Wumpus.Core.TextEncoder

Extended character code handling.

Wumpus uses SVG style escaping to embed character codes or names in regular strings:

 "regular ascii text &#egrave; more ascii text"

i.e. character names and codes are delimited by &# on the left and ; on the right.

In Wumpus strings both character names and character codes can be embedded - it seems conventional for PostScript to use names e.g.:

 (myst) show /egrave glyphshow (re) show

... and SVG to use codes, e.g.:

 mystère

To accommodate both Wumpus defines a TextEncoder record which provides a two-way mapping between character codes and glyph names for a character set.

Documentation

type GlyphName = String

type CharCode = Int

type PostScriptLookup = CharCode -> Maybe GlyphName

type SVGLookup = GlyphName -> Maybe CharCode

data TextEncoder

A TextEncoder

An instance needs:

• The functions for looking up codes by glyph-name and glyph-name by code.
• The name of the encoding - this is printed in the xml prologue of the SVG file as the encoding attribute. Latin 1's official name is "ISO-8859-1".
• Fallback glyph-names and char codes in case lookup fails.

Wumpus.Core.TextLatin1 defines an implementation for Latin 1.

Constructors

TextEncoder

Fields

ps_lookup :: PostScriptLookup

svg_lookup :: SVGLookup

svg_encoding_name :: String

ps_fallback :: GlyphName

svg_fallback :: CharCode

Instances

Monad m => ReaderM (PsT m) TextEncoder

Monad m => ReaderM (SvgT m) TextEncoder