THPDFParserRecoveryEvent Record
One structured cross-reference recovery event carrying kind, confidence, byte range and object identity
Declaration
THPDFParserRecoveryEvent= record
Kind: THPDFParserRecoveryEventKind;
Confidence: THPDFParserRecoveryConfidence;
FirstByte: Int64;
EndByte: Int64;
ObjectNumber: Integer;
GenerationNumber: Integer;
Message: AnsiString;
end;Unit
HPDFDoc
Fields
| Field | Purpose |
|---|---|
Kind | What happened: attempt started, limits overstepped, object or xref candidate accepted or rejected, xref section chosen, offsets rebuilt, or recovery succeeded or failed |
Confidence | Recovery confidence grade running from prcNone up to prcVerified |
FirstByte | First byte of the scanned or candidate region, or -1 where the event carries no range |
EndByte | One past the last byte of the region, or -1 when not available |
ObjectNumber | Object number of the candidate involved, or -1 when the event does not concern a specific object |
GenerationNumber | Generation number of the candidate object, or -1 when not available |
Message | Human-readable detail spelling out the decision |
Usage notes
- Events accumulate during a recovery-enabled load; read the lot with
GetLoadedParserRecoveryEventsor one by one withGetLoadedParserRecoveryEvent - When an index is out of range the getter still hands back a defaulted record whose byte and object fields are
-1, so treat negative values as absent - Pair
KindwithConfidenceto judge whether a rebuilt cross-reference came from a verified section or from a low-confidence object scan