Unit: PDFium
Example
Pdf.FileName := 'document.pdf';
Pdf.Password := 'password';
Pdf.Active := True; // open PDF document
Pdf.PageNumber := 1; // select page
ShowMessage(Pdf.Text); // extract text
Types
type TBookmark = record
Handle: FPDF_BOOKMARK;
Title: WString;
PageNumber: Integer;
Destination: TDestination;
DescendantCount: Integer;
IsOpen: Boolean;
Action: TPdfAction;
ActionPageNumber: Integer;
ActionDestination: TDestination;
ActionPath: WString;
HasColor: Boolean;
Color: TColor;
end;
type TBookmarks = array of TBookmark;
Bookmark data
type TDestination = record
Handle: FPDF_DEST;
Name: WString;
PageNumber: Integer;
ViewMode: TPdfDestinationViewMode;
ViewParameterCount: Integer;
ViewParameters: TPdfDestinationViewParameters;
HasX: Boolean;
X: Single;
HasY: Boolean;
Y: Single;
HasZoom: Boolean;
Zoom: Single;
end;
Named destination data
type TPdfDestinationViewMode = (dvmUnknown, dvmXYZ, dvmFit, dvmFitHorizontal, dvmFitVertical, dvmFitRectangle, dvmFitBoundingBox, dvmFitBoundingBoxHorizontal, dvmFitBoundingBoxVertical);
type TPdfDestinationViewParameters = array [0..3] of Single;
Destination view mode and its ordered native parameters
type TLinkAnnotation = record
Handle: FPDF_LINK;
Rectangle: TPdfRectangle;
PageNumber: Integer;
Action: TPdfAction;
ActionPageNumber: Integer;
ActionPath: WString;
Points: array of TQuadrilateralPoint;
end;
type TLinkAnnotations = array of TLinkAnnotation;
type TQuadrilateralPoint = array [1..4] of TPdfPoint;
Link annotation data
type TPdfPermission = (permPrint, permModify, permCopy, permAnnotate, permFillForms, permAccessibility, permAssemble, permPrintHighQuality);
type TPdfPermissions = set of TPdfPermission;
Typed document permission flags decoded from the native permission mask
type TPdfDocumentMetadata = record
Title: WString;
Author: WString;
Subject: WString;
Keywords: WString;
Creator: WString;
Producer: WString;
CreationDate: WString;
ModifiedDate: WString;
end;
One snapshot of the standard document Info dictionary fields
type TObjectType = (otUnknown, otText, otPath, otImage, otShading, otForm);
Determines type of the PDF object
type TPageMode = (pmNone, pmOutline, pmThumbs, pmFullScreen, pmOptionalContentGroup, pmAttachments, pmUnknown);
Determines how the PDF document should be displayed when opened
- pmNone - neither document outline nor thumbnail images visible
- pmOutline - document outline visible
- pmThumbs - thumbnail images visible
- pmFullScreen - full-screen mode, with no menu bar, window controls, or any other window visible
- pmOptionalContentGroup - optional content group panel visible
- pmAttachments - attachments panel visible
- pmUnknown - unknown page mode
type TPdfAction = (acUnsupported, acGoto, acGotoRemote, acUri, acLaunch);
Determines type of an action associated with a bookmark item
- acUnsupported - unsupported action type
- acGoto - go to a destination within current document
- acGotoRemote - go to a destination within another document
- acUri - Universal Resource Identifier, including web pages and other internet based resources
- acLaunch - launch an application or open a file
type TPdfAnnotation = record
Subtype: TPdfAnnotationSubtype;
Flags: TPdfAnnotationFlags;
HasColor: Boolean;
Color: TColor;
ColorAlpha: Byte;
HasInteriorColor: Boolean;
InteriorColor: TColor;
InteriorColorAlpha: Byte;
HasAttachmentPoints: Boolean;
AttachmentPoints: TQuadrilateralPoint;
Rectangle: TPdfRectangle;
ContentsText: WString;
AuthorText: WString;
end;
Annotation data
type TPdfAnnotationFlag = (afInvisible, afHidden, afPrint, afNoZoom, afNoRotate, afNoView, afReadOnly, afLocked, afToggleNoView);
type TPdfAnnotationFlags = set of TPdfAnnotationFlag;
Annotation flags
type TPdfAnnotationSubtype = (anUnknown, anText, anLink, anFreeText, anLine, anSquare, anCircle, anPolygon, anPolyLine, anHighlight, anUnderline, anSquiggly, anStrikeout, anStamp, anCaret, anInk, anPopup, anFileAttachment, anSound, anMovie, anWidget, anScreen, anPrinterMark, anTrapNet, anWaterMark, anThreed, anRichMedia, anXfaWidget, anRedact);
The annotation subtype
type TPdfBlendMode = (bmDefault, bmColor, bmColorBurn, bmColorDodge, bmDarken, bmDifference, bmExclusion, bmHardLight, bmHue, bmLighten, bmLuminosity, bmMultiply, bmNormal, bmOverlay, bmSaturation, bmScreen, bmSoftLight);
Determines how layers are blended into each other
type TPdfDisplayMode = (dmSingleContinuous, dmTwoPageContinuous, dmTwoPageContinuousWithCover);
Determines how the PDF document pages are displayed
- dmSingleContinuous - Display pages continuously in a single column
- dmTwoPageContinuous - Display pages continuously in two columns
- dmTwoPageContinuousWithCover - Display the first page alone, then display the remaining pages continuously in two columns
type TPdfFeature = (feUnknown, feXfaForm, fePortableCollection, feAttachment, feSecurity, feSharedReview, feSharedFormAcrobat, feSharedFormFilesystem, feSharedFormEmail, fe3dAnnotation, feMovieAnnotation, feSoundAnnotation, feScreenMediaAnnotation, feScreenRichMediaAnnotation, feAttachmentAnnotation, feSignatureAnnotation);
Feature of PDF document reported by OnUnsupportedFeature event
type TPdfFillMode = (fmNone, fmAlternate, fmWinding);
Determines how fill areas are formed when path intersects itself
- fmNone - no fill
- fmAlternate - even-odd parity rule
- fmWinding - nonzero winding rule
type TPdfFormFieldFlag = (ffReadOnly, ffRequired, ffNoExport, ffMultiline, ffChoiceCombo, ffChoiceEdit, ffPassword, ffChoiceMultiSelect);
type TPdfFormFieldFlags = set of TPdfFormFieldFlag;
Form field flags
type TPdfFormFieldInfo = record
Name: WString;
AlternateName: WString;
FieldType: TPdfFormFieldType;
Flags: TPdfFormFieldFlags;
OptionLabels: array of WString;
OptionSelected: array of Boolean;
FontSize: Single;
Checked: Boolean;
ExportValue: WString;
GroupCount: Integer;
GroupIndex: Integer;
end;
Form field information
type TPdfFormFieldType = (fiUnknown, fiPushButton, fiCheckBox, fiRadioButton, fiComboBox, fiListBox, fiTextField, fiSignature);
Determines the type of the form field
type TPdfFormType = (ftUnknown, ftNone, ftAcroForm, ftXfaFull, ftXfaForeground);
Determines type of forms in the PDF document
- ftUnknown - unknown form type
- ftNone - no forms in document
- ftAcroForm - forms are specified using AcroForm specification
- ftXfaFull - forms are specified using the entire XFA specification
- ftXfaForeground - forms are specified using the XFAF subset of XFA specification
type TPdfImage = record
Width: Integer;
Height: Integer;
Data: TBytes;
end;
Size in pixels and data of the PDF image object
type TPdfLineCap = (lcDefault, lcButt, lcRound, lcProjectingSquare);
The shape to be used at the ends of open subpaths
type TPdfLineJoin = (ljDefault, ljMiter, ljRound, ljBevel);
The shape to be used at the corners of paths that are stroked
type TPdfPoint = record
X, Y: Double;
end;
Point with X and Y coordinates
type TPdfRectangle = record
Left, Top, Right, Bottom: Double;
end;
type TPdfRectangles = array of TPdfRectangle;
Rectangle area with Left, Top, Right and Bottom coordinates
type TPdfFileIdentifierType = (fidPermanent, fidChanging);
Trailer file identifier selector. fidPermanent reads the first /ID entry and fidChanging reads the second /ID entry
type TPdfPageBox = (pbMedia, pbCrop, pbBleed, pbTrim, pbArt);
Page dictionary box selector used by PageBox, GetPageBox, and SetPageBox
type TPdfJavaScriptAction = record
Name: WString;
Script: WString;
end;
type TPdfJavaScriptActions = array of TPdfJavaScriptAction;
Document-level JavaScript name-tree action data
type TPdfSignature = record
Content: TBytes;
Encoding: WString;
Reason: WString;
Time: WString;
Permission: Integer;
end;
Digital signature information
type TPdfIntegerArray = array of Integer;
Integer array used by structure-tree marked-content ID lists
type TPdfSingleArray = array of Single;
Single-precision floating-point array used by text object character position setters
type TPdfPrintPageRange = record
StartPage: Integer;
EndPage: Integer;
end;
type TPdfPrintPageRanges = array of TPdfPrintPageRange;
One-based print page ranges from the document viewer preferences
type TPdfStructureElement = record
Level: Integer;
ParentIndex: Integer;
TypeName: WString;
ObjectType: WString;
Title: WString;
AlternateText: WString;
ActualText: WString;
Expansion: WString;
ID: WString;
Language: WString;
MarkedContentID: Integer;
MarkedContentIDs: TPdfIntegerArray;
ChildCount: Integer;
ChildMarkedContentIDs: TPdfIntegerArray;
AttributeCount: Integer;
ListNumbering: WString;
Scope: WString;
HasHeaders: Boolean;
HeadersIds: array of WString;
ColSpan: Integer;
RowSpan: Integer;
HasPrintField: Boolean;
NamespaceUri: WString;
RoleMappedNamespaceUri: WString;
RoleMappedTypeName: WString;
HasReference: Boolean;
HasAssociatedFile: Boolean;
AssociatedFileRelationship: WString;
AssociatedFileDescription: WString;
AnnotationSubtype: WString;
AnnotationContents: WString;
AnnotationAlternateText: WString;
IsArtifact: Boolean;
WidgetCount: Integer;
HasAppearance: Boolean;
LabelText: WString;
FieldAlternateName: WString;
end;
type TPdfStructureElements = array of TPdfStructureElement;
Flattened logical structure element data for tagged PDF pages
type TPdfFormSelectionRect = record
PageNumber: Integer;
Bounds: TPdfRectangle;
end;
One caller-owned page-space rectangle reported by native form-text selection
type TPdfStructureReadOptions = record
MaxDepth: Integer;
MaxElements: Integer;
MaxAttributeValues: Integer;
DeadlineMilliseconds: Cardinal;
IncludeAttributes: Boolean;
class function Default: TPdfStructureReadOptions; static;
end;
Resource budgets for iterative native structure-tree traversal
type TPdfStructureReadStatus = record
Completed: Boolean;
ElementsVisited: Integer;
AttributeValuesVisited: Integer;
DepthLimitReached: Boolean;
ElementLimitReached: Boolean;
AttributeLimitReached: Boolean;
DeadlineReached: Boolean;
VisitorStopped: Boolean;
end;
Completion and truncation details returned by structure traversal methods
type TPdfVersion = (pvUnknown, pv10, pv11, pv12, pv13, pv14, pv15, pv16, pv17);
The version of a PDF file or an explicit SaveAs target. Possible values are: unknown version, PDF version 1.0 to 1.7. Explicit SaveAs targets from PDF 1.3 through PDF 1.7 are normalised and validated against the selected PDF version
type TPrintPaperHandling = (phUndefined, phSimplex, phDuplexFlipShortEdge, phDuplexFlipLongEdge);
Print paper handling options
- phUndefined - undefined paper handling
- phSimplex - simplex printing
- phDuplexFlipShortEdge - duplex printing, flip on short edge
- phDuplexFlipLongEdge - duplex printing, flip on long edge
type TPdfPrintMode = (prEmf, prTextOnly, prPostScript2, prPostScript3, prPostScript2PassThrough, prPostScript3PassThrough, prEmfImageMasks, prPostScript3Type42, prPostScript3Type42PassThrough);
Windows PDFium print output mode selected by SetPdfPrintMode
- prEmf - EMF output
- prTextOnly - text-only output
- prPostScript2 - PostScript level 2 output
- prPostScript3 - PostScript level 3 output
- prPostScript2PassThrough - PostScript level 2 pass-through output
- prPostScript3PassThrough - PostScript level 3 pass-through output
- prEmfImageMasks - EMF output with image masks
- prPostScript3Type42 - PostScript level 3 Type 42 output
- prPostScript3Type42PassThrough - PostScript level 3 Type 42 pass-through output
type TRenderOption = (reAnnotations, reLcd, reNoNativeText, reGrayscale, reDebugInfo, reNoCatchException, reLimitCache, reHalftone, rePrinting, reReverseByteOrder, reNoSmoothText, reNoSmoothImage, reNoSmoothPath, reInvert, reNightDuotone);
type TRenderOptions = set of TRenderOption;
Page rendering flags
- reAnnotations - render annotations
- reLcd - text rendering optimised for LCD display
- reNoNativeText - don't use the native text output available on some platforms
- reGrayscale - greyscale output
- reDebugInfo - set if you want to get some debug info
- reNoCatchException - set if you don't want to catch exception
- reLimitCache - limit image cache size
- reHalftone - always use halftone for image stretching
- rePrinting - render for printing
- reReverseByteOrder - render in a reverse Byte order; this flag only enable when render to a bitmap
- reNoSmoothText - disable anti-aliasing on text
- reNoSmoothImage - disable anti-aliasing on images
- reNoSmoothPath - disable anti-aliasing on paths
- reInvert - invert rendered RGB channels as a TPdfView low-vision post-render transform
- reNightDuotone - map rendered luminance to TPdfView.NightDuotoneInk and TPdfView.NightDuotonePaper
type TRotation = (ro0, ro90, ro180, ro270);
The PDF page rotation. The page rotation is rotated clockwise
- ro0 - no page rotation
- ro90 - page rotation 90 degrees clockwise
- ro180 - page rotation 180 degrees clockwise
- ro270 - page rotation 270 degrees clockwise
type TSaveOption = (saNone, saIncremental, saNoIncremental, saRemoveSecurity);
Determines how the PDF document will be saved
- saNone - default save
- saIncremental - incremental save
- saNoIncremental - no incremental save
- saRemoveSecurity - remove security
type TSearchOption = (seCaseSensitive, seWholeWord, seConsecutive);
type TSearchOptions = set of TSearchOption;
Search flags
- seCaseSensitive - match case
- seWholeWord - match the whole word
- seConsecutive - don't skip past the current match to look for the next match
type TWebLink = record
Url: WString;
Rectangles: TPdfRectangles;
end;
type TWebLinks = array of TWebLink;
URL information and boundaries for the web link
type WString = {$ifdef D2009PLUS} string {$else} WideString {$endif D2009PLUS};
Wide string
Constants
const PointsPerInch: Integer = 72;
PDF uses points with a density of 72 points per inch
const FLAT_NORMALDISPLAY = 0;
const FLAT_PRINT = 1;
Flattening profiles passed to FlattenPage / FlattenAllPages. FLAT_NORMALDISPLAY targets screen display; FLAT_PRINT targets printed output
const FLATTEN_FAIL = 0;
const FLATTEN_SUCCESS = 1;
const FLATTEN_NOTHINGTODO = 2;
Result codes returned by FlattenPage
type TPdfReadingUnit = record
Text: WString;
Kind: TPdfContentFragmentKind;
HeadingLevel: Integer;
Page: Integer;
Rects: TPdfRectangles;
FragmentIndex: Integer;
SentenceIndex: Integer;
IsAlternateText: Boolean;
Language: WString;
end
type TPdfReadingUnits = array of TPdfReadingUnit;
Sentence-level reading units used by text-to-speech and assistive listen-reading workflows
type TPdfWordBox = record
Text: WString;
StartIndex: Integer;
Count: Integer;
Rect: TPdfRectangle;
Page: Integer;
end
type TPdfWordBoxes = array of TPdfWordBox;
Word-level text ranges and page rectangles used by synchronised reading highlights
type TPdfSpokenSpan = record
SpokenStart: Integer;
SpokenCount: Integer;
Page: Integer;
StartIndex: Integer;
Count: Integer;
end
type TPdfSpokenSpans = array of TPdfSpokenSpan;
Mapped runs that translate speech-engine offsets back to PDF page text
type TPdfSpokenText = record
Text: WString;
Spans: TPdfSpokenSpans;
end
type TPdfSpokenTexts = array of TPdfSpokenText;
Speech chunks and source mappings used by read-aloud workflows and TPdfReadingSession
type TPdfReadingSessionState = (pssIdle, pssSpeaking, pssPaused);
Playback state reported by TPdfReadingSession
Global Variables
var LibraryName: string = 'pdfium.dll';
Name of the standard PDFium library used when EnableV8Engine is False. Can be changed before calling LoadLibrary
var EnableV8Engine: Boolean = False;
Set to True before loading to select pdfium.v8.dll from the matching Win32 or Win64 DLLs folder