Extracts the page's text: every glyph its content shows, with its Unicode text, position, font, size and rendering mode.
public TextPage ExtractText()
ISO 32000-2 §9.4.4, §9.10, §14.8.2.6.2, §14.8.2.5 and §14.9; see TextExtractor. A tagged page reads in
logical content order, the order of its structure tree, with structure elements' ActualText and Lang applied
(Structure); other pages in content order. Text hidden by optional content and artifacts
are left out; invisible text and text replaced by ActualText are extracted and flagged (TextGlyphFlags).
A structure element whose content spans pages gives its replacement once, on the page where it shows its first glyph in
logical order, so pages read one after another hold it once, in whatever order they are read; on its other pages its glyphs
are still returned, in order, with their element and replacement flag, but each with an empty text range, so they add no
text, words or lines. When the element's content starts with items that show no glyph (an image) on an earlier page, that
page is extracted once per document to find out (in strict mode its deviations then throw here too). Positions are in
default user space. Annotation appearances are not part of the page's content and are not extracted.