format 2.1: the poem, and a note mark of a symbol and a number

From format 2.1 a poem comes before its verses, carries object number
0 and states the range of its verses, where it came after them sharing
the last verse's number. The text writer's rule, an object with no
text writes nothing, covers both forms and is unchanged; its comment,
in both languages, and the README now describe the poem as it is.

Spine now writes the markup's editor's notes, ~[* ]~ and ~[+ ]~, as
their own numbered series, marked *1, *2 ... and +1, +2 ..., in the
form the special notes already had. The note patterns took symbols
alone, so in every writer, both languages, these went through as text.
They now take [*+]+[0-9]* where they took symbols.

Over the markup document samples written by spine at 2.1: D == Gleam
byte for byte in all five formats; the same .typ from .ssp and
.ocda.db; spine's .typ body == this writer's; epubcheck clean; every
html link resolves. Against the same writers over 2.0 artefacts, two
things move, both the abstraction's doing: the_public_domain's html
and epub gain note 84, which 2.0 never gathered; and seven epubs lose
an empty body.xhtml page listed in the navigation, which was where the
2.0 poems went for want of a segment.

The repository is sisudoc-ocda-writer, in the singular: named for its
role, as its siblings are. It holds one writer, ocda-write, in each
language, taking the format as an option.

flake.lock, in dlang/ and gleam/, now pins the library at its
model 2.1 commit.

(assisted by Claude-Code)

1 files changed, 1 insertions, 1 deletions

diff --git a/dlang/src/write/rgx.d b/dlang/src/write/rgx.d
index 23c848d..f32df85 100644
--- a/dlang/src/write/rgx.d
+++ b/dlang/src/write/rgx.d
@@ -44,7 +44,7 @@ static struct Rgx {
   static inline_notes_al_regular_number_note = ctRegex!(`【(?P<num>\d+)\s+(?P<note>.+?)\s*】`, "mg");
   /+ ↓ and the same for a note marked with a run of stars or pluses
        instead of a number, which the markup also allows +/
-  static inline_notes_al_all_note = ctRegex!(`【(?P<num>\d+|(?:[*]|[+])+)\s+(?P<note>.+?)\s*】`, "mg");
+  static inline_notes_al_all_note = ctRegex!(`【(?P<num>\d+|[*+]+[0-9]*)\s+(?P<note>.+?)\s*】`, "mg");
 
   /+ ↓ tables: the object's text is rows and columns in one string +/
   static table_delimiter_row = ctRegex!("[ ]*\n", "mg");