GetXRefDiagnostics

Document management, Diagnostics, Recovery

Description

Returns a JSON report for the cross-reference and recovery path used to load the selected document

Function TPDFlib.GetXRefDiagnostics: WideString;

Empty string	No loaded document is selected and `LastErrorCode` is set to 401
JSON report	The report was returned

File positions	`fileSize`, `headerOffset`, `startXRefTokenOffset`, `declaredStartXRef`, `effectiveStartXRef`, `eofOffset` and `trailingBytes`
Cross-reference summary	`policy` is `strict`, `compatible` or `recovery`; `xrefType` is `table`, `stream`, `hybrid`, `mixed` or `unknown`; `hasXRefStream`, `hybrid`, `shiftedHeader`, `sectionCount` and `trailerCount` describe the selected chain
Entry counts	`entries` reports allocated slots and free, in-use, compressed and invalid entries
Decisions	`decisions` reports selected chain types, followed links, compatibility repairs and rejected sections in evaluation order
Sections	Each item reports declared and physical offsets, table or stream type, `startxref`, `prev` or `xrefstm` source, inclusive trailer dictionary boundaries, linked offsets, parse attempt, `compatibilityFlags` and success state
Recovery	`recovery` reports strict, repair or best-effort mode, whether fallback was attempted and succeeded, recovered object counts, the trailer source and ordered decisions

All byte positions are zero-based; unavailable positions are reported as -1

A shifted-header retry remains visible as attempt 2 so the declared and effective positions can be compared without hiding the failed first attempt

Recovery decisions can include raw-object-scan, recover-object-stream-members and synthesize-trailer