[Dave Airey]

The installation issue has been fixed on SSC by Kit Baum and Daniel Bischof. Many thanks.

[Daniel Bischof]

Dear all,

Thanks for all the feedback. I'm not really following this forum. Thus, usually German Rodriguez drops me an email outlining what I missed lately (#THX!)...Sorry for being quite ignorant about forum posts.

Indeed I used the more complex solution as I suspected it would be the better approach to solve the issues.

I was not aware that the schemes are no longer available on SSC. There was an issue with uploading which Kit has solved by now. They should be up and running with the complex fix.

I shall return to my code by the end of the week and replace the complex one with the simpler and more suitable one.

I will tweet when the newest version is online.

PS: I'm more active on twitter...

[George Hoffman]

i continue to have problems with the MRC and TFL schemes in Stata15.
i get the following error (repeated multiple times) when drawing a boxplot and scheme MRC:
(note: named style default not found in class alignstroke, default attributes used)

i have modified the scheme-mrc.scheme file by adding the line:
#include s2color

but this solution proposed by Jeff Pitblado from StataCorp does not address the error in the MRC/TFL schemes

is there a working generalizable solution to this schemes problem?

[wbuchanan]

George Hoffman have you tried to use brewscheme?

[George Hoffman]

wbuchanan - i have seen your presentation on brewscheme and read your suggestions here before. thank you for your suggestion. i have tried brewscheme but i am probably not up to extensive cooking of schemes.

my question was probably directly more to Stata developers, since a response earlier on this thread suggested that adding a revlevant #include statement would make older scheme files 'future proof'.

more specific to the error that in the MRC/TFL schemes: it seems that the term 'default' has specific meaning in the stata14 scheme file, but not in stata15; therefore an error is generated. but i am reaching here.

[wbuchanan]

I’m not sure what the MRC/TFL schemes are, but if you were trying to recreate the ggplot2 aesthetics there are examples in the documentation to handle that case if nothing else.