Re: Sun symbol (again!) [message #64405] |
Mon, 22 December 2008 00:06  |
Craig Markwardt
Messages: 1869 Registered: November 1996
|
Senior Member |
|
|
On Dec 21, 10:57 pm, David Fanning <n...@dfanning.com> wrote:
> Craig Markwardt writes:
>> I don't get it. Is there any reason that '!Mn!X' doesn't work?
>> TEXTOIDL() is capable of making this character with '\odot', in
>> contradiction to what the poster says below (I have a version of
>> TEXTOIDL from 1996).
>
> I think it works if you don't mind Hershey fonts in
> your PostScript output. But maybe you need something
> more elegant for formal presentations.
>
OK, I get it. It's rather funny that the Postscript symbol font
doesn't even contain the pure circle glyph. At least then you could
have cobble together a sun symbol with a circle and a dot. I still
don't think that the Hershey fonts are a horrible choice.
Craig
|
|
|
|
|
Re: Sun symbol (again!) [message #64409 is a reply to message #64408] |
Sun, 21 December 2008 18:16   |
Craig Markwardt
Messages: 1869 Registered: November 1996
|
Senior Member |
|
|
On Dec 21, 6:32 pm, Michael Williams <mjwilli...@gmail.com> wrote:
> Has anyone made any progress with getting a circular, typographically
> correct sun symbol using IDL? This is the last post on the issue, and
> I agree with its authors that the solutions from 2004 and 1997 are not
> satisfactory.
I don't get it. Is there any reason that '!Mn!X' doesn't work?
TEXTOIDL() is capable of making this character with '\odot', in
contradiction to what the poster says below (I have a version of
TEXTOIDL from 1996).
Craig
> On 11 Apr, 04:49, "jsch...@gmail.com" <jsch...@gmail.com> wrote:
>
>> I'd like to make a plot with a nice sun symbol.
>
>> It's been a few years since we had this discussion, and I was hoping
>> perhaps someone has some new insight.
>
>> I've read the newsgroup discussions from '04 and '97 and wasn't
>> particularly thrilled with any of the solutions.
>
>> Here are the things that won't / don't work for me.
>
>> 1) TeXtoIDL doesn't know \odot.
>
>> 2) sunsymbol.pro gives something unsatisfactory
>
>> http://web.mit.edu/jschwab/Public/sunsym.png
>
>> 3) The solution that initially seemed most inline with what I want was
>> modifying my ttfont.map and using the marvosym font. I don't find this
>> satisfactory because it's not a permanent solution since I don't want
>> to modify my ttfont.map each time and because I don't even approve of
>> the way it looks. (It's not symmetric!)
>
>> http://web.mit.edu/jschwab/Public/marvosym.png
>
>> What I'm left with is manually positioning the vector font symbol '!9!
>> Z(6E)!X'. It will work and gives me something that's not bad looking.
>> Unsurprisingly though, I find it rather unsatisfying.
>
>> If anyone has a better / more flexible solution, I'd love to hear it.
>
>
|
|
|
Re: Sun symbol (again!) [message #64460 is a reply to message #64408] |
Tue, 23 December 2008 16:56  |
Michael Williams
Messages: 17 Registered: December 2008
|
Junior Member |
|
|
On 22 Dec, 03:57, David Fanning <n...@dfanning.com> wrote:
> Here is an article that describes one possibility using
> freely available (and free) true-type fonts:
>
> http://www.dfanning.com/misc_tips/sun_symbol.html
For the benefit of future Googlers, the sun symbol in marvosym, while
better than many other solutions, is not a circle but a backwards
"o" (note the thickness and angle of the stroke on the right and left-
hand sides). An astronomical sun symbol is a perfect circle with a
point at its centre: http://en.wikipedia.org/wiki/Sun_symbol. Users of
OS X have access to (and licenses for!) a number of TrueType fonts
which contain this symbol and presumably following your procedure to
get a True Type symbol with these fonts would also work. These fonts
include Apple Symbols and Arial Unicode MS. Presumably the same is
true on Windows and Linux.
-- Mike
|
|
|
Re: Sun symbol (again!) [message #64461 is a reply to message #64408] |
Tue, 23 December 2008 16:41  |
Michael Williams
Messages: 17 Registered: December 2008
|
Junior Member |
|
|
On 22 Dec, 03:57, David Fanning <n...@dfanning.com> wrote:
> Craig Markwardt writes:
>> I don't get it. Is there any reason that '!Mn!X' doesn't work?
>> TEXTOIDL() is capable of making this character with '\odot', in
>> contradiction to what the poster says below (I have a version of
>> TEXTOIDL from 1996).
>
> I think it works if you don't mind Hershey fonts in
> your PostScript output. But maybe you need something
> more elegant for formal presentations.
I don't mind Hershey fonts for this one character! Is there any way to
mix them up with real Postscript fonts in the same output? This would
be ideal! I want my sun symbol to be a perfect circle, which Hershey
fonts do admirably, but I don't want the bowls of my "o" and "a" and
"e" to be perfect circles since, well, it's not the 1960s any more,
and computers are capable of better typography! : )
I would prefer to avoid TrueType since the output is visibly blotchy
within individual letters for unknown reasons (you can see this on
your example: http://www.dfanning.com/misc_tips/sun_symbol_6.png)
> Here is an article that describes one possibility using
> freely available (and free) true-type fonts:
>
> http://www.dfanning.com/misc_tips/sun_symbol.html
Thanks very much for writing that up! Very useful information. I'm
afraid none of those really suit my purposes, since I require
publication quality typography. For the time being I will open the
files in Adobe Illustrator and manually edit the Postscript fonts.
At the risk of wandering off-topic, as someone new to IDL, its
typographic capabilities seem to me to be extremely eccentric, crufty
and archaic compared to other graphing toolkits.
|
|
|
Re: Sun symbol (again!) [message #64462 is a reply to message #64407] |
Tue, 23 December 2008 16:34  |
Michael Williams
Messages: 17 Registered: December 2008
|
Junior Member |
|
|
On 22 Dec, 05:28, anand <jana...@gmail.com> wrote:
> the Sun symbol can be put in plots or postscript files using the
> program sunsymbol.pro which you can find athttp://idlastro.gsfc.nasa.gov/ftp/pro/plot/sunsymbol.pro
>
> you can use it as:
> ytitle='height (R'+sunsymbol()+')'
> to express the y-axis title in terms of solar radius.
Thanks. The result (which I understand is a combination of a letter
"o" and a dot) is not great. When Times is used (which is my
preference) it looks nothing like a sun symbol, which is not
surprising given how it is assembled. Helvetica is better but still
far from symmetrical, as can be seen on David Fanning's page.
|
|
|
Re: Sun symbol (again!) [message #64464 is a reply to message #64409] |
Tue, 23 December 2008 16:32  |
Michael Williams
Messages: 17 Registered: December 2008
|
Junior Member |
|
|
On 22 Dec, 02:16, Craig Markwardt <cbmarkwa...@gmail.com> wrote:
> On Dec 21, 6:32 pm, Michael Williams <mjwilli...@gmail.com> wrote:
>
>> Has anyone made any progress with getting a circular, typographically
>> correct sun symbol using IDL? This is the last post on the issue, and
>> I agree with its authors that the solutions from 2004 and 1997 are not
>> satisfactory.
>
> I don't get it. Is there any reason that '!Mn!X' doesn't work?
> TEXTOIDL() is capable of making this character with '\odot', in
> contradiction to what the poster says below (I have a version of
> TEXTOIDL from 1996).
Hi Craig,
those two solutions only work with Hershey fonts, which I would prefer
to avoid for aesthetic reasons.
-- Mike
|
|
|