Organization of the concept of "anytext" as special case of the concept of "texty term"
So I was thinking about what the type of a "text thing" could be and this seems appropriate:
Categorization always categorizes a term as not belonging to this tree or else as belonging to exactly one of the leaf types. Types nearer the root type just regroup sutypes in a meaningful way (for the programmer).
+--- var?: an unbound variable; can become anything | | | +--- openvarlist?: nonempty open list of > 0 vars | | +--- openlist? -----+--- opencharylist?: nonempty open list of > 0 char, may contain vars | | | +--- opencodeylist?: nonempty open list of > 0 unicode code points, may contain vars | | +--- varlist?: list of N > 0 vars; edge case | | +--- changylist? ---+--- charylist?: list of C > 0 chars and V > 0 vars | | | +--- codeylist?: list of C> 0 unicode code points and V > 0 vars | | +--- emptylist!: empty list; edge case as not sure whether charlist or codelist | | | +--- textlist! ---+--- charlist!: nonempty list of chars (atoms of length 1) | | | | | +--- codelist!: nonempty list of unicode code points (integers between 0 and 0x10FFFF) | | texty* ---+ +--- text! ---+ | | | | | | +--- atom!: Prolog atoms including the empty atom '' (but not the empty list ) | | | | +---anytext! ---+ +--- stringy! ---+ | | | +--- string!: SWI-Prolog strings including the empty string "" | +--- number! acceptable because a number can be transformed into text (according to some unspecified convention...)
Show me the code you say?
An implementation of the above type tree can be found as pre-release here