Add option to print only the first flat on each line

[davidweichiang]

Illustrates gregorio-project/gregorio#157.
Goes with gregorio-project/gregorio#1607.

[rpspringuel]

If I'm understanding the Dominican convention correctly, then the the notes at the g position (staff space in the middle) are supposed to be ta's not ti's after a flat appears on a line through the end of the line. Applying this convention should result in the following in the left-hand score:

1. A flat appears in the first syllable, so the punctum inclinatum in the first syllable is ta and the flat in the syllable "Ma-" at the end of the line is redundant (though perhaps desired, given how far it is from the initial flat).
2. The first ti/ta note is in "-a" and it is flatted (a ta). From there we should have a ta in "por-" and "ma-" without needing to show the flat (as is the case).
3. The initial punctum ( "-la") and the punctum inclinatum in"ma-" are a ti, as we haven't encountered a flat yet on this line. "-cur-" contains a flat, so its note is a ta, as are the notes in "-re", "ca-", "qui", and "-rat". None of these later syllables need the flat (as shown).
4. The first note in the ti/ta position is the ta in "na-". From there we have ta's in "-tu-" and "mi-" (without flat signs, as shown).
5. The first note in the ti/ta position in the ta in "Ge-". From there we should have ta's in "Vir-" and "pri-" without showing a flat (as is shown, and mentioned in the test note).
6. We have a ti in "-ri-", but then an explicit ta in "-e-". From there "ab", "o-" (twice), "il-", and "A-" are all ta's without need of a flat (as shown).
7. "-ca-" is a ti and "mi-" is an explicit ta.

Applying the same rules to the right-hand score and organizing everything into a table gives me the following:

Syllable	Line in RH score	RH ti/ta	Line in LH score	LH ti/ta
Al-	1	ta (explicit)	1	ta (explicit)
Al-	1	ta	1	ta
Ma-	1	ta (explicit)	1	ta (explicit)
-a	2	ta (explicit)	2	ta (explicit)
por-	2	ta	2	ta
ma-	2	ta	2	ta
-la	3	ti	2	ta
ma-	3	ti	2	ta
-cur-	3	ta (explicit)	3	ta (explicit)
-re	3	ta	3	ta
ca-	3	ta	3	ta
qui	3	ta	3	ta
-rat	3	ta	3	ta
na-	4	ta (explicit)	4	ta (explicit)
-tu-	4	ta	4	ta
mi-	4	ta	4	ta
Ge-	5	ta (explicit)	4	ta (explicit)
Vir-	5	ta	5	ti
pri-	5	ta	5	ti
-ri-	6	ti	5	ti
-e-	6	ta (explicit)	5	ta (explicit)
ab	6	ta	6	ti
o-	6	ta	6	ti
o-	6	ta	6	ti
il-	6	ta	6	ti
A-	6	ta	6	ti
-ca-	7	ti	6	ti
mi-	7	ta (explicit)	6	ta (explicit)

Have I understood and applied the Dominican rule appropriately? If so, this is a good test with several syllables showing a discrepancy between the two scores. What would be most useful is a list of the "correct" choice. The gabc should show this (by explicitly including the gx where it would be needed in each syllable if that syllable started a line), but it might be useful to call that out in the test text or a comment just so people realize what's up when they come upon this test without the current context.

[davidweichiang]

That's right. I've updated the test to include the score as it was originally printed.

I am not sure why the Ma in Mater has an explicit flat. As I understand it, it wasn't really needed.

[rpspringuel]

Okay, I understand the test now and think it looks good. My only remaining quibble is with calling the note "B" since square notes are technically a moveable do system and thus don't map to exact pitches. But as I said, it's a quibble and doesn't really affect the understanding of the tests so I wouldn't change it.

This test is a bit different in that one document shows both failing test and the passing test, but that's because the underlying gabc is different in these two cases. I think that's a good reason for leaving this as is.