- From: Ian Hickson <ian@hixie.ch>
- Date: Tue, 6 May 2008 21:27:51 +0000 (UTC)
On Tue, 6 May 2008, Mathieu HENRI wrote: > > I can't see any mention that support for Text is optional. Is it > reasonable with browsers approaching major releases ? I don't understand. > The TextMetrics MUST return have 2 Float readonly properties : the width > and height. The width alone is not enough. The width isn't even the bounding box width, swo in fact even the width isn't what you typically want unless you're just laying out text side by side or doing something like centering it, which is what the current API is for. There is a note in the spec to the effect that the current metrics aren't enough; we'll add something in due course, once we know what people really need. (For typography purposes, what you want is the em box, the nominal layout width, and the font's default line-height. Currently the first two of these are available, but not the third.) > The methods fillText() and strokeText() could have a 5th optional > boolean argument called wrapText ( false by default ) to specify if the > text should wrap or the font-size be decrease in case the text is wider > than maxWidth. There is a note in the spec specifically indicating that multiline text will never be provided in this API. We will provide drawElement()-like APIs in preference at some point, if ever. > It is unclear whether the transformations affect the text. Step 6 says "Paint the hypothetical inline box as the shape given by the text's glyphs, as transformed by the current transformation matrix", do you have a suggestion for hwo to make it clearer? -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Tuesday, 6 May 2008 14:27:51 UTC