Zig Requires Chilkat v11.7.0+

Get a PDF Field Value by its Label (Value to the Right)

See more PDF Text Extraction Examples

Demonstrates the "label" option, which captures the text that belongs to a caption such as "Invoice Number:" without needing coordinates. With labelValue "right", the value is the rest of the label's line. Label matching ignores case, extra spaces and a trailing colon. Zig

const std = @import("std");
const chilkat = @import("chilkat");

pub fn main(init: std.process.Init) !void {
    const alloc = init.arena.allocator();

    // This example requires the Chilkat API to have been previously unlocked.
    // See Global Unlock Sample for sample code.

    // The invoice used in this example is available at invoice_labels.pdf

    const pdf = try chilkat.Pdf.init();
    defer pdf.deinit();
    pdf.loadFile("qa_data/pdf/invoice_labels.pdf") catch {
        std.debug.print("{s}\n", .{try pdf.getLastErrorText(alloc)});
        return;
    };

    const json = try chilkat.JsonObject.init();
    defer json.deinit();
    const json_out = try chilkat.JsonObject.init();
    defer json_out.deinit();

    // The invoice has lines such as "Invoice Number:   INV-2026-0412".
    // The label option restricts the "text" operation to the text belonging to the label.
    json.updateString("label", "Invoice Number") catch {};

    // "right" means the value is the rest of the label's line.
    // (The default, "auto", uses the rest of the line if there is text there, otherwise the block below.)
    json.updateString("labelValue", "right") catch {};

    pdf.extractText("text", json, json_out) catch {
        std.debug.print("{s}\n", .{try pdf.getLastErrorText(alloc)});
        return;
    };

    // The value is returned as the text.
    std.debug.print("Invoice Number: {s}\n", .{try json_out.stringOf(alloc, "text")});

    // The labelMatches array tells where the label was found and how it appeared in the PDF.
    std.debug.print("Found on page {d} as \"{s}\"\n", .{ json_out.intOf("labelMatches[0].page"), try json_out.stringOf(alloc, "labelMatches[0].label") });

    // Matching ignores case and the trailing colon, so these all work:
    json.updateString("label", "CUSTOMER:") catch {};
    pdf.extractText("text", json, json_out) catch {
        std.debug.print("{s}\n", .{try pdf.getLastErrorText(alloc)});
        return;
    };

    std.debug.print("Customer: {s}\n", .{try json_out.stringOf(alloc, "text")});

    json.updateString("label", "due date") catch {};
    pdf.extractText("text", json, json_out) catch {
        std.debug.print("{s}\n", .{try pdf.getLastErrorText(alloc)});
        return;
    };

    std.debug.print("Due Date: {s}\n", .{try json_out.stringOf(alloc, "text")});

    // The totals at the foot of the invoice are right-aligned labels with the amount beside them.
    json.updateString("label", "Total Due") catch {};
    pdf.extractText("text", json, json_out) catch {
        std.debug.print("{s}\n", .{try pdf.getLastErrorText(alloc)});
        return;
    };

    std.debug.print("Total Due: {s}\n", .{try json_out.stringOf(alloc, "text")});

    // A label that is not found is not an error.  The text is empty, labelMatches is empty,
    // and the quality report has an issue with the code "labelNotFound".
    json.updateString("label", "Tracking Number") catch {};
    pdf.extractText("text", json, json_out) catch {
        std.debug.print("{s}\n", .{try pdf.getLastErrorText(alloc)});
        return;
    };

    if (json_out.sizeOfArray("labelMatches") == 0) {
        std.debug.print("Tracking Number: (not found) {s}\n", .{try json_out.stringOf(alloc, "quality.pages[0].issues[0].code")});
    }

    // Sample output:

    // Invoice Number: INV-2026-0412
    // Found on page 1 as "Invoice Number:"
    // Customer: Northwind Traders
    // Due Date: April 13, 2026
    // Total Due: $1,234.53
    // Tracking Number: (not found) labelNotFound
}