The library and the binary now deny clippy's unwrap_used, expect_used,
panic, unimplemented, todo and unreachable lints outside unit tests, and
forbid unsafe code. Malformed input must give an error or a warning,
never a panic.
The 36 offending calls are replaced without changing behaviour:
- Byte reads that converted a slice of checked length to an array now
use the `Encoding::*_at` decoders, `first_chunk`, or a `Cursor`
helper that returns `Result`; the fallbacks are the existing results
for data of the wrong length (`Value::Other`, a skipped value, a
short-data error).
- The fixed-offset readers of the header and the master pages build the
array with `array::from_fn`; they index as before.
- `Node::set` returns the node it set, so `TextFramePreference` insets
no longer look it up again with `expect`.
- The embedded value files fall back to no values if one fails to
parse, and a block with a version bound that is not a number does not
apply. A new unit test checks that every committed file parses and
that every bound is a number.
- `Xml::end` without an open element fails a debug assertion and
writes nothing in a release build.
diff_outputs.py over the corpus (without own/): 0 differing outputs,
0 changed failures, 0 changed warnings.
Idea from storytold/designcraft#98.
Groups that contain themselves or nest over 100 deep, style group cycles,
table counts larger than the cell data, slot numbers outside their page,
cyclic slotted-record continuations and non-finite numbers now give a
warning or an error instead of a crash or invalid output. Corpus output
is unchanged (tools/diff_outputs.py).
- Bullet list: the font style can be followed by 1 and four bytes.
- InDesign 2.0: in-object strings start with 1, not 2.
- InDesign 2.0-5.0: XML tag names are a flag byte and a string, in
either byte order.
- InDesign 3.0 and 4.0: the document's hyperlink lists start at offset
0 and include page destinations.
- XMP: the packet can be shorter than its object; take its length.
- InDesign 1.x files (no signature, DOCUMENT at 0x5C) are reported as
not supported; files whose page directory lies beyond the database
pages as having no object database.
- Schema: leave out hyperlink sources that hold an anchored object, and
the characters U+FFFE and U+FFFF, which XML does not allow.
tests/corpus.rs accepts the two rejections above and skips the paths
listed in the local corpus/exclude.txt.
The byte order flag applies to object data only: database pages stay
little-endian. Object data is decoded in the file's byte order, and in
big-endian data the string tag and the flag byte before it are swapped.
The two big-endian samples are also the only files from before
InDesign 7.0, and some of their layouts differ: page transform and
bounds chunks, font records, XML tag names and nodes, story strands and
40-byte guide records are read; text frame settings, document
preferences, a composite font and a section style are left out with a
warning. The 40-byte guides of an InDesign 7.0 file are now read too.
docs/format/big-endian.md records the evidence. The corpus tests now
cover all byte orders.
Objects are sequences of (u32 ID, u32 length, data) chunks. Add a cursor
for chunk data with the text segment encoding (single-byte and UTF-16
runs) and the in-object string format. `indd dump` prints an object's
chunks.
The master page holds three tree roots and counts: objects, the class of
each UID, and UIDs without a class. Database::class_of returns an
object's class ID, and `indd objects` prints it.
Decode the database pages: page trailers and types, the logical page
table, the object B+ tree keyed by (UID, segment), data pages, slotted
pages and continued records. Database::object returns any object's byte
stream. Add `indd objects` and `indd object` commands.
Every object of all 351 little-endian corpus files reads without an
inconsistency. Big-endian databases report Unsupported for now.
Add a project CLAUDE.md with the clean-room rules.