Nomad changes

docs/snippets/zig/config/advanced_config.md

@@ -0,0 +1,58 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "use_cache": true,
+        \\  "enable_quality_processing": true,
+        \\  "ocr": {
+        \\    "backend": "tesseract",
+        \\    "language": "eng"
+        \\  },
+        \\  "chunking": {
+        \\    "max_characters": 1000,
+        \\    "overlap": 200,
+        \\    "embedding": {
+        \\      "model": {"type": "preset", "name": "balanced"},
+        \\      "batch_size": 32,
+        \\      "normalize": true
+        \\    }
+        \\  },
+        \\  "language_detection": {
+        \\    "enabled": true,
+        \\    "min_confidence": 0.8,
+        \\    "detect_multiple": false
+        \\  },
+        \\  "keywords": {
+        \\    "algorithm": "yake",
+        \\    "max_keywords": 10,
+        \\    "min_score": 0.1,
+        \\    "ngram_range": [1, 3],
+        \\    "language": "en"
+        \\  },
+        \\  "token_reduction": {
+        \\    "mode": "moderate",
+        \\    "preserve_important_words": true
+        \\  },
+        \\  "postprocessor": {
+        \\    "enabled": true
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/chunking_config.md

@@ -0,0 +1,30 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "chunking": {
+        \\    "max_characters": 1000,
+        \\    "overlap": 200,
+        \\    "chunker_type": "markdown",
+        \\    "prepend_heading_context": true
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/config_basic.md

@@ -0,0 +1,26 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "use_cache": true,
+        \\  "enable_quality_processing": true
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/config_discover.md

@@ -0,0 +1,25 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    // The Zig binding accepts JSON config strings. To use a discovered config
+    // file, load it from disk into a string and pass it through unchanged.
+    const cwd = std.fs.cwd();
+    const config_json = cwd.readFileAlloc(allocator, "kreuzberg.json", 1 << 20) catch |err| switch (err) {
+        error.FileNotFound => try allocator.dupe(u8, "{}"),
+        else => return err,
+    };
+    defer allocator.free(config_json);
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{result_json});
+}
+```

docs/snippets/zig/config/config_ocr.md

@@ -0,0 +1,28 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "ocr": {
+        \\    "backend": "tesseract",
+        \\    "language": "eng"
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("scanned.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/config_programmatic.md

@@ -0,0 +1,39 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    // Build the config JSON programmatically with std.json.
+    var arena = std.heap.ArenaAllocator.init(allocator);
+    defer arena.deinit();
+    const a = arena.allocator();
+
+    var root = std.json.ObjectMap.init(a);
+    try root.put("use_cache", std.json.Value{ .bool = true });
+    try root.put("enable_quality_processing", std.json.Value{ .bool = true });
+
+    var ocr = std.json.ObjectMap.init(a);
+    try ocr.put("backend", std.json.Value{ .string = "tesseract" });
+    try ocr.put("language", std.json.Value{ .string = "eng+deu" });
+    try root.put("ocr", std.json.Value{ .object = ocr });
+
+    var chunking = std.json.ObjectMap.init(a);
+    try chunking.put("max_characters", std.json.Value{ .integer = 1000 });
+    try chunking.put("overlap", std.json.Value{ .integer = 200 });
+    try root.put("chunking", std.json.Value{ .object = chunking });
+
+    const config_value = std.json.Value{ .object = root };
+    var buffer = std.ArrayList(u8).init(a);
+    try std.json.stringify(config_value, .{}, buffer.writer());
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, buffer.items);
+    defer std.heap.c_allocator.free(result_json);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{result_json});
+}
+```

docs/snippets/zig/config/document_structure_config.md

@@ -0,0 +1,25 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "include_document_structure": true
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/element_based_output.md

@@ -0,0 +1,25 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "result_format": "element_based"
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/embedding_config.md

@@ -0,0 +1,34 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "chunking": {
+        \\    "max_characters": 1000,
+        \\    "overlap": 200,
+        \\    "embedding": {
+        \\      "model": {"type": "preset", "name": "balanced"},
+        \\      "batch_size": 16,
+        \\      "normalize": true,
+        \\      "show_download_progress": true
+        \\    }
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/html_output.md

@@ -0,0 +1,29 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "output_format": "html",
+        \\  "html_output": {
+        \\    "theme": "github",
+        \\    "embed_css": true
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/keyword_extraction_config.md

@@ -0,0 +1,31 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "keywords": {
+        \\    "algorithm": "yake",
+        \\    "max_keywords": 10,
+        \\    "min_score": 0.1,
+        \\    "ngram_range": [1, 3],
+        \\    "language": "en"
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/language_detection_config.md

@@ -0,0 +1,29 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "language_detection": {
+        \\    "enabled": true,
+        \\    "min_confidence": 0.8,
+        \\    "detect_multiple": true
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/ocr_dpi_config.md

@@ -0,0 +1,32 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "images": {
+        \\    "extract_images": true,
+        \\    "target_dpi": 300,
+        \\    "max_image_dimension": 4096,
+        \\    "auto_adjust_dpi": true,
+        \\    "min_dpi": 150,
+        \\    "max_dpi": 600
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/pdf_config.md

@@ -0,0 +1,30 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "pdf_options": {
+        \\    "extract_images": true,
+        \\    "passwords": ["password123"],
+        \\    "extract_metadata": true,
+        \\    "extract_annotations": true
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("encrypted.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/pdf_hierarchy_config.md

@@ -0,0 +1,32 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "pdf_options": {
+        \\    "hierarchy": {
+        \\      "enabled": true,
+        \\      "k_clusters": 6,
+        \\      "include_bbox": true,
+        \\      "ocr_coverage_threshold": 0.5
+        \\    }
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/postprocessor_config.md

@@ -0,0 +1,28 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "postprocessor": {
+        \\    "enabled": true,
+        \\    "enabled_processors": ["whitespace_normalizer", "unicode_normalizer"]
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/quality_processing_config.md

@@ -0,0 +1,26 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "enable_quality_processing": true,
+        \\  "use_cache": true
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/tesseract_config.md

@@ -0,0 +1,33 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "ocr": {
+        \\    "backend": "tesseract",
+        \\    "language": "eng+deu",
+        \\    "tesseract_config": {
+        \\      "language": "eng+deu",
+        \\      "psm": 6,
+        \\      "oem": 3
+        \\    }
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("scanned.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```

docs/snippets/zig/config/token_reduction_config.md

@@ -0,0 +1,28 @@
+```zig title="Zig"
+const std = @import("std");
+const kreuzberg = @import("kreuzberg");
+
+pub fn main() !void {
+    var gpa = std.heap.GeneralPurposeAllocator(.{}){};
+    defer _ = gpa.deinit();
+    const allocator = gpa.allocator();
+
+    const config_json =
+        \\{
+        \\  "token_reduction": {
+        \\    "mode": "moderate",
+        \\    "preserve_important_words": true
+        \\  }
+        \\}
+    ;
+
+    const result_json = try kreuzberg.extract_file_sync("document.pdf", null, config_json);
+    defer std.heap.c_allocator.free(result_json);
+
+    const owned = try allocator.dupe(u8, result_json);
+    defer allocator.free(owned);
+
+    const stdout = std.io.getStdOut().writer();
+    try stdout.print("{s}\n", .{owned});
+}
+```