---

Text Storage

ATSUI doesn’t allocate and manage buffers for your text; you are responsible for memory management. Instead ATSUI caches the text that you want rendered, storing the cached text as a sequence of glyph codes. Because ATSUI uses Unicode (UTF-16), each character uses at least 2 bytes of storage. (Surrogate and other characters may require more than 2 bytes of storage.) The storage order is the order in which text is stored. The text stored in memory is the source text; the text displayed is the display text, as shown in Figure 1-3.

Display order is the left-to-right (or top-to-bottom) order in which glyphs are drawn. ATSUI expects you to store your text in input order, which is the “logical” order, or the order in which the characters, not glyphs, would be read or pronounced in the language of the text. Because text of different languages may be read from left to right, right to left, or top to bottom, the input order is not necessarily the same as the display order of the text when it is drawn. Your application needs to differentiate between the order in which the character codes are stored and the order in which the corresponding glyphs are displayed. Figure 1-3 shows Hebrew glyphs that are stored one way and displayed another way.

As shown in Figure 1-3, the character codes that make up the text are numbered using zero-based offsets. Therefore, the first character code in the figure has an offset of 0.

ATSUI also uses a zero-based numbering scheme to index the glyphs that are actually displayed. The glyph index, which gives the glyph’s position in the display order, always starts at 0. Each glyph has a single index, even though its character code is 2 bytes (one Unichar) long.

---