Drop the undocumented orientation-swap on moved guides; Alt again previews guide.move with copy, matching new-guide Alt orientation only.
Co-authored-by: Cursor <cursoragent@cursor.com>
Match the public shortcut reference: Alt while dragging converts horizontal and vertical, with snapping on the new axis; guide.move accepts vertical on the index form.
Co-authored-by: Cursor <cursoragent@cursor.com>
PNG, PDF, EPS, legacy `.ai`, packaged zips and saved `.vectorcraft` files
record the wall-clock time they were made in, so the same commands run twice
gave different bytes although the pixels matched (#2880):
PDF /CreationDate and /ModDate D:20261018155816 vs D:20261018155817
PNG the `Creation Time` tEXt entry
.vectorcraft one byte of the stored date
SVG carries no time, and already repeated
`metadata::output_time` is the time a file records: `SOURCE_DATE_EPOCH` when it
holds a non-negative whole number of seconds, else the clock, as the
reproducible-builds specification defines it. A date a command is given
(`file.new {created}`, a save's `modified`) still wins over the variable, and a
blank, negative or non-numeric value is ignored so that an export is never
refused over the environment.
`metadata::now_unix` keeps the real clock and now has exactly two callers, both
of which must not be pinnable from the environment: the recovery heartbeat, and
the seed PDF encryption draws on.
A .ai saved without PDF compatibility has only a placeholder page in
its PDF part. document.open reads such a file from its editing data,
but with `pages` given it failed with "the file holds only a
placeholder page", even for "1" on a file of one page.
When `pages` names every page of the file, document.open now reads
the editing data as it does without `pages`.
Docs: docs/mcp.md.
#2784 made zh-hant a language that translates every status and error message while #2930 added the shared parameter-check errors to the others; main's shared_param_check_errors_are_translated failed on zh-hant.
Preview in Browser wrote every page to one folder,
VectorCraft Save for Web in the system's temporary folder. A second
preview of a document with the same title, from the same app or from
another one running, overwrote the first preview's page, and any image
file of the same name (PNG-8 and PNG-24 both write .png), before the
browser read them. The same overwrite made
dialogs::tests_save_for_web::preview_in_browser_opens_the_page fail now
and then, when another Save for Web test in the same binary, or a test
run in another checkout on the machine, previewed Untitled-1 at the
same time.
Each preview now goes to a new folder inside that one, named after the
process id and a count of the process's previews. A name that an
earlier process with the same id left there is skipped for the next
count. The page keeps its name and its images folder. A preview first
removes the entries of VectorCraft Save for Web that were last changed
an hour ago or earlier, which also clears the files the single folder
held before.
Docs: docs/target-app-parity.md.
Saving a version 7 .ai wrote the title, spot color names and gradient
names as the UTF-8 bytes of their characters: a spot color named Café
was written (Caf\303\251) in the header's custom colors and in each x.
Type and, since #2722, layer names are written in Windows-1252, the
encoding the file installs with TE.
These names now go through #2722's Windows-1252 path, so Café is
written (Caf\351). A character outside Windows-1252 is written as ?,
with one warning for all of these names. A spot color keeps one name
in the header, in each x and X and in gradient stops, and a gradient
keeps one name in its definition and in each Bg. Two spot colors or
two gradients whose names become the same that way are numbered (? 2).
Docs: docs/file-format-parity.md.
An image in a .ai's editing data whose operands had no matrix, fewer
than 11 numbers, or a matrix of other than six numbers was left out
with no warning. So was one whose matrix maps it onto a line or a
point. The reader's warnings did not mention such an image, while the
reader warned "an image VectorCraft can't read was left out" for an
image whose size or bit depth it can't use.
Each of those exits now adds the same warning, as does the one where
the decoded pixels don't make an image.
A PDF or PDF-compatible .ai saved with Preserve Editing carries the
native document as an embedded file, with a hash of its pages at the
end of the file's /Desc. The import read /Desc with from_utf8_lossy and
looked for "hash ". /Desc is a PDF text string, which may be written as
UTF-16BE with a byte order mark (ISO 32000-1 7.9.2.2). Written that
way it holds no UTF-8 "hash ", so the import found no hash, took the
editing data as intact and reopened the native document with no
warning, even after another app had changed the pages.
/Desc is now read with text_string, the importer's decoder for text
strings, so the hash is found in UTF-16BE, UTF-8 and PDFDocEncoding.
Changed pages open as artwork with the "changed in another app"
warning. The embedded file's /UF and /F names are decoded the same
way, so editing data named only in UTF-16BE is found.
A /Desc without a hash is still trusted, as before.
