Sorry, but when we speak of "syntax highlightning" for natural language what is the actual goal?
Are we trying to color differently verbs, nouns, adverbs and prepositions? So that the "goal" is to properly decide if "lie" is a verb or a noun?
Or are we trying to colour subject, verb, object and other elements of the sentence? So that "Rome destroyed Carthago" will have a different color for Rome then the sentence "Hannibal tried to destroy Rome"?
In general, code has "reserved words" and the rest is either a "name" (variables, constants, literals) or a ... Well... "Verb". Like functions, procedures, methods.
In some rare case (function pointers, closures) you have "verbs" that can be used as "nouns" but you completely lack concepts like dative, accusative and so on.
I think that this really breaks down as an analogy when you try to adapt syntax parsing to natural language.
Are we trying to color differently verbs, nouns, adverbs and prepositions? So that the "goal" is to properly decide if "lie" is a verb or a noun?
Or are we trying to colour subject, verb, object and other elements of the sentence? So that "Rome destroyed Carthago" will have a different color for Rome then the sentence "Hannibal tried to destroy Rome"?
In general, code has "reserved words" and the rest is either a "name" (variables, constants, literals) or a ... Well... "Verb". Like functions, procedures, methods.
In some rare case (function pointers, closures) you have "verbs" that can be used as "nouns" but you completely lack concepts like dative, accusative and so on.
I think that this really breaks down as an analogy when you try to adapt syntax parsing to natural language.