calling \GreHyph with an argument

[eroux]

It would be useful sometimes to be able to reduce the size of the hyphen, but depending on the space between the syllables: having a long hyphen when the space is very big, and a shorter one when the space can be shrinked. I think the best way to do that is by

• calling \GreHyph more consistently when inserting an hyphen (sometimes - is used)
• calling \GreHyph with an argument (ignored by default?) being the width between the two syllables

[henryso]

It seems that the \GreHyph vs - dichotomy was done on purpose (there are comments to that effect). Before we start changing it, does anyone remember why that was done?

[eroux]

I'm affraid I don't...

[henryso]

Researching this a bit more, there is an option to make the \GreHyph hyphen protrude when at the end of a line by setting \greseteolhyphen{zero}, so if we change all - to \GreHyph and set that zero option, hyphens within a line will protrude into the next syllable, which is most likely not a desirable result. This also means that gregorio (the executable) emitting \GreHyph for all hyphens entered in gabc is probably wrong. How should we proceed here?

[henryso]

There are 3 types of hyphens:

• a hyphen within a syllable (entered in gabc)
• a hyphen at the end of a syllable, but not at the end of a line
• a hyphen at the end of a line

Either of the latter two may be "forced" by entering a hyphen at the end of a syllable in gabc.

My proposal:

• A hyphen entered in gabc within a syllable will be \GreIntraSyllableHyphen, which will typeset a hyphen, regardless of any other settings
• A hyphen entered in gabc at the end of a syllable will continue to be \GreForceHyphen (as it is now), which will force a hyphen to appear at the end of the current syllable.
• A hyphen at the end of a syllable but not at the end of a line will be \gre@hyphen#1, which will typeset a hyphen that is not affected by the \greseteolhyphen setting. #1 is the amount of space between the syllable texts on either side.
• A hyphen at the end of a syllable will be \gre@eolhyphen, which will typeset a hyphen and be affected by the \greseteolhyphen setting.

Comments?

[eroux]

I agree, but I think the best would be to make something a bit more generic, like a \gre@hyphen with two arguments, the first being the type (among the 4 you described), and the second the space between the texts of this syllable and of the next one... \GreForceHyphen, etc. would call it internally. How does it sound? Also, we must think of a way for users to override the \gre@hyph macro in a user-friendly way...

[henryso]

What you're suggesting doesn't entirely make sense so me. The second argument only makes sense for hyphens at the end of syllables which are not at the end of a line.

What is the second argument for a hyphen that is not at the end of a syllable? 0? The width of a hyphen?

What is the second argument for a hyphen at the end of a line? Infinity? 0?

Also, \GreForceHyphen does not emit a hyphen. It sets a flag that causes the hyphen determination code to always generate a hyphen (either interline end-of-line, depending on the final position).

[eroux]

The second argument would ignored in these cases yes. Ok for \GreForceHyphen, let's ignore it then

[eroux]

let's finish it in 5.0

[henryso]

Is the thing left to do here the override of \gre@hyph in a user-friendly way, or is there something else missing?

[eroux]

I remember something about being able to control the length of the hyphen to put at the end of a line when it's automatically added in Lua... I'm affraid I can't have a clearer answer without digging in the code (which won't be this week I'm affraid...)