29.1 The dense variable-length token model

In the dense variable-length model of input, one or more successful calls of marpa_r_alternative() must be immediately previous to every call to marpa_r_earleme_complete(). Note that, for a variable-length input model to be “dense” according to this definition, at least one successful call of marpa_r_alternative() must be immediately previous to each call to marpa_r_earleme_complete(). Recall that, in this document, we say that a marpa_r_alternative() call is “immediately previous” to a marpa_r_earleme_complete() call iff that marpa_r_earleme_complete() call is the first marpa_r_earleme_complete() call after the marpa_r_alternative() call.

In the dense model of input, after a successful call of marpa_r_alternative(), the earleme variables are as follows:

• The furthest earleme will be max(old_f, old_c+length),
  • where old_f is the furthest earleme before the call to marpa_r_alternative(),
  • old_c is the value of the current earleme before the call to marpa_r_alternative(), and
  • length is the length of the token read.
• marpa_r_alternative() never changes the latest or current earleme.

In the dense variable-length model of input, the following are also true:

• The effect of the marpa_r_earleme_complete() mutator on the earleme variables is the same as for the basic models of input. See The standard model of input.
• the latest earleme is always the same as the current earleme. In fact, the latest earleme and the current earleme are always the same, except in the fully general model of input.