extrac prices pdfco

2025-11-17 15:05:10 +01:00
parent 6b65b31932
commit 1804605cad
16 changed files with 1131 additions and 330 deletions
@@ -0,0 +1,217 @@
+<?php
+header("Content-Type: application/json; charset=utf-8");
+
+// ---------------------------------------------------------
+// 1) CONFIGURAZIONE
+// ---------------------------------------------------------
+$apiKey    = "info@claudiosironi.com_Qfh02D7sAvi2tcx3ZchHpusNaBquCKhJw81fEnkHe2ersQDVOex4IokhCCzaFAz1";
+$fileToken = "filetoken://61a780917907f86a340290d22c449357dc68950e9066bd67b2";
+
+// opzionale: passare un token via GET
+if (isset($_GET['token']) && $_GET['token'] !== "") {
+    $fileToken = $_GET['token'];
+}
+
+// ---------------------------------------------------------
+// 2) PDF.CO → JSON2 (TUTTE LE PAGINE)
+// ---------------------------------------------------------
+$endpoint = "https://api.pdf.co/v1/pdf/convert/to/json2";
+
+$payload = [
+    "url"          => $fileToken,
+    "inline"       => true,
+    "detectTables" => true,
+    "pages"        => "all"
+];
+
+$ch = curl_init($endpoint);
+curl_setopt($ch, CURLOPT_POST, true);
+curl_setopt($ch, CURLOPT_HTTPHEADER, [
+    "Content-Type: application/json",
+    "x-api-key: $apiKey"
+]);
+curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($payload));
+curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
+
+$response = curl_exec($ch);
+
+if (!$response) {
+    echo json_encode(["error" => "Errore CURL: " . curl_error($ch)]);
+    exit;
+}
+
+curl_close($ch);
+
+$json = json_decode($response, true);
+
+// ---------------------------------------------------------
+// 3) VALIDAZIONE
+// ---------------------------------------------------------
+if (!isset($json["body"]["document"]["page"])) {
+    echo json_encode([
+        "error" => true,
+        "message" => "PDF.co JSON senza pagine valide",
+        "raw" => $json
+    ]);
+    exit;
+}
+
+$pages = $json["body"]["document"]["page"];
+
+// ---------------------------------------------------------
+// 4) FUNZIONI UTILI
+// ---------------------------------------------------------
+
+// Formato A – 291 02F
+function isCodeTypeA($txt)
+{
+    return preg_match('/^\d{3}\s+[0-9A-Z]{2,3}$/', $txt);
+}
+
+// Formato C – blocchi ripetuti (499 3A 14 5.475)
+function extractTypeCBlocks($line)
+{
+    preg_match_all(
+        '/(\d{3})\s+([0-9A-Z]{1,3})\s+(\d{2})\s+(\d{1,3}\.\d{3})/',
+        $line,
+        $m,
+        PREG_SET_ORDER
+    );
+
+    $out = [];
+    foreach ($m as $b) {
+        $out[] = [
+            "codice"     => $b[1],
+            "variante"   => $b[2],
+            "dimensione" => $b[3],
+            "prezzo"     => floatval(str_replace(".", "", $b[4]))
+        ];
+    }
+    return $out;
+}
+
+// ---------------------------------------------------------
+// 5) PARSER PER UNA SINGOLA RIGA (ricostruita)
+// ---------------------------------------------------------
+function parseLine($line)
+{
+    // FORMATO A
+    if (preg_match('/^(\d{3}\s+[0-9A-Z]{2,3})\s+(.*)$/', $line, $mA)) {
+
+        $codice = trim($mA[1]);
+        $resto  = trim($mA[2]);
+
+        preg_match_all('/\d{1,3}[.,]\d{2,3}/', $resto, $matchesPrezzi);
+
+        if (count($matchesPrezzi[0]) >= 2) {
+
+            $descr = trim(preg_replace('/\d{1,3}[.,]\d{2,3}/', '', $resto));
+
+            $vals = array_map(function ($v) {
+                return floatval(str_replace(",", ".", str_replace(".", "", $v)));
+            }, $matchesPrezzi[0]);
+
+            return [
+                "type" => "A",
+                "codice" => $codice,
+                "descrizione" => $descr,
+                "prezzi" => $vals
+            ];
+        }
+    }
+
+    // FORMATO B1
+    if (preg_match(
+        '/^(\d{3}\s+[A-Z]{1,3})\s+([A-Za-zÀ-ù0-9\s]+?)\s+((?:\d{1,3}[.,]\d{2,3}\s*){4,})$/',
+        $line,
+        $mB
+    )) {
+        $codice = trim($mB[1]);
+        $descr  = trim($mB[2]);
+        $valuesString = trim($mB[3]);
+
+        preg_match_all('/\d{1,3}[.,]\d{2,3}/', $valuesString, $vv);
+
+        $vals = array_map(function ($v) {
+            return floatval(str_replace(",", ".", str_replace(".", "", $v)));
+        }, $vv[0]);
+
+        return [
+            "type" => "B1",
+            "codice" => $codice,
+            "descrizione" => $descr,
+            "prezzi" => $vals
+        ];
+    }
+
+    // FORMATO C (ripetuto in riga)
+    $blocks = extractTypeCBlocks($line);
+
+    if (!empty($blocks)) {
+
+        $desc = $line;
+        foreach ($blocks as $b) {
+            $pattern = sprintf(
+                '/%s\s+%s\s+%s\s+\d{1,3}\.\d{3}/',
+                $b["codice"],
+                $b["variante"],
+                $b["dimensione"]
+            );
+            $desc = preg_replace($pattern, "", $desc);
+        }
+
+        $desc = trim($desc);
+
+        return [
+            "type" => "C",
+            "descrizione" => $desc,
+            "varianti" => $blocks
+        ];
+    }
+
+    // Nessun match
+    return null;
+}
+
+// ---------------------------------------------------------
+// 6) PROCESSAMENTO DI TUTTE LE PAGINE
+// ---------------------------------------------------------
+$items = [];
+
+foreach ($pages as $p) {
+
+    if (!isset($p["row"])) continue;
+
+    // Ricostruisco tutte le vere righe (“flat lines”)
+    $flatLines = [];
+
+    foreach ($p["row"] as $row) {
+        $line = "";
+        foreach ($row["column"] as $col) {
+            if (!empty($col["text"]["text"])) {
+                $line .= " " . trim($col["text"]["text"]);
+            }
+        }
+        $line = trim($line);
+        if ($line !== "") {
+            $flatLines[] = $line;
+        }
+    }
+
+    // Applico i parser su ogni riga ricostruita
+    foreach ($flatLines as $l) {
+        $parsed = parseLine($l);
+        if ($parsed !== null) {
+            $items[] = $parsed;
+        }
+    }
+}
+
+// ---------------------------------------------------------
+// 7) OUTPUT FINALE
+// ---------------------------------------------------------
+echo json_encode([
+    "status" => "ok",
+    "total"  => count($items),
+    "items"  => $items
+], JSON_PRETTY_PRINT);