#include <KernelApi.h>
Public Attributes | |
| WORD | left |
| WORD | top |
| WORD | width |
| WORD | height |
| float | pointSize |
| WORD | capHeight |
| WORD | baseLine |
| WORD | zone |
| WCHAR | code |
| BYTE | err |
| BYTE | reserved_b |
| BYTE | cntChoices |
| BYTE | cntSuggestions |
| DWORD | ndxChoices |
| WORD | fontAttrib |
| WORD | ndxFontFace |
| DWORD | info |
| WORD | makeup |
| BYTE | widthULdot |
| BYTE | widthULgap |
| WORD | cellNum |
| BYTE | ndxFGColor |
| BYTE | ndxBGColor |
| short | lang |
| short | lang2 |
| DWORD | ndxExt |
The LETTER structure.
This is a recognized data structure. As the result of the recognition process the recognition data will consist of this type of structure for each recognized character. This is the most detailed information available about the recognized characters.
See the usage of alternatives and handling of spaces.
pointSize is not a replacement for the field fontSize in the CSDK versions 12.x (that's the cause of the name change):pointSize is filled only on textual PDF inputs. For others, it is used only internally during page formatting (only at RecApiPlus level).capHeight is always available and could be used for replacing fontSize. Rough approximation: fontSize = capHeight * 100 / dpi (fontSize in CSDK 12.7 is calculated in this way).left, top, width, height, capHeight, baseLine, widthULdot, widthULgap) is expressed in pixel coordinates mapped to the image specified when getting the letters. left, top, width, height) of a character usually contains that single character only, but sometimes more characters are recognized together in one step, in which case all those characters have the same bounding box. | WORD LETTER::baseLine |
Y coordinate of the baseline in pixels. In case of vertical text this is an X coordinate. In CCJK vertical text the baseline is at the middle of the characters by definition. In case of barcodes this coordinate is independent of the direction of the barcode, it is always the Y coordinate of the bottom of the barcode.
| WORD LETTER::capHeight |
Expresses a measure of the capital letter height in pixels. See notes for more info!
| WORD LETTER::cellNum |
Index of the cell in the cell list which contains the character (applicable only for WT_TABLE zones). Index of the text-line form-element object which contains the character (in case of WT_FORM zones).
| BYTE LETTER::cntChoices |
Number of related choices continuously placed in the external choice string.
| BYTE LETTER::cntSuggestions |
Number of related suggestions continuously placed in the external suggestion array.
| WCHAR LETTER::code |
Character code in 16-bit UNICODE. This is the first choice of the recognition or UNICODE_REJECTED for rejected characters.
| BYTE LETTER::err |
Confidence number expressing both the first guess' recognition certainty (code member) and also the word certainty. For more information see the section confidence reporting.
| WORD LETTER::fontAttrib |
Font information about the recognized character. Used by the OCR engines. See its possible bits.
| WORD LETTER::height |
Height of the character rectangle in pixels.
| DWORD LETTER::info |
Additional information about the character. See its possible bits and the macros for easier handling this information.
| short LETTER::lang |
This is used to declare which language the recognized word belongs to. See Language of a word.
| short LETTER::lang2 |
See field lang.
| WORD LETTER::left |
Left boundary of the rectangle containing the character in pixels.
| WORD LETTER::makeup |
Since the recognition data does not contain extra characters for marking the line ends, paragraphs, pages, etc., these items of information are stored for the particular characters in this field. It can be any binary OR-ed combination of the possible formatting attributes.
| BYTE LETTER::ndxBGColor |
Index of the background color within the palette of the recognition data. See kRecGetLetterPalette.
| DWORD LETTER::ndxChoices |
Index of the second choice in the external choice string.
| DWORD LETTER::ndxExt |
Not documented.
| BYTE LETTER::ndxFGColor |
Index of the foreground color within the palette of the recognition data. See kRecGetLetterPalette.
| WORD LETTER::ndxFontFace |
Index of the logical font definition placed in an external font array.
| DWORD LETTER::ndxSuggestions |
If this LETTER is not a space, this member is the index of the first suggestion in the external suggestion string. (This member makes a union together with spcInfo.)
| float LETTER::pointSize |
Font size in points. See notes for more info!
| BYTE LETTER::reserved_b |
Internal use only.
| LSPC LETTER::spcInfo |
If this LETTER is a space additional information is available here. (This member makes a union together with ndxSuggestions.)
| WORD LETTER::top |
Top boundary of the rectangle containing the character in pixels.
| WORD LETTER::width |
Width of the character rectangle in pixels. See also the section Handling of spaces about 0-width spaces.
| BYTE LETTER::widthULdot |
Width of a dot in pixels if the "underline" is underdots in reality. 0 if simple underline, 0 if nothing. It also gives this information in the case of dotleaders. (See LSPC.)
| BYTE LETTER::widthULgap |
Width of a gap in pixels if the "underline" is underdots in reality. 0 if simple underline, 0 if nothing. It also gives this information in the case of dotleaders. (See LSPC.)
| WORD LETTER::zone |
Index of the zone in the zone list which contains the character.