Monkeybread Software - DynaPDF Manual

DynaPDF Manual - Page 481

Function Reference

Page 481 of 839

value of leading must not be considered to calculate the text position.

TextRise

The text rise is already included in the text transformation matrix tm. The

value is always 0.0.

WordSP

Current word spacing, info only.

CharSP

Current character spacing, info only.

HScale

Current horizontal scaling, info only.

LineWidth

The current line width; it can be ignored if the text is not stroked.

Text

The raw text line that was found in the content stream. This is a non-null-

terminated binary string; it can contain single or multi-byte characters.

This member should no longer be used to extract or replace text. Use the

member Kerning instead.

TextLen

The string length in bytes, not characters! If TextLen is zero no text was

found or an error occurred.

RawKern

The raw kerning array. The text in this array points directly into the

source text buffer. The strings are not null-terminated.

Kerning

Already translated UTF-16 Unicode kerning array. A kerning record

consists of the members Advance, Text (UTF-16), Length, and Width. The

member Advance represents the amount in which the text is moved on the

x-axis. Negative values move the cursor to the right, positive to the left.

Normal kerning between characters is usually always less than the half

space width which can be taken from the structure TPDFStack. This makes

it easy to determine whether a space character is emulated with kerning

space. Note that negative values move the cursor to right! A positive value

of Advance cannot emulate a space character because the cursor is moved

to the wrong direction. The parameter Text contains the Unicode text that

is not null-terminated. Length is the text length in characters. The text

width and advance are calculated in text space.

Note that it is not always possible to translate a text to Unicode. The

translation will fail if a CID font depends on an external CMap file that is

not available. External CMaps should always be loaded with

SetCMapDir() before GetPageText() is executed the first time.

It is also possible that a PDF file contains an invalid encoding or

ToUnicode CMap. Because the encoding or ToUnicode CMap is only used