From f108f026b9ed7a963bff2361f786d3c766a64c3b Mon Sep 17 00:00:00 2001
From: Lorenz Stechauner <lorenz.stechauner@necronda.net>
Date: Sun, 13 Jul 2025 22:04:06 +0200
Subject: [PATCH] organic/pdf: Add pdf upload feature

---
 www/organic/pdf.php | 58 ++++++++++++++++++++++++++++++++++++---------
 1 file changed, 47 insertions(+), 11 deletions(-)

diff --git a/www/organic/pdf.php b/www/organic/pdf.php
index d19495d..abd21f7 100644
--- a/www/organic/pdf.php
+++ b/www/organic/pdf.php
@@ -1,9 +1,9 @@
 <?php
 
-if ($_SERVER['REQUEST_METHOD'] !== 'GET' && $_SERVER['REQUEST_METHOD'] !== 'HEAD') {
+if ($_SERVER['REQUEST_METHOD'] !== 'GET' && $_SERVER['REQUEST_METHOD'] !== 'POST' && $_SERVER['REQUEST_METHOD'] !== 'HEAD') {
     header('Status: 405');
     header('Content-Length: 0');
-    header('Allow: GET, HEAD');
+    header('Allow: GET, POST, HEAD');
     exit;
 }
 
@@ -14,7 +14,7 @@ if ($info !== '') {
     exit;
 }
 
-$url = str_replace(' ', '+', $_GET['url']);
+$url = isset($_GET['url']) ? str_replace(' ', '+', $_GET['url']) : null;
 $format = $_GET['format'] ?? 'json';
 
 function get_address($array, $from, $to): array {
@@ -42,7 +42,31 @@ function jenc($data): string {
 
 if ($format === 'text') {
     header('Content-Type: text/plain; charset=UTF-8');
-    passthru("curl -s '" . escapeshellarg($url) . "' | pdftotext -raw - -");
+
+    if ($_SERVER['REQUEST_METHOD'] === 'POST') {
+        $fd_spec = [
+            0 => ["pipe", "r"],  // stdin
+            1 => ["pipe", "w"],  // stdout
+            2 => ["pipe", "w"],  // stderr
+        ];
+        $process = proc_open(['pdftotext',  '-raw', '-', '-'], $fd_spec, $pipes);
+        $input = fopen("php://input", "rb");
+        while (!feof($input)) {
+            if (($buffer = fread($input, 8192)) === false)
+                break;
+            fwrite($pipes[0], $buffer);
+        }
+        fclose($input);
+        fclose($pipes[0]);
+
+        fpassthru($pipes[1]);
+        fclose($pipes[1]);
+        $stderr = stream_get_contents($pipes[2]);
+        fclose($pipes[2]);
+        $return_value = proc_close($process);
+    } else {
+        passthru("curl -s '" . escapeshellarg($url) . "' | pdftotext -raw - -");
+    }
 } else if ($format === 'json') {
     header('Content-Type: application/json; charset=UTF-8');
 
@@ -51,13 +75,25 @@ if ($format === 'text') {
         1 => ["pipe", "w"],  // stdout
         2 => ["pipe", "w"],  // stderr
     ];
-    $process = proc_open(
-        ['bash', '-c',
-            "curl -s " . escapeshellarg($url) . " | " .
-            "pdftotext -raw - -"],
-        $fd_spec,
-        $pipes
-    );
+    if ($_SERVER['REQUEST_METHOD'] === 'POST') {
+        $process = proc_open(['pdftotext',  '-raw', '-', '-'], $fd_spec, $pipes);
+        $input = fopen("php://input", "rb");
+        while (!feof($input)) {
+            if (($buffer = fread($input, 8192)) === false)
+                break;
+            fwrite($pipes[0], $buffer);
+        }
+        fclose($input);
+    } else {
+        $process = proc_open(
+            ['bash', '-c',
+                "curl -s " . escapeshellarg($url) . " | " .
+                "pdftotext -raw - -"],
+            $fd_spec,
+            $pipes
+        );
+    }
+
     fclose($pipes[0]);
     $text = stream_get_contents($pipes[1]);
     fclose($pipes[1]);