Re: Phantom Error Messages in v9
- To: mathgroup at smc.vnet.net
- Subject: [mg129329] Re: Phantom Error Messages in v9
- From: James Stein <mathgroup at stein.org>
- Date: Sat, 5 Jan 2013 02:17:25 -0500 (EST)
- Delivered-to: l-mathgroup@mail-archive0.wolfram.com
- Delivered-to: l-mathgroup@wolfram.com
- Delivered-to: mathgroup-newout@smc.vnet.net
- Delivered-to: mathgroup-newsend@smc.vnet.net
On Fri, Jan 4, 2013 at 12:43 PM, McHale, Paul <Paul.McHale at excelitas.com>
wrote:
> Hi James,****
>
> ** **
>
> I am experiencing a similar problem. Have you found a resolution?****
>
> ** **
>
> Thanks,****
>
> Paul
>
Yes & No. An updated (and clearer) notebook showing the problem is here:
https://www.dropbox.com/s/wmqtsl41521x00s/SpuriousMsgs.nb
Someone at Wolfram responded on Jan 2 2013 as follows:
I have reproduced this problem with spurious error messages. It appears to
be caused by the fonts used in your Input cell (recreating the cell with
the default fonts does not produce the error messages). I also note that
this appears to be related to dynamic highlighting - turning dynamic
highlighting off in the Preference menu (using the appropriate checkbox in
the Interface tab) also resolves the problem.
I have filed a report with our developers on this issue so that the problem
can be fixed in a future version of Mathematica.
to which I responded:
Thank you for investigating this issue. I have a comment and a conclusion:
It is true that in the file I sent you, the font was smaller than normal.
This happened during my efforts to reproduce the bug in a simple case.
However the bug does occur in the "normal" font, provided that **some** of
the text in the usage-string is reduced in size (by selecting it and
pressing command-minus.) Reducing text like this is normal -- I learned to
do it by copying the style Mathematica employs for its own usage statements.
For example, if you evaluate "?Part", you will see that Mathematica's usage
statement for 'Part' contains both italicized and size-reduced portions --
and this evokes the bug. Therefore the bug is NOT caused by a
"non-standard" font; it is caused by some change in v9 which broke
Mathematica-style usage statements (when written by users). Therefore the
bug might be deemed to have higher priority than otherwise.
I have thousands of such usage statements in hundreds of notebooks in
dozens of projects -- and I learned the style by copying Wolfram!! With
version 9, I have learned to ignore the errors in the Messages window, as
they are 99.99% erroneous -- they appear by the dozens as I edit any of my
many notebook files.
Thanks again
-- James Stein