Tk_ComputeTextLayoutSection: Tk Library Procedures (3)
Return to Main Contents
NAMETk_ComputeTextLayout, Tk_FreeTextLayout, Tk_DrawTextLayout, Tk_UnderlineTextLayout, Tk_PointToChar, Tk_CharBbox, Tk_DistanceToTextLayout, Tk_IntersectTextLayout, Tk_TextLayoutToPostscript - routines to measure and display single-font, multi-line, justified text.
#include <tk.h> Tk_TextLayout Tk_ComputeTextLayout(tkfont, string, numChars, wrapLength, justify, flags, widthPtr, heightPtr) void Tk_FreeTextLayout(layout) void Tk_DrawTextLayout(display, drawable, gc, layout, x, y, firstChar, lastChar) void Tk_UnderlineTextLayout(display, drawable, gc, layout, x, y, underline) int Tk_PointToChar(layout, x, y) int Tk_CharBbox(layout, index, xPtr, yPtr, widthPtr, heightPtr) int Tk_DistanceToTextLayout(layout, x, y) int Tk_IntersectTextLayout(layout, x, y, width, height) void Tk_TextLayoutToPostscript(interp, layout)
These routines are for measuring and displaying single-font, multi-line, justified text. To measure and display simple single-font, single-line strings, refer to the documentation for Tk_MeasureChars. There is no programming interface in the core of Tk that supports multi-font, multi-line text; support for that behavior must be built on top of simpler layers. Note that unlike the lower level text display routines, the functions described here all operate on character-oriented lengths and indices rather than byte-oriented values. See the description of Tcl_UtfAtIndex for more details on converting between character and byte offsets.
The routines described here are built on top of the programming interface described in the Tk_MeasureChars documentation. Tab characters and newline/return characters may be treated specially by these procedures, but all other characters are passed through to the lower level.
Tk_ComputeTextLayout computes the layout information needed to display a single-font, multi-line, justified string of text and returns a Tk_TextLayout token that holds this information. This token is used in subsequent calls to procedures such as Tk_DrawTextLayout, Tk_DistanceToTextLayout, and Tk_FreeTextLayout. The string and tkfont used when computing the layout must remain valid for the lifetime of this token.
Tk_FreeTextLayout is called to release the storage associated with layout when it is no longer needed. A layout should not be used in any other text layout procedures once it has been released.
Tk_DrawTextLayout uses the information in layout to display a single-font, multi-line, justified string of text at the specified location.
Tk_UnderlineTextLayout uses the information in layout to display an underline below an individual character. This procedure does not draw the text, just the underline. To produce natively underlined text, an underlined font should be constructed and used. All characters, including tabs, newline/return characters, and spaces at the ends of lines, can be underlined using this method. However, the underline will never be drawn outside of the computed width of layout; the underline will stop at the edge for any character that would extend partially outside of layout, and the underline will not be visible at all for any character that would be located completely outside of the layout.
Tk_PointToChar uses the information in layout to determine the character closest to the given point. The point is specified with respect to the upper-left hand corner of the layout, which is considered to be located at (0, 0). Any point whose y-value is less that 0 will be considered closest to the first character in the text layout; any point whose y-value is greater than the height of the text layout will be considered closest to the last character in the text layout. Any point whose x-value is less than 0 will be considered closest to the first character on that line; any point whose x-value is greater than the width of the text layout will be considered closest to the last character on that line. The return value is the index of the character that was closest to the point. Given a layout with no characters, the value 0 will always be returned, referring to a hypothetical zero-width placeholder character.
Tk_CharBbox uses the information in layout to return the bounding box for the character specified by index. The width of the bounding box is the advance width of the character, and does not include any left or right bearing. Any character that extends partially outside of layout is considered to be truncated at the edge. Any character that would be located completely outside of layout is considered to be zero-width and pegged against the edge. The height of the bounding box is the line height for this font, extending from the top of the ascent to the bottom of the descent; information about the actual height of individual letters is not available. For measurement purposes, a layout that contains no characters is considered to contain a single zero-width placeholder character at index 0. If index was not a valid character index, the return value is 0 and *xPtr, *yPtr, *widthPtr, and *heightPtr are unmodified. Otherwise, if index did specify a valid, the return value is non-zero, and *xPtr, *yPtr, *widthPtr, and *heightPtr are filled with the bounding box information for the character. If any of xPtr, yPtr, widthPtr, or heightPtr are NULL, the corresponding value is not calculated or stored.
Tk_DistanceToTextLayout computes the shortest distance in pixels from the given point (x, y) to the characters in layout. Newline/return characters and non-displaying space characters that occur at the end of individual lines in the text layout are ignored for hit detection purposes, but tab characters are not. The return value is 0 if the point actually hits the layout. If the point didn't hit the layout then the return value is the distance in pixels from the point to the layout.
Tk_IntersectTextLayout determines whether a layout lies entirely inside, entirely outside, or overlaps a given rectangle. Newline/return characters and non-displaying space characters that occur at the end of individual lines in the layout are ignored for intersection calculations. The return value is -1 if the layout is entirely outside of the rectangle, 0 if it overlaps, and 1 if it is entirely inside of the rectangle.
Tk_TextLayoutToPostscript outputs code consisting of a Postscript array of strings that represent the individual lines in layout. It is the responsibility of the caller to take the Postscript array of strings and add some Postscript function operate on the array to render each of the lines. The code that represents the Postscript array of strings is appended to interp->result.
DISPLAY MODELWhen measuring a text layout, space characters that occur at the end of a line are ignored. The space characters still exist and the insertion point can be positioned amongst them, but their additional width is ignored when justifying lines or returning the total width of a text layout. All end-of-line space characters are considered to be attached to the right edge of the line; this behavior is logical for left-justified text and reasonable for center-justified text, but not very useful when editing right-justified text. Spaces are considered variable width characters; the first space that extends past the edge of the text layout is clipped to the edge, and any subsequent spaces on the line are considered zero width and pegged against the edge. Space characters that occur in the middle of a line of text are not suppressed and occupy their normal space width.
Tab characters are not ignored for measurement calculations. If wrapping is turned on and there are enough tabs on a line, the next tab will wrap to the beginning of the next line. There are some possible strange interactions between tabs and justification; tab positions are calculated and the line length computed in a left-justified world, and then the whole resulting line is shifted so it is centered or right-justified, causing the tab columns not to align any more.
When wrapping is turned on, lines may wrap at word breaks (space or tab characters) or newline/returns. A dash or hyphen character in the middle of a word is not considered a word break. Tk_ComputeTextLayout always attempts to place at least one word on each line. If it cannot because the wrapLength is too small, the word will be broken and as much as fits placed on the line and the rest on subsequent line(s). If wrapLength is so small that not even one character can fit on a given line, the wrapLength is ignored for that line and one character will be placed on the line anyhow. When wrapping is turned off, only newline/return characters may cause a line break.
When a text layout has been created using an underlined tkfont, then any space characters that occur at the end of individual lines, newlines/returns, and tabs will not be displayed underlined when Tk_DrawTextLayout is called, because those characters are never actually drawn - they are merely placeholders maintained in the layout.
This document was created by man2html, using the manual pages.
Time: 17:23:14 GMT, May 21, 2013