Configuration

vyuh_docx is headless, so it has no global app configuration. Configuration is local to the reader, builder, exporter, layout, and diagnostic APIs that need it.

PDF Export Configuration

final exporter = PdfExporter(
  pageWidth: 612,
  pageHeight: 792,
  marginTop: 72,
  marginBottom: 72,
  marginLeft: 72,
  marginRight: 72,
  fontSize: 11,
  compressContent: true,
  watermark: 'CONTROLLED',
  watermarkColorHex: 'D9D9D9',
  watermarkOpacity: 0.18,
  creationDate: trustedIssuedAt,
  strictFonts: true,
);

exporter.registerFont('Arial', arialBytes);
final bytes = exporter.exportToBytes(document);

Option	Default	Purpose
pageWidth	612.0	Default page width in points. Used when the document section does not override page geometry.
pageHeight	792.0	Default page height in points.
marginTop	72.0	Default top margin in points.
marginBottom	72.0	Default bottom margin in points.
marginLeft	72.0	Default left margin in points.
marginRight	72.0	Default right margin in points.
fontSize	11	Base font size used when the document has no more specific style.
compressContent	true	Compress PDF content streams. Disable for raw PDF debugging.
watermark	null	Optional diagonal watermark across every page.
watermarkColorHex	D9D9D9	Watermark fill color without #.
watermarkOpacity	0.18	Watermark alpha from 0 to 1.
creationDate	wall clock	Fixed PDF creation time for deterministic output.
strictFonts	false	Throw if text cannot be represented by registered fonts.
registerFont	none	Register host-supplied font bytes by font family name.

Layout Engine Configuration

Most callers use PdfExporter, which creates PdfLayoutEngine from section data. Use the layout engine directly for diagnostics or custom rendering.

final layout = PdfLayoutEngine(
  pageWidth: 612,
  pageHeight: 792,
  marginTop: 72,
  marginBottom: 72,
  marginLeft: 72,
  marginRight: 72,
).buildLayoutDocument(document.elements);

The exporter also passes section-derived configuration such as header reserve, footer reserve, columns, document grid line pitch, and default tab stop.

Reader Configuration

Readers are intentionally low ceremony:

Reader	Configuration model
DocxReader.load(path)	File path only. Reads the DOCX package and resolves relationships, styles, numbering, sections, comments, revisions, and media.
DocxReader.loadFromBytes(bytes)	Byte payload only. Use this for server, browser, storage, or upload flows.
SfdtReader.read(json)	SFDT JSON string.
SfdtReader.readMap(map)	Already-decoded SFDT map.
HtmlParser / MarkdownParser	Parser object plus source content.

Builder Configuration

The builder API configures the document through fluent calls and ends with build():

final document = docx()
  .h1('Procedure')
  .p('Use the approved equipment.')
  .table(rows: [
    ['Step', 'Owner'],
    ['Review', 'QA'],
  ])
  .build();

For lower-level generation, use DocxDocumentBuilder and model classes directly when a typed field is needed before a fluent helper exists.

Exporter Configuration

Exporter	Configuration
DocxExporter	Writes the model to DOCX bytes or file. Configuration is mostly document-driven.
SfdtExporter	Writes the model to Syncfusion JSON. Configuration is document-driven.
HtmlExporter	Writes HTML output. Configuration is document-driven.
PdfExporter	Exposes page defaults, compression, watermark, timestamp, and font strictness.

Deterministic Review Setup

For review tests and controlled rendering, prefer:

final exporter = PdfExporter(
  creationDate: DateTime.utc(2026, 6, 20),
  strictFonts: true,
  compressContent: false,
);

compressContent: false makes PDF content streams easier to inspect in tests. creationDate removes wall-clock drift. strictFonts prevents silent glyph loss.