it is in COMPLETE_MESSAGES, but the untranslated report listed 1,842
status and error messages with no Italian row, so with the interface in
Italian those messages showed in English.
- it.tsv gains 1,802 `@msg` rows. The translations are VectorCraft's
own, written from the English meaning with the catalog's terms, and
await native review. Placeholders, backticked names, command ids,
double-quoted values and bare true, false and null stay as the
English writes them.
- The untranslated report now lists 40 Italian messages. 38 of those
keys get Italian rows from open pull requests 7211, 7214 and 7116,
so they are left out here. Menu strings, `tl!` literals, undo labels
and swatch names are unchanged.
- The Italian row of docs/localization-parity.md gives the catalog's
current counts, and docs/target-app-parity.md has a line for the
change.
Session::replace_document on a tab other than the current one cleared
the bounds of type set in a font named like a composite font and left
that type with estimated bounds until the document came to the front.
- The replacement's type without bounds is measured at once, in the
document's own composite fonts, lent for the measuring. The front
document stays the one whose composite fonts are registered.
- When the document then comes to the front unchanged, its type in its
own composite fonts is not measured again. Type set in a composite
font only another document defines is still measured again, as
missing, and so is type set in a composite font name that another
thread's loan held with a different definition.
The German catalog lacked a row for 1,842 of the status and error
messages that the untranslated report finds in the sources, so the
status bar showed those messages in English.
- de.tsv gains 1,842 `@msg` rows: the errors of commands and their
parameter checks, scripts, file import and export, printing and
plug-ins. The translations are VectorCraft's own, written from the
English meaning with the catalog's conventions and its existing
feature names. Placeholders, parameter names, command ids and
literal values stay as the English writes them.
- The untranslated report now lists no German messages. Menu strings,
`tl!` literals, undo labels and swatch names are unchanged.
- Two of the messages have two placeholders in a row. The catalog does
not load such a row, so those two are still shown in English.
- The German row of docs/localization-parity.md gives the catalog's
current message counts.
Docs: docs/target-app-parity.md, docs/localization-parity.md.
A legacy .ai graph (%_Gs ... %_GS) opened as plain art, with its table,
type, axes and settings left out. A grouped column, stacked column, line
or area graph now keeps them on its group when every record has the
shape and values the version 7 specification's appendix A gives, the
table holds at least one number, the drawn art strokes the plot's left
and bottom edges, and each series group is tagged once as a whole
column, holds only paths and paints them with one fill and one stroke.
The saved art stays as it is on open; Graph Data redraws the graph from
its data afterward, with each series' fill and stroke kept.
Any other graph (pie and scatter graphs, designs, customizations, record
shapes of later versions, tables past 10,000 categories, 256 series or
2,600,000 cells) keeps its art as plain objects with the existing note.
Saving a graph to .ai, in either form, now warns that its data and
settings are not written.
Right-to-left type on a path that overflows searched for the start of its
text that fits with a binary search, which assumed that a longer start
never fits once a shorter one does not. A letter tracked to a negative
advance breaks that: with steps of 6, 6, -3 and 6 points in a 10 point
span, the first three letters fit while the first two do not, and the
search showed only the first letter.
- Without hanging punctuation, and with no spacing or on a path of one
straight segment, every start is measured in O(n log n): each glyph
keeps its place in the whole text's visual order, except the trailing
whitespace and controls that UAX #9 rule L1 sets to the paragraph's
level.
- With hanging punctuation, or spacing on a curve, when a letter has a
negative advance, a start is a candidate only when its glyphs, in its
own visual order and each at the least step it can take, reach no
farther than the path's length. This lower bound holds with negative
advances.
- At most 32 candidates are arranged from the longest. When none of
them fits, a binary search over the rest shows a start that fits,
which is not always the longest.
- With hanging punctuation, or spacing on a curve, text with no negative
advance keeps the binary search.
Right-aligned and centered type on a curve with Spacing was placed using
the width its glyphs have at the start bracket, while the steps were
measured where the line was placed, so a line that fits could end past
the end bracket and lose glyphs. It is now placed using the width of the
steps it has where it is placed.
A path directly in a graph series lost what these commands gave it when
the graph was drawn again, for example after graph.setData:
- While Global Edit is on in an isolated graph, a match that is a series
path, or a series group whose paths take the edit, now has what
changed on its paths stored for their series: the paint, weight,
opacity and blend mode of the topmost visible fill and stroke, and the
object's opacity and blend mode. The edited match is stored by its own
command as before.
- addon.arrange.randomizeOpacity and addon.arrange.sort with apply
"opacity" now store the object opacity, with the blend mode, of the
series paths whose opacity they change. When they change several
paths of one series, the frontmost of them sets what the graph
stores, as for the other commands.
The graph stores them in the same undo step as the edit. The params
text of addon.arrange.randomizeOpacity and addon.arrange.sort,
docs/mcp.md and docs/target-app-parity.md describe the new behavior.
The check after each edit that the selected objects stay within the
canvas measured the whole of every selected path. Moving one anchor of
a 10,000-anchor path measured all 9,999 segments. A path or compound
path whose anchors and handles are all finite and inside the canvas
range now passes without being measured, because its curves stay inside
their control points. A path with a point outside that range is still
measured, so the check accepts and refuses the same edits as before.
path.moveAnchor to the point the anchor is already at, and
path.setHandle to the point the handle is already at, now leave the
document as it is and record no undo step.
vectorcraft_geom::path::measured_segments counts the segments that
PathData::bounds measures, and the new engine tests use it.
Two contours of an area-type frame now count as meeting along a slanted
edge only when their edges lie exactly on one line, checked with an exact
cross product, and the intervals on either side of a seam join only
where both edges cross the middle of the band at the same x, found along
the same line. Before, both checks allowed a few units in the last place
of the coordinates, which near the edge of the canvas (x = 3,999,000)
is wider than a real gap of one to four units, so a line ran across the
gap outside the frame.
An edge divided by an anchor exactly on it still joins, including where
the divided edge's own crossing would round to another x.
A 3D effect on a fill or stroke worked on copies of the object's fills
and strokes, and the copies were looked up among the object's items by
identity, so a live-effect plug-in before that 3D effect was given
`_context.item` "object" and a deferred stage there was listed with no
item. The drawn art, the 3D meshes and Expand as Wireframes now pass the
original item, so the plug-in is given `{"fill": i}` or `{"stroke": i}`
as docs/plugins.md describes. The paints of a type's first run still have
no item.
Select > Start Global Edit was enabled for any plain group, so a group
holding type, an image or nothing was offered in the Select menu and the
Properties panel's Quick Actions, the Control bar showed its Global Edit
Options, and starting then refused it. The enabled check now applies the
same support check as starting to every selected object, a group's
contents included.
The answer is kept per document revision and selected object, so the
Control bar, which asks every frame, walks a group once per change.
Docs: docs/mcp.md, docs/target-app-parity.md (a new line; the existing
Global Edit line no longer says the menu item checks only for a group).
A document's composite fonts were lent by writing them into the font
database's shared map, and a second lend of another definition of the same
name was skipped while the first was held. A document drawn in a tiled or
floating window while the front document drew, a document placed in one
that defines the name differently, and an export made while another
document is in front could draw their type in the other document's
definition. Registering the front document's fonts during a render could
also change the definition mid-render.
Lending now applies on the lending thread only: each guard pushes the
document's definitions onto a per-thread stack that lookups read before the
database's own fonts, latest loan first, and removes them when it drops,
also when the thread unwinds. The guard is neither Send nor Sync. Lending no
longer changes the database or its generation; while a thread lends
definitions other than the database's, FontDb::generation gives a number of
its own for them, so kept layouts and render caches made in one document's
definitions are not used for another's.
Delete Artboards, Delete Empty Artboards and Cut Artboard removed the
chosen artboards one at a time, walking every ruler guide and the layer
tree once per artboard. They now share one helper that removes the
artboards in one walk and their guides in one more, leaving the same
artboards, guides and selected guides as before.
Empty Artboards in Artwork Cleanup had its own copy of the one-pass
removal. It now calls the same helper, and its test reads that helper's
walk counter.
264 Hindi rows kept English words in a Hindi sentence ("or", "give",
"see"), dropped the word that names what is wrong ("unknown align" read
as "unknown `left`"), reversed a condition ("not editing an opacity
mask" read as "do not edit an opacity mask"), translated enum values the
command reads in English, broke a list of accepted values
("(vertical|horizontal|free)" read as "(vertical Chhorizontal
Chlorfree)"), or repeated a parameter as a "size: आकार" prefix. Each row
now says what its English says, with parameter names, enum values,
placeholders and value lists unchanged, and "positive" reads धनात्मक.
Two tests check every row: no English connecting word outside
backticks, quotes, braces and brackets, and every `a|b|c` list of an
`@msg` row kept as written.
The help page says a wrap object's Offset may be positive or negative
(https://help.adobe.com/archive/en_US/illustrator/cs4/illustrator_cs4_help.pdf#page=307).
Text Wrap Options, the commands, the MCP tool and .ai import already
kept a negative value, but layout set it to 0.
- Layout keeps the signed offset, within -1000 to 1000 pt like the
commands and .ai import (a value that is not a number counts as 0).
- Around a wrap object, a negative offset shrinks the object by that
distance from each edge, holes included, so lines beside, above and
below it run that far into it. A shape narrower than twice the
distance no longer moves text aside. A line may get less room than
the exact shrunk shape leaves, for example where several edges come
close to one piece of the band, or where a band holds more than 64
anchor heights. It never gets less room than an offset of 0 leaves.
- With Invert Wrap, a negative offset keeps the text that far inside
the object.
- Zero and positive offsets up to 1000 pt lay out as before. A larger
positive offset lays out as 1000 pt.
- The params text of object.textWrap.make and object.textWrap.options
and the MCP text_wrap offset description give the range and what a
negative value does.
Docs: docs/target-app-parity.md.
An object in an opacity mask's art is written with its art style, as on
the layers, so a mask whose art has Offset Path opens from the editing
data with the effect live and draws the masked object as it did. This
covers a hidden art root and masks inside another mask's art.
Geometry effects on mask art that no art style holds, such as Roughen,
are applied to the art written, in the masks of layers, symbols and
pattern tiles. Before, the mask's art was written without them.
On a canvas narrower than one row of the bar, the action buttons ran
past its right edge: only the font controls moved to a second row, and
keep_in moved the bar without shrinking it.
- The actions now wrap, in order, onto rows under the row with the
handle, the chips and More Options. Their widths are measured with
the button's font, and the bar's place on the canvas uses the same
layout from the first frame and after a resize.
- An action wider than a row, or one that would need a row past the
canvas's height, is listed in More Options.
- When the handle, the chips and More Options don't fit one row, the
chips take rows after the actions. Without a row to spare, the font
controls are left out. A canvas too small for the handle and More
Options shows no bar.
Docs: docs/target-app-parity.md.
With downsampling on, each piece of an image bent by a Warp effect was
resampled for its own placed size, so a warp wrote one image XObject for
each size a piece was placed at (four for a 4 x 4 warp that stretches
across, eight for a colorized one with its soft masks). The pieces are
now resampled for the widest and the highest size any piece places the
image at, and all paint that one image. Pieces off the page count too,
and another placement of the same image keeps its own size.
Without downsampling, a colorized image and a CMYK image with a grid
mask no longer key their cache by placed size, so each is decoded and
encoded once for all the sizes it is placed at, as plain images are.
When a recovery copy could not be written, the status bar said only
"Couldn't save recovery data: <error>", and a copy that failed on the
background worker showed as "Saving recovery data for <title> failed:
<error>". Neither said which folder was in use or how to pick another.
The message keeps the error text (and the document's title when one
copy fails), then names the folder copies go to, says whether it is
the default recovery folder or the folder set in Preferences, and
names the setting that picks another folder: Preferences → File
Handling → Data Recovery → Folder (recoveryFolder). The recovery code
writes the same line to the log. The folder is the one in use when the
round started, so a preference changed while a copy runs does not
change the folder the message names.
A copy that fails now reports its own message, so the job's result is
Ok rather than an error, and the job no longer adds "<label> failed:
<error>" over it. A store the host installed (browser storage) has no
folder to name: its failed copy shows "Saving recovery data for
<title> failed: <error>", as a copy on the worker did before, also
when the copy was written without the worker.
- engine: recovery::folder returns the folder in use and whether the
recoveryFolder preference names it; recovery::store uses it.
- Four @msg rows in each complete-message catalog, and in the bg, bs,
da and id catalogs, which translate the old message.
- Control replies to a client confined to read roots show a recovery
folder outside them, and the paths under it, by the folder's name, as
they already show other paths outside the roots. That covers the
folder in use and the last eight folders a failure message named.
- docs/target-app-parity.md: a line next to the Data Recovery Folder
preference.
BFlat