Granular markup of fragmented verse lines not supported by API when counting lines

[ingoboerner]

I just discovered this when relaxing closed attribute value lists for #67:
In the legacy ODD I documented the use of the attribute @part (of att.fragmentable but only included the values "I","M","F".)

We have an example from RusDraCor in which a verse line is split across the speeches of several characters:

<sp who="#TretijMistik">
                                  <speaker>Третий мистик</speaker>
                                  <l part="F">Наступит событие.</l>
                                 </sp>
                                 <sp who="#Pero">
                                  <speaker>Пьеро</speaker>
                                  <l>О, вечный ужас, вечный мрак!</l>
                                 </sp>
                                 <sp who="#PervyjMistik">
                                  <speaker>Первый мистик</speaker>
                                  <l part="I">Ты ждешь?</l>
                                 </sp>
                                 <sp who="#VtorojMistik">
                                  <speaker>Второй мистик</speaker>
                                  <l part="M">Я жду.</l>
                                 </sp>
                                 <sp who="#TretijMistik">
                                  <speaker>Третий мистик</speaker>
                                  <l part="F">Уж близко прибытие:</l>
                                  <l>За окном нам ветер подал знак.</l>
                                 </sp>

Not sure if this is actually the right usage of the @part because then we would consider that the "parts" identified with the attribute @part belong to a single verse line:

"Ты ждешь? Я жду. Уж близко прибытие:"

Not sure if this makes sense in this case.

From the Guidelines on the use of F,M,I: See this https://www.tei-c.org/release/doc/tei-p5-doc/en/html/CO.html#COVE

In this encoding, the part attribute is used, as with div, to indicate that the last two l elements should be regarded as the initial and final parts of a single line, rather than as two lines.

My point is not so much about our example not really showing the right usage of the @part attribute, but rather about the problem when we introduce a means of marking up fragmented verse lines:

The API does return a number of verse lines per play and per corpus. I suppose the algorithm does not take the usage of @part into account. So if we take the Guidelines seriously, consecutive verse lines having a @part attribute should be counted as a single verse line, but I think they will be counted individually. I will check the code at some later stage when I get to the features anyways.

IMHO it doesn't make sense to change anything now (apart from the example, if there are better ones), but at some point we either need to decide how we go about this counting. An option could be to leave everything as is but just document that we can not take fragmented verses into account, but doing so without saying is probably not the best idea.