Initial commit

.gitignore

@@ -0,0 +1,3 @@
+.devenv/
+vendor/
+.php-cs-fixer.cache

.php-cs-fixer.dist.php

@@ -0,0 +1,17 @@
+<?php
+
+declare(strict_types=1);
+
+use PhpCsFixer\Config;
+use PhpCsFixer\Finder;
+
+return (new Config())
+    ->setRiskyAllowed(false)
+    ->setRules([
+        '@PER-CS' => true
+    ])
+    ->setFinder(
+        (new Finder())
+            ->in(__DIR__)
+    )
+;

composer.json

@@ -0,0 +1,19 @@
+{
+    "name": "armelnl/scandalous",
+    "description": "A wrapper around LiteParse for PDF extraction",
+    "type": "library",
+    "license": "MIT",
+    "autoload": {
+        "psr-4": {
+            "Scandalous\\": "src/"
+        }
+    },
+    "require": {
+	"php": "^8.4",
+        "symfony/process": "^8.1"
+    },
+    "require-dev": {
+        "friendsofphp/php-cs-fixer": "^3.95",
+        "phpstan/phpstan": "^2.2"
+    }
+}

devenv.lock

@@ -0,0 +1,65 @@
+{
+  "nodes": {
+    "devenv": {
+      "locked": {
+        "dir": "src/modules",
+        "lastModified": 1780630679,
+        "narHash": "sha256-hhQyVAYmNKziZ0T+T4Gsk0PYmnz4vdzOzpkJAmDASKM=",
+        "owner": "cachix",
+        "repo": "devenv",
+        "rev": "90ed6227ab389dd4e874a69a724f25dba312b754",
+        "type": "github"
+      },
+      "original": {
+        "dir": "src/modules",
+        "owner": "cachix",
+        "repo": "devenv",
+        "type": "github"
+      }
+    },
+    "nixpkgs": {
+      "inputs": {
+        "nixpkgs-src": "nixpkgs-src"
+      },
+      "locked": {
+        "lastModified": 1778507786,
+        "narHash": "sha256-HzSQCKMsMr8r55LwM1JuzIOB+8bzk0FEv6sItKvsfoY=",
+        "owner": "cachix",
+        "repo": "devenv-nixpkgs",
+        "rev": "8f24a228a782e24576b155d1e39f0d914b380691",
+        "type": "github"
+      },
+      "original": {
+        "owner": "cachix",
+        "ref": "rolling",
+        "repo": "devenv-nixpkgs",
+        "type": "github"
+      }
+    },
+    "nixpkgs-src": {
+      "flake": false,
+      "locked": {
+        "lastModified": 1778274207,
+        "narHash": "sha256-I4puXmX1iovcCHZlRmztO3vW0mAbbRvq4F8wgIMQ1MM=",
+        "owner": "NixOS",
+        "repo": "nixpkgs",
+        "rev": "b3da656039dc7a6240f27b2ef8cc6a3ef3bccae7",
+        "type": "github"
+      },
+      "original": {
+        "owner": "NixOS",
+        "ref": "nixpkgs-unstable",
+        "repo": "nixpkgs",
+        "type": "github"
+      }
+    },
+    "root": {
+      "inputs": {
+        "devenv": "devenv",
+        "nixpkgs": "nixpkgs"
+      }
+    }
+  },
+  "root": "root",
+  "version": 7
+}

devenv.nix

@@ -0,0 +1,20 @@
+{
+  pkgs,
+  lib,
+  config,
+  ...
+}:
+{
+  packages = [
+    pkgs.liteparse
+  ];
+
+  languages = {
+    php = {
+      enable = true;
+      packages = {
+        composer = pkgs.phpPackages.composer;
+      };
+    };
+  };
+}

src/Contract/ProcessRunnerInterface.php

@@ -0,0 +1,11 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Scandalous\Contract;
+
+interface ProcessRunnerInterface
+{
+    /** @param string[] $command */
+    public function run(array $command): string;
+}

src/Contract/TextExtractorInterface.php

@@ -0,0 +1,10 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Scandalous\Contract;
+
+interface TextExtractorInterface
+{
+    public function extract(string $file): string;
+}

src/Engine/LiteParseExtractor.php

@@ -0,0 +1,21 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Scandalous\Engine;
+
+use Scandalous\Contract\ProcessRunnerInterface;
+use Scandalous\Contract\TextExtractorInterface;
+
+final class LiteParseExtractor implements TextExtractorInterface
+{
+    public function __construct(
+        private ProcessRunnerInterface $runner,
+        private string $binary = 'lit',
+    ) {}
+
+    public function extract(string $file): string
+    {
+        return $this->runner->run([$this->binary, 'parse', $file]);
+    }
+}

src/Process/ProcessRunner.php

@@ -0,0 +1,20 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Scandalous\Process;
+
+use Scandalous\Contract\ProcessRunnerInterface;
+use Symfony\Component\Process\Process;
+
+final class ProcessRunner implements ProcessRunnerInterface
+{
+    /** @param string[] $command */
+    public function run(array $command): string
+    {
+        $process = new Process($command);
+        $process->mustRun();
+
+        return $process->getOutput();
+    }
+}

src/Scandalous.php

@@ -0,0 +1,19 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Scandalous;
+
+use Scandalous\Contract\TextExtractorInterface;
+
+class Scandalous
+{
+    public function __construct(
+        private readonly TextExtractorInterface $extractor,
+    ) {}
+
+    public function extract(string $pdfPath): string
+    {
+        return $this->extractor->extract($pdfPath);
+    }
+}