A graph in a legacy .ai, or in a current file's editing data, is written
as `%_Gs`, its data and settings on `%_` lines, its drawn art in a group
with a `%_... Go` tag after each object, and `%_GS`. The reader stopped
on the graph operators as ones it doesn't know, so such a file did not
open through its layers.
The reader now leaves the graph's records out with their operands and
reads the drawn art as usual: the bars, lines and labels come in as the
groups, paths and type they are written as. Paint operators on `%_`
lines between `Gt` and `GT` set nothing. One note per file says the
graphs came in as plain art without their data or settings. A graph
that starts inside another, or doesn't end before its layer does, keeps
the art read so far with a second note. Graph operators outside a graph
stay unknown.
In editing data and legacy .ai files, a lower-case l, c, v or y makes the
anchor the segment ends on smooth and an upper-case L, C, V or Y makes it a
corner. The reader used to guess each anchor's kind from whether its
handles are in line, so a lower-case l always read as a corner and an
upper-case C with handles in line read as smooth. A closed path's first
anchor takes the case of the segment back to it. Points and handles do not
move. Paths, compound paths, clipping paths, guide paths, area type frames
and type-on-a-path paths all keep the kinds.
Add the progress line for the symbol editing follow-ups next to the Symbol
editing mode entry in docs/target-app-parity.md, with the tests that cover
them.
Symbol editing copies the editing layer's art into the symbol before Text Wrap, list numbering and threaded re-flow update that layer, so moving a wrap object in the edit left the symbol, its other instances and the saved file with the old wrap. The edit now syncs the symbol again after those passes, and when that changes the symbol, inline graphics of it are refreshed and the threads holding them re-flow.
The appearance changes of the instances on the layers now follow the edited objects through the same pass as the instances in other symbols, graph designs, pattern tiles and brushes. That pass also reaches instances used as opacity-mask art, which keep their changes on the same objects when an edit reorders them and drop the change to an object the edit deletes.
Entering symbol editing mode from an instance puts a copy of the symbol's art on the editing layer in the instance's place, and the instance stays in the document. The view drew the instance dimmed under the copy, so art with transparency showed darker, and its edges showed a dimmed fringe.
The renderer now leaves that instance out of the screen view while the editing layer is isolated, with its masks and effects, and in Outline view inside a clip group. Exports and thumbnails draw it as before.
Editing a dynamic symbol moved each placed instance's appearance changes with the objects they
name, but an instance stored in the art of another symbol kept its changes at the old paths, so
placing that symbol showed a change on the wrong object. Those instances, opacity masks included,
now follow the edit too and keep only the changes to objects the art still has.
symbol.edit checks the new placement before it ends the symbol being edited, so a call with an out-of-range x or y fails and leaves the current edit open. While a symbol is edited, object.isolate refuses art outside the editing layer, and object.exitIsolation goes back to the editing layer only from a group inside it; from any other isolation it leaves symbol editing mode.
Headless tests cover a double-click on a symbol in Thumbnail View and List View, the panel
menu's Edit Symbol with and without a selected instance, the Control bar's Edit Symbol button,
and the Control bar's Exit Symbol Editing Mode and Exit Isolation Mode buttons inside the
symbol's art. The panel's module comment now says a double-click opens the symbol in symbol
editing mode, and docs/mcp.md lists symbol.edit's ids parameter.
Horizontal Scale and Vertical Scale accept any number above 0 and at most 4000000 %, and Baseline Shift accepts -4000000 to 4000000 pt, the limits the document model keeps a character style in. Character Rotation is stored from -180 to 180 degrees, as text.setFormat stores it. A scale outside its limit shows the new message "{label} must be a number above 0 and at most {max}", translated in every catalog that has Basic Character Formats.
charStyle.new now validates attrs against the target text's attributes, so a
language tag that is not in the list but that the selected run already holds
(a native file can store any tag) is kept. Any other unlisted tag is still an
error.
Character Style Options and New Character Style get an Advanced
Character Formats section with Horizontal Scale, Vertical Scale,
Baseline Shift, Character Rotation and Language. Empty fields stay
unset, an explicit 100%, 0 degrees or English: USA stays distinct from
unset, and Reset Panel on the section clears only these five. Scales
take 1% to 10000% and baseline shift -1296 pt to 1296 pt, as
text.setRangeStyle does; rotation is stored from -180 to 180 degrees.
Language sets the language the characters are shaped in; hyphenation
and spelling do not change.
A field still as it opened keeps its stored value without the range
check, so a style captured from a scaled type object can be renamed.
The same applies to Basic's Size and Leading minimum.
charStyle.edit keeps a language tag that is not in the list when the
style already has it, so a style read from a file with such a tag can
be edited.
The OpenType panel gets a Position menu after Figure, with Default
Position, Superscript, Subscript, Numerator and Denominator. `numr` and
`dnom` join the extra features (the existing bits keep their indices),
so they round-trip through the stored tags and reach the shaper.
text.setStyle and text.setRangeStyle take `featureChanges`, which turns
features on or off on top of each character's own features in one
command and keeps every tag the list does not name. The Figure menu,
the Position menu and the checkboxes send it, so on a selection whose
characters have different features each character keeps its other
features. Reset OpenType Features still sends `features`.
Catalog rows for Position:, Default Position, Numerator and
Denominator; one line each in docs/type-parity.md and
docs/target-app-parity.md.
A layer among the targets already stood only for its art that can be
edited. An object named in ids is now checked the same way: a hidden or
locked object, or one on a locked artboard, is refused with an error
instead of becoming the mask, being masked or getting an empty mask.
object.mesh.create with `at` ignored a failed line placement and still replaced the
path with a bare 1x1 mesh. When the new mesh can't place its lines at the point, the
object is left as it is, and when nothing was converted the command fails with
"point is outside the object" and adds no undo step.
BFlat