[R] some questions about font

[Bobai Li]

Dear Paul,

Thank you for your help!

1. I know how to use "expression(italic(whatever))", but I don't know 
how to make "expression(whatever)" to produce italic font.

2. With regard to the CM font, I guess psfrag may be to best way to get 
around the problem. Indeed, I have been using psfrag since I am also 
using LaTeX. My only complaint is that I have yet to find out an easy 
way to export  the modified  graph  back to  the EPS format (to be used 
in other documents).  I know how to do it the hard way, but here I am 
talking about handling some 100 graphic files (It would be more 
efficient to have a R file to create all graphic files with psfrag tags 
and a latex file to replace the tags and export individual graphs to 
EPS; and it would be ideal if the latex file can also be generated 
automatically from R)

3. If you don't mind, I would like ask another question: how to write a 
function f(x) to define---      x <<- "x"

Many thanks for your help!

Bobai




Paul Murrell wrote:

> Hi
>
>
> Paul Murrell wrote:
>
>> Hi
>>
>>
>> Bobai Li wrote:
>>
>>>
>>> Hi,
>>>
>>> I have been using R to create some mathematical and statistical 
>>> graphs for a book manuscript, but I got some problems:
>>>
>>> 1)  Some web positngs said that default typeface for math 
>>> expressions is italic, but in my system (R 2.01 on WinXP), the 
>>> default is regular font.
>>> How can I change the default to ilatic?
>>
>>
>>
>>
>> expression(italic(whatever))
>>
>>
>>> 2)  When use ComputerModern font,  (i.e., 
>>> family=c("CM_regular_10.afm","CM_boldx_10.afm","cmti10.afm","cmbxti10.afm","CM_symbol_10.afm") 
>>> ), some accented symbols are not available. For example, 
>>> "expression(hat(beta))" will produce warning message like "font 
>>> metrics unknown for character 94."
>>
>>
>>
>>
>> That appears to be a bug (that requires changes to the PostScript 
>> device driver).  A nastyish workaround for hat(beta) is 
>> widehat(beta), but I suspect there are other problems that this will 
>> not solve.
>
>
>
> Some good news and some bad news.
>
> The good news is that this can be worked around fairly simply (i.e., 
> without installing a new version of R) as follows:
> (i) make a copy of $R_HOME/library/grDevices/afm/ISOLatin1.enc (or 
> possibly $R_HOME/afm/ISOLatin1.enc, depending on your R version) and 
> call the copy CMISOLatin1.enc.  Modify CMISOLatin1.enc so that the 
> second line starts with CMISOLatin1Encoding (rather than 
> ISOLatin1Encoding) and (further down the file) change \asciicircum to 
> \circumflex and \asciitilde to \tilde.
> (ii) when opening your PostScript device, as well as specifying the 
> family argument as you do above, specify encoding="CMISOLatin1".
>
> The bad news is that R does not position the accents as well as LaTeX 
> does it (e.g., a hat is not located above a beta in quite the same 
> place).
>
> Do you know about PSfrag ...?
>
> Paul