From 8ff6721b9b87e200bfe9c5bd73d3c1cdb61c3bf7 Mon Sep 17 00:00:00 2001
From: Ben Harris <ben@tilde.team>
Date: Wed, 17 Aug 2022 15:20:41 -0400
Subject: [PATCH] add fulltext.php

---
 fulltext/.gitignore    |   1 +
 fulltext/composer.json |  14 +++
 fulltext/composer.lock | 224 +++++++++++++++++++++++++++++++++++++++++
 fulltext/fulltext.php  |  23 +++++
 4 files changed, 262 insertions(+)
 create mode 100644 fulltext/.gitignore
 create mode 100644 fulltext/composer.json
 create mode 100644 fulltext/composer.lock
 create mode 100644 fulltext/fulltext.php

diff --git a/fulltext/.gitignore b/fulltext/.gitignore
new file mode 100644
index 0000000..57872d0
--- /dev/null
+++ b/fulltext/.gitignore
@@ -0,0 +1 @@
+/vendor/
diff --git a/fulltext/composer.json b/fulltext/composer.json
new file mode 100644
index 0000000..615a6b0
--- /dev/null
+++ b/fulltext/composer.json
@@ -0,0 +1,14 @@
+{
+    "name": "benharri/fulltext",
+    "type": "project",
+    "require": {
+        "j0k3r/php-readability": "^2.0"
+    },
+    "license": "MIT",
+    "authors": [
+        {
+            "name": "Ben Harris",
+            "email": "ben@tilde.team"
+        }
+    ]
+}
diff --git a/fulltext/composer.lock b/fulltext/composer.lock
new file mode 100644
index 0000000..04d244b
--- /dev/null
+++ b/fulltext/composer.lock
@@ -0,0 +1,224 @@
+{
+    "_readme": [
+        "This file locks the dependencies of your project to a known state",
+        "Read more about it at https://getcomposer.org/doc/01-basic-usage.md#installing-dependencies",
+        "This file is @generated automatically"
+    ],
+    "content-hash": "d5efce9175131a522632dce527f8a3ce",
+    "packages": [
+        {
+            "name": "j0k3r/php-readability",
+            "version": "2.0.1",
+            "source": {
+                "type": "git",
+                "url": "https://github.com/j0k3r/php-readability.git",
+                "reference": "6689f199562836142fa4c3e82b1885ff56255ccb"
+            },
+            "dist": {
+                "type": "zip",
+                "url": "https://api.github.com/repos/j0k3r/php-readability/zipball/6689f199562836142fa4c3e82b1885ff56255ccb",
+                "reference": "6689f199562836142fa4c3e82b1885ff56255ccb",
+                "shasum": ""
+            },
+            "require": {
+                "ext-mbstring": "*",
+                "masterminds/html5": "^2.7",
+                "php": ">=7.2.0",
+                "psr/log": "^1.0"
+            },
+            "require-dev": {
+                "friendsofphp/php-cs-fixer": "^3.0",
+                "monolog/monolog": "^1.24|^2.1",
+                "phpstan/phpstan": "^1.3",
+                "phpstan/phpstan-phpunit": "^1.0",
+                "rector/rector": "^0.12.15",
+                "symfony/phpunit-bridge": "^4.4|^5.3|^6.0"
+            },
+            "suggest": {
+                "ext-tidy": "Used to clean up given HTML and to avoid problems with bad HTML structure."
+            },
+            "type": "library",
+            "autoload": {
+                "psr-4": {
+                    "Readability\\": "src/"
+                }
+            },
+            "notification-url": "https://packagist.org/downloads/",
+            "license": [
+                "Apache-2.0"
+            ],
+            "authors": [
+                {
+                    "name": "Jeremy Benoist",
+                    "email": "jeremy.benoist@gmail.com",
+                    "homepage": "http://www.j0k3r.net",
+                    "role": "Developer"
+                },
+                {
+                    "name": "DitherSky",
+                    "homepage": "https://github.com/Dither",
+                    "role": "Developer (https://github.com/Dither/full-text-rss)"
+                },
+                {
+                    "name": "Keyvan Minoukadeh",
+                    "email": "keyvan@keyvan.net",
+                    "homepage": "http://keyvan.net",
+                    "role": "Developer (ported original JS code to PHP)"
+                },
+                {
+                    "name": "Arc90",
+                    "homepage": "http://arc90.com",
+                    "role": "Developer (original JS version)"
+                }
+            ],
+            "description": "Automatic article extraction from HTML",
+            "keywords": [
+                "article",
+                "article extraction",
+                "content",
+                "content extraction",
+                "extraction",
+                "html"
+            ],
+            "support": {
+                "issues": "https://github.com/j0k3r/php-readability/issues",
+                "source": "https://github.com/j0k3r/php-readability/tree/2.0.1"
+            },
+            "funding": [
+                {
+                    "url": "https://github.com/j0k3r",
+                    "type": "github"
+                }
+            ],
+            "time": "2022-06-13T07:13:23+00:00"
+        },
+        {
+            "name": "masterminds/html5",
+            "version": "2.7.5",
+            "source": {
+                "type": "git",
+                "url": "https://github.com/Masterminds/html5-php.git",
+                "reference": "f640ac1bdddff06ea333a920c95bbad8872429ab"
+            },
+            "dist": {
+                "type": "zip",
+                "url": "https://api.github.com/repos/Masterminds/html5-php/zipball/f640ac1bdddff06ea333a920c95bbad8872429ab",
+                "reference": "f640ac1bdddff06ea333a920c95bbad8872429ab",
+                "shasum": ""
+            },
+            "require": {
+                "ext-ctype": "*",
+                "ext-dom": "*",
+                "ext-libxml": "*",
+                "php": ">=5.3.0"
+            },
+            "require-dev": {
+                "phpunit/phpunit": "^4.8.35 || ^5.7.21 || ^6 || ^7"
+            },
+            "type": "library",
+            "extra": {
+                "branch-alias": {
+                    "dev-master": "2.7-dev"
+                }
+            },
+            "autoload": {
+                "psr-4": {
+                    "Masterminds\\": "src"
+                }
+            },
+            "notification-url": "https://packagist.org/downloads/",
+            "license": [
+                "MIT"
+            ],
+            "authors": [
+                {
+                    "name": "Matt Butcher",
+                    "email": "technosophos@gmail.com"
+                },
+                {
+                    "name": "Matt Farina",
+                    "email": "matt@mattfarina.com"
+                },
+                {
+                    "name": "Asmir Mustafic",
+                    "email": "goetas@gmail.com"
+                }
+            ],
+            "description": "An HTML5 parser and serializer.",
+            "homepage": "http://masterminds.github.io/html5-php",
+            "keywords": [
+                "HTML5",
+                "dom",
+                "html",
+                "parser",
+                "querypath",
+                "serializer",
+                "xml"
+            ],
+            "support": {
+                "issues": "https://github.com/Masterminds/html5-php/issues",
+                "source": "https://github.com/Masterminds/html5-php/tree/2.7.5"
+            },
+            "time": "2021-07-01T14:25:37+00:00"
+        },
+        {
+            "name": "psr/log",
+            "version": "1.1.4",
+            "source": {
+                "type": "git",
+                "url": "https://github.com/php-fig/log.git",
+                "reference": "d49695b909c3b7628b6289db5479a1c204601f11"
+            },
+            "dist": {
+                "type": "zip",
+                "url": "https://api.github.com/repos/php-fig/log/zipball/d49695b909c3b7628b6289db5479a1c204601f11",
+                "reference": "d49695b909c3b7628b6289db5479a1c204601f11",
+                "shasum": ""
+            },
+            "require": {
+                "php": ">=5.3.0"
+            },
+            "type": "library",
+            "extra": {
+                "branch-alias": {
+                    "dev-master": "1.1.x-dev"
+                }
+            },
+            "autoload": {
+                "psr-4": {
+                    "Psr\\Log\\": "Psr/Log/"
+                }
+            },
+            "notification-url": "https://packagist.org/downloads/",
+            "license": [
+                "MIT"
+            ],
+            "authors": [
+                {
+                    "name": "PHP-FIG",
+                    "homepage": "https://www.php-fig.org/"
+                }
+            ],
+            "description": "Common interface for logging libraries",
+            "homepage": "https://github.com/php-fig/log",
+            "keywords": [
+                "log",
+                "psr",
+                "psr-3"
+            ],
+            "support": {
+                "source": "https://github.com/php-fig/log/tree/1.1.4"
+            },
+            "time": "2021-05-03T11:20:27+00:00"
+        }
+    ],
+    "packages-dev": [],
+    "aliases": [],
+    "minimum-stability": "stable",
+    "stability-flags": [],
+    "prefer-stable": false,
+    "prefer-lowest": false,
+    "platform": [],
+    "platform-dev": [],
+    "plugin-api-version": "2.2.0"
+}
diff --git a/fulltext/fulltext.php b/fulltext/fulltext.php
new file mode 100644
index 0000000..c8d38cf
--- /dev/null
+++ b/fulltext/fulltext.php
@@ -0,0 +1,23 @@
+<?php
+
+include_once 'vendor/autoload.php';
+
+use Readability\Readability;
+
+$url = urldecode($_GET["url"]);
+
+$html = file_get_contents($url);
+
+$readability = new Readability($html, $url);
+$result = $readability->init();
+
+if ($result) {
+    header('Content-type: text/plain');
+    echo $readability->getTitle()->textContent;
+    echo PHP_EOL;
+    echo wordwrap($readability->getContent()->textContent);
+}
+else {
+    echo 'failed';
+}
+