Nodered PDF Parseer

flow.json

[
    {
        "id": "d197b28e9395b61b",
        "type": "ui-file-input",
        "z": "e93f3133440568a7",
        "group": "8948f2c8ad9f604f",
        "name": "",
        "order": 1,
        "width": 0,
        "height": 0,
        "topic": "topic",
        "topicType": "msg",
        "label": "File Input",
        "icon": "paperclip",
        "allowMultiple": false,
        "accept": "",
        "className": "",
        "x": 600,
        "y": 340,
        "wires": [
            [
                "81ca42c6330d3361",
                "d649e707401cb0a7"
            ]
        ]
    },
    {
        "id": "1308ad842a729aae",
        "type": "debug",
        "z": "e93f3133440568a7",
        "name": "SUCCESS",
        "active": true,
        "tosidebar": true,
        "console": false,
        "tostatus": false,
        "complete": "true",
        "targetType": "full",
        "statusVal": "",
        "statusType": "auto",
        "x": 790,
        "y": 380,
        "wires": []
    },
    {
        "id": "81ca42c6330d3361",
        "type": "function",
        "z": "e93f3133440568a7",
        "name": "Parse PDF",
        "func": "/**\n * Extracts text content from a PDF buffer using the globally available pdf-parse library.\n * Assumes 'pdfParse' has been added to functionGlobalContext in settings.js.\n * Expects msg.payload to be a Buffer containing the PDF data.\n * Sends the result to one of two outputs:\n *  - Output 1: Success - msg.payload contains the extracted text (string).\n *                     msg.pdfInfo contains metadata from pdf-parse.\n *  - Output 2: Failure - Original msg with msg.error containing the error object.\n *\n * Requires the 'pdf-parse' library (https://www.npmjs.com/package/pdf-parse)\n * Library MUST be configured in settings.js -> functionGlobalContext.\n * @param {NodeMessage & {payload: Buffer}} msg The input message.\n */\n\n// --- Input Validation ---\n// Check if pdfParse is available globally (configured in settings.js)\nif (typeof pdfParse === 'undefined') {\n    // node.error can take the msg object as the second argument\n    node.error(\"Global 'pdfParse' not found. Ensure 'pdf-parse' is installed and added to functionGlobalContext in settings.js, then restart Node-RED.\", msg);\n    // Send to error output\n    node.send([null, msg]);\n    return; // Stop execution\n}\n\nif (!msg.payload || !Buffer.isBuffer(msg.payload)) {\n    // CORRECTED: node.warn only takes one argument (the message string)\n    node.warn(\"msg.payload is missing or not a Buffer.\");\n    // Send to error output\n    msg.error = new Error(\"Input payload is not a valid Buffer.\");\n    node.send([null, msg]);\n    return; // Stop execution\n}\n\nconst pdfBuffer = msg.payload;\n// node.log only takes one argument\nnode.log(`Received PDF buffer, size: ${pdfBuffer.length} bytes. Starting parsing.`);\n\n// --- Asynchronous PDF Parsing ---\n// Use the globally available pdfParse function\npdfParse(pdfBuffer)\n    .then(function(data) {\n        // --- Success ---\n        // node.log only takes one argument\n        node.log(`Successfully parsed PDF: ${data.numpages} pages, ${data.text.length} characters.`);\n\n        // Prepare the output message for the success path (Output 1)\n        msg.payload = data.text; // Replace buffer with extracted text\n\n        // Optionally add other extracted info\n        msg.pdfInfo = {\n            numPages: data.numpages,\n            numRender: data.numrender,\n            info: data.info,\n            metadata: data.metadata,\n            version: data.version\n        };\n\n        // Send the message to the first output\n        node.send([msg, null]);\n\n    })\n    .catch(function(err) {\n        // --- Failure ---\n        // node.error can take the msg object as the second argument\n        node.error(\"Error parsing PDF: \" + err.message, msg);\n\n        // Prepare the output message for the error path (Output 2)\n        msg.error = err; // Add the error object\n\n        // Send the original message (with error info) to the second output\n        node.send([null, msg]);\n    });\n\n// Return null because the work is asynchronous and handled by node.send()\nreturn null;",
        "outputs": 2,
        "timeout": 0,
        "noerr": 0,
        "initialize": "",
        "finalize": "",
        "libs": [
            {
                "var": "pdfParse",
                "module": "pdf-parse"
            }
        ],
        "x": 610,
        "y": 400,
        "wires": [
            [
                "1308ad842a729aae"
            ],
            [
                "dd667b4ee49e474c"
            ]
        ]
    },
    {
        "id": "dd667b4ee49e474c",
        "type": "debug",
        "z": "e93f3133440568a7",
        "name": "FAIL",
        "active": true,
        "tosidebar": true,
        "console": false,
        "tostatus": false,
        "complete": "true",
        "targetType": "full",
        "statusVal": "",
        "statusType": "auto",
        "x": 770,
        "y": 420,
        "wires": []
    },
    {
        "id": "d649e707401cb0a7",
        "type": "debug",
        "z": "e93f3133440568a7",
        "name": "RAW UPLOAD",
        "active": true,
        "tosidebar": true,
        "console": false,
        "tostatus": false,
        "complete": "true",
        "targetType": "full",
        "statusVal": "",
        "statusType": "auto",
        "x": 780,
        "y": 340,
        "wires": []
    },
    {
        "id": "01fb87c40c24d01c",
        "type": "comment",
        "z": "e93f3133440568a7",
        "name": "UI Element, kann ueber die einstellungen optisch angepasst werden",
        "info": "UI Element,",
        "x": 280,
        "y": 340,
        "wires": []
    },
    {
        "id": "8948f2c8ad9f604f",
        "type": "ui-group",
        "name": "Group 1",
        "page": "afdf2c859e3351de",
        "width": 6,
        "height": 1,
        "order": 1,
        "showTitle": true,
        "className": "",
        "visible": true,
        "disabled": false,
        "groupType": "default"
    },
    {
        "id": "afdf2c859e3351de",
        "type": "ui-page",
        "name": "Page 1",
        "ui": "c44752c1f7bc883a",
        "path": "/page1",
        "icon": "home",
        "layout": "grid",
        "theme": "91ce8000ccb00b38",
        "breakpoints": [
            {
                "name": "Default",
                "px": 0,
                "cols": 3
            },
            {
                "name": "Tablet",
                "px": 576,
                "cols": 6
            },
            {
                "name": "Small Desktop",
                "px": 768,
                "cols": 9
            },
            {
                "name": "Desktop",
                "px": 1024,
                "cols": 12
            }
        ],
        "order": 1,
        "className": "",
        "visible": "true",
        "disabled": "false"
    },
    {
        "id": "c44752c1f7bc883a",
        "type": "ui-base",
        "name": "My Dashboard",
        "path": "/dashboard",
        "appIcon": "",
        "includeClientData": true,
        "acceptsClientConfig": [
            "ui-notification",
            "ui-control"
        ],
        "showPathInSidebar": false,
        "headerContent": "page",
        "navigationStyle": "default",
        "titleBarStyle": "default",
        "showReconnectNotification": true,
        "notificationDisplayTime": 1,
        "showDisconnectNotification": true,
        "allowInstall": true
    },
    {
        "id": "91ce8000ccb00b38",
        "type": "ui-theme",
        "name": "Default Theme",
        "colors": {
            "surface": "#ffffff",
            "primary": "#0094CE",
            "bgPage": "#eeeeee",
            "groupBg": "#ffffff",
            "groupOutline": "#cccccc"
        },
        "sizes": {
            "density": "default",
            "pagePadding": "12px",
            "groupGap": "12px",
            "groupBorderRadius": "4px",
            "widgetGap": "12px"
        }
    }
]