PdfDocument class - dart_pdf_editor library

Properties

catalog → CosDictionary: no setter
cos → CosDocument: The underlying COS-level document, for anything not surfaced here yet.
final
hashCode → int: The hash code for this object.
no setterinherited
info → Map<String, String>: Document information dictionary (/Title, /Author, ...) as text.
no setter
pageCount → int: The number of reachable page leaves. Deliberately NOT the root /Count: real-world files lie about it (and fuzzed ones make it a reference to a stream), and an index below pageCount must never make page throw. The walk caches in _leafCache; /Count is still used as the subtree-skipping hint inside page.
no setter
pages → List<PdfPage>: Every reachable page in document order, resolved in one page-tree walk.
no setter
runtimeType → Type: A representation of the runtime type of the object.
no setterinherited
version → String: no setter

Methods

applyIncrementalUpdate(Uint8List newBytes) → Set<int>: Feeds an append-only incremental revision into this open document in place (see CosDocument.applyIncrementalUpdate) and re-walks the page tree, so a render worker can reflect one page's edit without re-parsing the whole document. newBytes must begin with this document's current bytes (cos.bytes). Returns the set of redefined COS object numbers. Throws when the update cannot be applied incrementally; the caller should re-open instead.
invalidatePageCache() → void: Drops cached page-tree state. Editing code calls this after structural changes (reorder, removal, insertion) so lookups re-walk the tree.
noSuchMethod(Invocation invocation) → dynamic: Invoked when a nonexistent method or property is accessed.
inherited
openAppended(Uint8List bytes, {String password = ''}) → PdfDocument: Opens bytes - this document's bytes with incremental updates appended (e.g. a PdfEditor save over it), or the prefix an earlier revision ended at - with this document's already authenticated keys, so a freshly written revision of an encrypted file reopens without its password, and an undo does not re-run the password check. A revision whose /Encrypt differs authenticates password instead. See CosDocument.openAppended.
page(int index) → PdfPage: Returns page index (zero-based), with inheritable attributes (Resources, MediaBox, CropBox, Rotate) resolved along the tree path.
pageIndexOf(CosDictionary pageDict) → int: Zero-based index of a page dictionary, or -1 if it isn't a leaf of this document's page tree. Resolved objects are cached by reference, so identity comparison is sound. Used to resolve link destinations.
rollbackTo(int length) → ({Set<int> changed, int steps})?: Takes this open document back to an earlier revision it advanced past with applyIncrementalUpdate (see CosDocument.rollbackTo) and drops the page-tree caches, so a render worker can follow an undo without re-parsing the document or losing the objects the undone revisions did not touch. length is that revision's byte length; cos.bytes must still hold it as their prefix. Returns the redefined COS object numbers and how many revisions were undone, or null - nothing changed - when length is not a revision this document can roll back to; re-open instead.
toString() → String: A string representation of this object.
inherited
withIncrementalUpdate(Uint8List newBytes) → PdfDocument: Folds an append-only revision into the COS layer (see applyIncrementalUpdate) and returns a fresh PdfDocument over it.

Operators

operator ==(Object other) → bool: The equality operator.
inherited

Static Methods

open(Uint8List bytes, {String password = '', List<int>? populatedRanges}) → PdfDocument: Opens a document. For encrypted files password is tried first as the user and then as the owner password (the empty default is what most owner-locked business documents expect); a wrong password throws CosPasswordException.
openSource(PdfByteSource source, {String password = '', PdfSourceLoadOptions options = const PdfSourceLoadOptions()}) → Future<PdfDocument>: Opens a document from an asynchronous, random-access source - a remote file over HTTP Range requests, a local file read on demand, or any other PdfByteSource. Only the bytes the parser needs (header, cross-reference chain, and the live objects) are fetched, falling back to a full download when the source cannot serve useful ranges or the file is malformed.