.dynamodb/
+
+# TernJS port file
+
+.tern-port
+
+# Stores VSCode versions used for testing VSCode extensions
+
+.vscode-test
+
+# yarn v2
+
+.yarn/cache
+.yarn/unplugged
+.yarn/build-state.yml
+.yarn/install-state.gz
+.pnp.*
+
+# IntelliJ based IDEs
+.idea
+
+# Finder (MacOS) folder config
+.DS_Store
+
+res
+test
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..7b993cb
--- /dev/null
+++ b/README.md
@@ -0,0 +1,15 @@
+# latex-lsp
+
+To install dependencies:
+
+```bash
+bun install
+```
+
+To run:
+
+```bash
+bun run index.ts
+```
+
+This project was created using `bun init` in bun v1.0.20. [Bun](https://bun.sh) is a fast all-in-one JavaScript runtime.
diff --git a/bun.lockb b/bun.lockb
new file mode 100755
index 0000000..6d49e0e
Binary files /dev/null and b/bun.lockb differ
diff --git a/index.ts b/index.ts
new file mode 100644
index 0000000..816b1ef
--- /dev/null
+++ b/index.ts
@@ -0,0 +1,16 @@
+import fs from 'fs';
+import process from 'process';
+import { getDiagnostics, parseLsp } from './parser';
+
+if (process.argv.length !== 3) {
+  console.log("Please provide only one pass");
+  process.exit(1);
+}
+
+const file = fs.readFileSync(process.argv[2]).toString();
+
+const res = parseLsp(file);
+
+fs.writeFileSync('./res', res.text);
+
+console.log(await getDiagnostics(file));
diff --git a/lsp.ts b/lsp.ts
new file mode 100644
index 0000000..980f666
--- /dev/null
+++ b/lsp.ts
@@ -0,0 +1,218 @@
+import fs from 'fs';
+import * as process from 'process';
+import { getDiagnostics } from './parser';
+
+export function log(str: string) {
+  fs.appendFileSync('/home/sylv/latex-lsp/test', str + "\n");
+}
+
+export function error(str: string) {
+  log("ERROR:" + str);
+  process.exit(1);
+}
+
+let init = false;
+
+type RPCResult = {
+  jsonrpc: string,
+  id: string,
+} & ({
+  result: any,
+} | {
+  error: any,
+})
+
+type RPCRequest = {
+  id: string;
+  jsonrpc: string,
+  method: string,
+  params: string,
+}
+
+type InitRequest = RPCRequest & {
+  method: 'initialize',
+}
+
+// ASSUMES that is has the right format
+function sendRPC(obj: any) {
+  const msg = JSON.stringify(obj);
+  const buff = Buffer.from(msg, 'utf-8');
+  
+  const fullmessage = `Content-Length: ${buff.byteLength}\r\n\r\n${msg}`;
+
+  log("Sending message: " + fullmessage.replace(/\n/g, '\\n').replace(/\r/g, '\\r'))
+
+  process.stdout.cork()
+  process.stdout.write(fullmessage);
+  process.stdout.uncork();
+}
+
+function sendRPCMessage(req: RPCRequest, result: any = undefined, error: any = undefined) {
+  if (!result && !error) {
+    error("Invalid rpc message to send")
+    return;
+  }
+
+  sendRPC({
+    id: req.id,
+    jsonrpc: req.jsonrpc,
+    result,
+    error
+  } as RPCResult)
+}
+
+function notifyRPC(method: string, params: any) {
+  sendRPC({
+    jsonrpc: '2.0',
+    method,
+    params,
+  })
+}
+
+type TextDocumentDidOpen = RPCRequest & {
+  method: 'textDocument/didOpen',
+  params: {
+    textDocument: {
+      uri: string;
+      languageId: string;
+      version: number;
+      text: string;
+    }
+  }
+}
+
+type Position = {
+  line: number,
+  character: number
+}
+
+type Range = {
+  start: Position,
+  end: Position,
+}
+
+export type Dialog = {
+  range: Range,
+  message: string,
+  severity: Severity,
+}
+
+export enum Severity {
+  Error = 1,
+  Warning = 2,
+  Information = 3,
+  Hint = 4,
+}
+
+async function handleJSON(req: RPCRequest) {
+  log(`New Message entered: method: ${req.method}`);
+
+  if (init) {
+    if (req.method === 'initialized') {
+      // On init confirm do nothing
+      return;
+    } else if (req.method === 'textDocument/didOpen') {
+      const reqO: TextDocumentDidOpen = req as any;
+
+      if (reqO.params.textDocument.languageId !== 'latex') {
+        error(`This server only supports latex! Got: ${reqO.params.textDocument.languageId}`)
+        return;
+      }
+
+      const params = {
+        uri: reqO.params.textDocument.uri,
+        diagnostics: (await getDiagnostics(reqO.params.textDocument.text)),
+      }
+
+      notifyRPC('textDocument/publishDiagnostics', params);
+
+
+      //error(`TODO ${req.method}`);
+    } else {
+      error(`Handle: ${req.method} after init`);
+    }
+  } else {
+    if (req.method === 'initialize') {
+      log("Recived init");
+
+      init = true;
+
+      log(JSON.stringify(req));
+
+      sendRPCMessage(req, {
+        capabilities: {
+          diagnosticProvider: {
+            // TODO change in the future when the server also checks the ib file as well
+            interFileDependencies: false,
+            workspaceDiagnostics: false,
+          }
+        },
+        serverInfo: {
+          name: "Super Cool latex server",
+          version: "1.0.0",
+        }
+      });
+
+      return;
+    } else {
+      error(`Expected init method found '${req.method}'`)
+    }
+  }
+}
+
+
+export function handleData(data: Buffer) {
+  let strData = data.toString();
+  let jsonData = strData.split('\n');
+
+  let req: RPCRequest;
+
+  if (jsonData.length === 3) {
+    try {
+      req = JSON.parse(jsonData[2]);
+    } catch (e) {
+      log("Failed to parse json 1/2! Msg:\n");
+      log(strData.replace(/\n/g, '\\n').replace(/\r/g, '\\r'));
+      log("\n\nFailed to parse json 2/2! JSON Line:\n");
+      log(jsonData[2]);
+      return;
+    }
+    handleJSON(req);
+    return;
+  }
+
+  for (let i = 2; i < jsonData.length; i += 2) {
+
+    let jsonLine = jsonData[i];
+    if (i != jsonData.length - 1) {
+      let index = jsonLine.indexOf('Content-Length');
+      if (index == -1) {
+        error("Handling multiline expected 'Content-Length'");
+        return;
+      }
+
+      try {
+        req = JSON.parse(jsonLine.substring(0, index));
+      } catch (e) {
+        log("Failed to parse json 1/2! Msg:\n");
+        log(strData.replace(/\n/g, '\\n').replace(/\r/g, '\\r'));
+        log("\n\nFailed to parse json 2/2! JSON Line:\n");
+        log(jsonLine.substring(0, index));
+        return;
+      }
+    } else {
+      try {
+        req = JSON.parse(jsonLine);
+      } catch (e) {
+        log("Failed to parse json 1/2! Msg:\n");
+        log(strData.replace(/\n/g, '\\n').replace(/\r/g, '\\r'));
+        log("\n\nFailed to parse json 2/2! JSON Line:\n");
+        log(jsonLine);
+        return;
+      }
+
+    }
+
+    handleJSON(req);
+  }
+}
diff --git a/lsp_run.ts b/lsp_run.ts
new file mode 100644
index 0000000..b686078
--- /dev/null
+++ b/lsp_run.ts
@@ -0,0 +1,13 @@
+import { error, handleData, log } from "./lsp";
+
+console.log = log;
+
+process.stdin.resume();
+
+process.stdin.on('data', function(data) {
+  try {
+    handleData(data);
+  } catch (e) {
+    error("TRY GOT ERROR:" + e)
+  }
+})
diff --git a/package.json b/package.json
new file mode 100644
index 0000000..43881a7
--- /dev/null
+++ b/package.json
@@ -0,0 +1,11 @@
+{
+  "name": "latex-lsp",
+  "module": "index.ts",
+  "type": "module",
+  "devDependencies": {
+    "@types/bun": "latest"
+  },
+  "peerDependencies": {
+    "typescript": "^5.0.0"
+  }
+}
\ No newline at end of file
diff --git a/parser.ts b/parser.ts
new file mode 100644
index 0000000..1682911
--- /dev/null
+++ b/parser.ts
@@ -0,0 +1,474 @@
+import fs from 'fs';
+
+import process from 'process';
+import { Severity, type Dialog } from './lsp';
+
+
+type ParseResultConversion = {
+  length: number,
+  original_length?: number,
+  original_position: number,
+  position: number,
+  type: 'text' | 'h1' | 'h2'
+};
+
+type ParseResult = {
+  text: string,
+  originalString: string,
+  conversions: ParseResultConversion[],
+}
+
+// Returns the number of character skiped
+// This puts the i at the \n so it's skiped
+function parseComment(text: string, curPos: number): number {
+  for (let i = curPos + 1; i < text.length; i++) {
+    const char = text[i];
+    if (char === '\n') {
+      return i - curPos;
+    }
+  }
+  return text.length - curPos;
+}
+
+function createPartition(text: string, startPos: number, curPos: number, result: string): [ParseResultConversion[], string] | null {
+  if (startPos >= curPos || text.substring(startPos, curPos).match(/^\s*$/)) {
+    return null;
+  }
+
+  var t = text.substring(startPos, curPos);
+
+  if (t.indexOf('\n') == -1) {
+    return [[{
+      length: curPos - startPos,
+      position: result.length,
+      original_position: startPos,
+      type: 'text',
+    }], t];
+  }
+
+  const split = t.split('\n');
+
+  let nt = "";
+  const convs: ParseResultConversion[] = [];
+
+  let n = startPos;
+
+  let pos = result.length;
+
+  for (const line of split) {
+    let nLine = line.replace(/^\s*/, '');
+    n += line.length - nLine.length;
+    nt += nLine + '\n';
+    convs.push({
+      length: nLine.length,
+      original_position: n,
+      position: pos,
+      type: 'text'
+    });
+    pos += nLine.length + 1;
+    n += nLine.length + 1;
+  }
+
+  return [convs, nt];
+
+}
+
+function isChar(charCode: number): boolean {
+  return (charCode >= 92 && charCode <= 122) || (charCode >= 65 && charCode <= 90);
+}
+function readBalanced(endChar: string, startChar: string, text: string, curPos: number): number {
+  let bal = 1;
+  for (let i = curPos; i < text.length; i++) {
+    const char = text[i];
+    if (char == endChar) {
+      if (bal == 1) {
+        return i - curPos + 1;
+      } else {
+        bal -= 1;
+      }
+    } else if (char == startChar) {
+      bal += 1;
+    }
+  }
+  throw new Error("Can not find end of balance read")
+}
+
+function isWhiteSpace(char: string): boolean {
+  return [' ', '\t', '\n'].includes(char);
+}
+
+function parseCommand(text: string, curPos: number, result: string): [number, ParseResultConversion, string] | [number] {
+
+  if (text.length - 1 == curPos) {
+    throw new Error("The latex file has the wrong format the file can not end with a empty command");
+  }
+
+  if (text[curPos + 1] === '\\') {
+    return [1, {
+      length: 1,
+      position: result.length,
+      original_position: curPos + 1,
+      type: 'text'
+    }, '\\'];
+  } else if (text[curPos + 1] === '%') {
+    return [1, {
+      length: 1,
+      position: result.length,
+      original_position: curPos + 1,
+      type: 'text'
+    }, '%'];
+  }
+
+  let commandName = "";
+  let commandNameFinished = false;
+
+  // TODO store the location of the opts and args
+  let options = [];
+  let args = [];
+
+  let findEnd = false;
+
+
+  let len = 0;
+
+  for (let i = curPos + 1; i < text.length; i++) {
+    const char = text[i];
+    if (isChar(char.charCodeAt(0))) {
+      if (!commandNameFinished) {
+        commandName += char;
+      } else {
+        len = i;
+        findEnd = true;
+        break;
+      }
+    } else if (char === '[') {
+      commandNameFinished = true;
+      const len = readBalanced(']', '[', text, i + 1);
+      options.push(text.substring(i + 1, i + len))
+      i += len;
+    } else if (char === '{') {
+      commandNameFinished = true;
+      const len = readBalanced('}', '{', text, i + 1);
+      args.push(text.substring(i + 1, i + len))
+      i += len;
+    } else if (isWhiteSpace(char)) {
+      len = i;
+      findEnd = true;
+      break;
+    } else {
+
+      if (char == '.' || char == ',') {
+        if (commandNameFinished) {
+          len = i;
+          findEnd = true;
+          break;
+        }
+      }
+
+      console.log(text.substring(i - 20, i + 20));
+      console.log('Char:', char.charCodeAt(0));
+      throw new Error("TODO handle not char chars in the parse command function");
+    }
+  }
+
+  if (!findEnd) {
+    throw new Error("Could not end of the command");
+  }
+
+  len = len - curPos;
+
+  switch (commandName) {
+    case 'documentclass':
+    case 'usepackage':
+    case 'graphicspath':
+    case 'hypersetup':
+    case 'pagestyle':
+    case 'fancyhead':
+    case 'fancyfoot':
+    case 'renewcommand':
+    case 'setlength':
+    case 'addbibresource':
+    case 'date':
+    case 'maketitle':
+    case 'newpage':
+    case 'tableofcontents':
+    case 'includegraphics':
+    case 'appendix':
+    case 'printbibliography':
+      return [len];
+    case 'title':
+    case 'author':
+    case 'end':
+      console.log("TODO: add way to check the", commandName)
+      return [len];
+
+    case 'cite':
+      console.log("TODO check if it exists on the bibliography");
+      return [len];
+
+    case 'begin':
+    case 'item':
+      console.log("TODO handle", commandName, ":", args)
+      return [len];
+
+    case 'section':
+      return [len, {
+        length: args[0].length + 1,
+        original_length: len,
+        original_position: curPos - commandName.length,
+        position: result.length,
+        type: 'h1',
+      }, args[0] + '\n']
+
+    case 'subsection':
+      return [len, {
+        length: args[0].length + 1,
+        original_length: len,
+        original_position: curPos - commandName.length,
+        position: result.length,
+        type: 'h2',
+      }, args[0] + '\n']
+
+    default:
+      console.log("Command name:", commandName, "options:", options, "args:", args);
+      throw new Error("TODO handle this case");
+  }
+
+
+}
+
+function readUntil(text: string, char: string, curPos: number): number {
+  for (let i = curPos; i < text.length; i++) {
+    if (text[i] === char) {
+      return i - curPos;
+    }
+  }
+
+  throw new Error("Could not find matching pair");
+}
+
+export function parseLsp(text: string): ParseResult {
+  const result: ParseResult = {
+    text: '',
+    originalString: text,
+    conversions: [],
+  };
+
+  let conversionStartPosition = 0;
+
+  for (let i = 0; i < text.length; i++) {
+    let char = text[i];
+
+    if (char === '%') {
+      //console.log("Found comment");
+
+      const possiblePartition = createPartition(text, conversionStartPosition, i - 1, result.text);
+
+      if (possiblePartition) {
+        const [conv, toAdd] = possiblePartition;
+        result.conversions = result.conversions.concat(conv)
+        result.text += toAdd;
+      }
+
+      const len = parseComment(text, i);
+
+      i += len;
+
+      // Skip the begining \n
+      conversionStartPosition = i + 1;
+
+    } else if (char === '\\') {
+      //console.log("Found command")
+
+      const possiblePartition = createPartition(text, conversionStartPosition, i - 1, result.text);
+
+      if (possiblePartition) {
+        const [conv, toAdd] = possiblePartition;
+        result.conversions = result.conversions.concat(conv);
+        result.text += toAdd;
+      }
+
+      const res = parseCommand(text, i, result.text);
+
+      if (res.length === 1) {
+        const [len] = res;
+        i += len;
+      } else {
+
+        const [len, conv, toAdd] = res;
+
+        result.conversions.push(conv);
+        result.text += toAdd;
+        i += len;
+      }
+
+      conversionStartPosition = i + 1;
+    } else if (char === '$') {
+      console.log('Found math expr')
+      if (text[i + 1] === '$') {
+        throw new Error("Handle double math expression");
+      }
+
+      const possiblePartition = createPartition(text, conversionStartPosition, i - 1, result.text);
+
+      if (possiblePartition) {
+        const [conv, toAdd] = possiblePartition;
+        result.conversions = result.conversions.concat(conv);
+        result.text += toAdd;
+      }
+
+      const len = readUntil(text, '$', i + 1);
+
+      i += len + 1;
+
+      conversionStartPosition = i + 1;
+    } else {
+      //console.log(char);
+    }
+
+  }
+
+  return result;
+}
+
+function getLineAndChar(lineIndex: number[], offset: number): [number, number] {
+  let l = 0;
+  let r = lineIndex.length;
+
+  while (r >= l) {
+    const i = Math.floor((r + l) / 2);
+    //console.log(i, l ,r, offset, lineIndex[i]);
+
+    if (lineIndex[i + 1] < offset) {
+      l = i + 1;
+      continue;
+    }
+    if (lineIndex[i] > offset) {
+      r = i - 1;
+      continue;
+    }
+    return [i, offset - lineIndex[i]];
+  }
+
+  return [-1, -1];
+}
+
+
+function getOriginalPostion(res: ParseResult, offset: number): number {
+  let l = 0;
+  let r = res.conversions.length;
+
+  while (r >= l) {
+    const i = Math.floor((r + l) / 2);
+    const conv = res.conversions[i];
+
+    if (conv.position > offset) {
+      r = i - 1;
+      continue;
+    }
+
+    if (conv.position + conv.length < offset) {
+      l = i + 1;
+      continue;
+    }
+
+    return conv.original_position + (offset - conv.position);
+  }
+
+  return -1;
+}
+
+function buildLineIndex(file: string) {
+  const lines = file.split('\n');
+  let i = 0;
+  const lineIndex = [0];
+  for (const line of lines) {
+    i += line.length + 1;
+    lineIndex.push(i);
+  }
+  return lineIndex;
+}
+
+export async function getDiagnostics(file: string): Promise<Dialog[]> {
+
+  const res = parseLsp(file);
+
+  const formData = new URLSearchParams();
+
+  formData.set('text', res.text);
+  formData.set('language', 'en-GB');
+  formData.set('username', process.env.USERNAME ?? '');
+  formData.set('apiKey', process.env.APIKEY ?? '');
+  formData.set('level', 'picky');
+
+  const rawRes = await fetch('https://api.languagetoolplus.com/v2/check', {
+    method: 'POST',
+    headers: {
+      'Accept': 'application/json',
+    },
+    body: formData,
+  });
+
+  if (rawRes.status !== 200) {
+    process.exit(2);
+  }
+  const body = await rawRes.json();
+
+  type Match = {
+    message: string,
+    shortMessage: string,
+    offset: number,
+    length: number,
+    replacements: {
+      value: string
+    }[],
+    context: {
+      text: string,
+      offset: number,
+      length: number
+    },
+    sentence: string,
+    rule: {
+      id: string,
+      subId: string,
+      description: string,
+      urls: {
+        value: string
+      }[],
+      issueType: string,
+      category: {
+        id: string,
+        name: string
+      }
+    }
+  }
+
+  const lineIndex = buildLineIndex(file);
+
+  const diags = [];
+
+  for (const i of body.matches) {
+    const match: Match = i;
+    const original_position = getOriginalPostion(res, match.offset);
+    if (original_position == -1) {
+      console.log("Could not find the original position")
+      continue;
+    }
+    const [startLine, startChar] = getLineAndChar(lineIndex, original_position);
+    const [endLine, endChar] = getLineAndChar(lineIndex, original_position + match.length);
+
+    const range = {
+      start: { line: startLine, character: startChar },
+      end: { line: endLine, character: endChar },
+    }
+
+    diags.push({
+      range,
+      severity: Severity.Error,
+      message: match.message,
+    })
+  }
+
+  return diags;
+}
+
diff --git a/tsconfig.json b/tsconfig.json
new file mode 100644
index 0000000..dcd8fc5
--- /dev/null
+++ b/tsconfig.json
@@ -0,0 +1,22 @@
+{
+  "compilerOptions": {
+    "lib": ["ESNext"],
+    "target": "ESNext",
+    "module": "ESNext",
+    "moduleDetection": "force",
+    "jsx": "react-jsx",
+    "allowJs": true,
+
+    /* Bundler mode */
+    "moduleResolution": "bundler",
+    "allowImportingTsExtensions": true,
+    "verbatimModuleSyntax": true,
+    "noEmit": true,
+
+    /* Linting */
+    "skipLibCheck": true,
+    "strict": true,
+    "noFallthroughCasesInSwitch": true,
+    "forceConsistentCasingInFileNames": true
+  }
+}