Re: Meaning of NSFont italicAngle?

FROM : Henry McGilton
DATE : Thu Nov 04 20:45:40 2004

On Thursday, November 4, 2004, at 10:47 AM, Thomas Lachand-Robert wrote:

>
> Le 4 nov. 04, à 18:28, Henry McGilton a écrit :
>

>>
>> On Wednesday, November 3, 2004, at 02:15 PM, Thomas Lachand-Robert 
>> wrote:
>>

>>> Hi all,
>>>
>>> I would like to know the true meaning of italicAngle for a NSFont. 
>>> The doc says:
>>>
>>> - (float)italicAngle
>>> Returns the receiver’s italic angle, the amount that the font is 
>>> slanted in degrees clockwise from the vertical, as read from its AFM 
>>> file.

>>
>> What 'doc' are you reading?  My documentation of NSFont says:
>>
>>        Returns the receiver's italic angle, the amount that the font
>>        is slanted in degrees counterclockwise from the vertical,
>>        as read from its AFM file.
>>
>> Note  'counterclockwise'.  Given that context, a clockwise slant 
>> should
>> be interpreted as a negative number.  In general, Adobe PostScript 
>> (and
>> by extension, font metrics) use positive values to mean 
>> counterclockwise
>> rotations.
>>

>
> I return you the question. (Maybe you got the doc for Tiger?) I just 
> copied and paste from Apple's developer doc. I get the same on the web 
> site:
> http://developer.apple.com/documentation/Cocoa/Reference/
> ApplicationKit/ObjC_classic/Classes/NSFont.html#//apple_ref/doc/uid/
> 20000382/italicAngle



I was referencing my 10.2 developer reference documentation.  The NSFont
page says its last update was July 1, 2002.  That's where I got the
paragraph with the 'counterclockwise' in it.

I also checked with the Adobe Font Metrics Specification, Version 4.1,
dated 7 October 1998.  It states that the Italic Angle is a 
'counterclockwise'
angle.

The reference on Apple's web site that you provided above was updated on
August 31, 2004.

So it looks as if there are two or possibly even three problems.

1.  Apple have decided to change the convention for the direction
    of angles.

    This scenario is unlikely, as NSBezierPath and NSAffineTransform
    appear to be staying with the counterclockwise convention.

2.  Somebody injected a  'bug'  into the NSFont documentation in the
    past couple of years.

    This scenario sounds more likely to me.

3.  The angles measured on the screen and the angles as reported by
    NSFont's  italicAngle  method appear to be off by a factor of
    ten.

I poked through the archives at mamsam and cocoabuilder, and the
only references to  italicAngle  are our discussion.

Would somebody at Apple please confirm the truth of both the
documentation and NSFont behaviour, before we file bugs?

I am surprised that this has not cropped up before.  Maybe
none of applications using the text system know about using
italic correction --- I know for sure Micro$oft Internet
Exploder doesn't.

>
> Anyway I don't mind much the sign, since I know how the fonts are 
> slanted in my case. I am looking for the amount.
>

>>> And 1.5 degree is a very small angle, so it's probably not that
>>> (but in radians, it's too much!). Could it be the tangent of the 
>>> angle?

>>
>> I checked the AFM file for Times-Italic.  The AFM file says that the
>> Italic angle is  -15.5.  I used TextEdit to display some Times-Italic
>> text at large size, and, using my trusty Protractors application, 
>> measured
>> the angle as being approximately 15 degrees clockwise from the 
>> vertical.
>>
>> I also checked the AFM file for Helvetica-Oblique.  The AFM file says 
>> that the
>> Italic angle is  -12.0.  I used TextEdit to display some 
>> Helvetica-Oblique
>> text at large size, and, using my trusty Protractors application, 
>> measured
>> the angle as being approximately 12 degrees clockwise from the 
>> vertical.
>>
>> I also checked for Optima-Oblique (which for some obscure reason 
>> Cocoa wants
>> to be called Optima-Italic).  The AFM file says Italic angle is 
>> -11.0 (this
>> is an old AFM file).  The system reports the angle as  -1.2.
>>
>> Very unlikely that the number is the tangent.  The tangent of -15.5 
>> degrees
>> is about -0.278.
>>
>> There does seem to be some kind of anomaly, though.  It's as if the 
>> system is
>> off by a factor of ten in every case.
>>
>> Could somebody at Apple comment on this anomaly?
>>

> Yes I agree with you: it's probably a factor 10 or something. Anyway 
> you proved that it is NOT the value in the AFM file, so the doc is 
> definitely wrong.
>
> Thanks for your comments and experiments,



    Regards,
       ........  Henry


===============================+============================
  Henry McGilton, Boulevardier |    Trilithon Software
    Objective-C/Java Composer  |    Seroia Research
-------------------------------+----------------------------
  mailto:<email_removed>  |  http://www.trilithon.com
                                |
===============================+============================


===============================+============================
  Henry McGilton, Boulevardier |    Trilithon Software
    Objective-C/Java Composer  |    Seroia Research
-------------------------------+----------------------------
  mailto:<email_removed>  |  http://www.trilithon.com
                                |
===============================+============================

===============================+============================
  Henry McGilton, Boulevardier |    Trilithon Software
    Objective-C/Java Composer  |    Seroia Research
-------------------------------+----------------------------
  mailto:<email_removed>  |  http://www.trilithon.com
                                |
===============================+============================

===============================+============================
  Henry McGilton, Boulevardier |    Trilithon Software
    Objective-C/Java Composer  |    Seroia Research
-------------------------------+----------------------------
  mailto:<email_removed>  |  http://www.trilithon.com
                                |
===============================+============================