Re: regtest bombs out with guilev2

[David Kastrup]

Thomas Morley <address@hidden> writes:

> 2018-02-03 20:47 GMT+01:00 David Kastrup <address@hidden>:
>> Thomas Morley <address@hidden> writes:
>>
>>> Hi,
>>>
>>> I was testing Dans recent patches (issues 5267/8/9) with my
>>> guile-v2-setup (with guile-2.2.3 now)
>>> While I've got a successful make, make doc bombs out at
>>> input/regression/markup-partial.ly.
>>>
>>> Processing the derived minimal below
>>>
>>> \markup bold-red = \markup \bold \with-color #red \etc
>>> \markup \bold-red "Bold red."
>>>
>>> returns:
>>>
>>> ~/lilypondH/Test/forum$ lilypond-git-guile-2.2 atest-70.ly
>>> GNU LilyPond 2.21.0
>>> Processing `atest-70.ly'
>>> Parsing...Backtrace:
>>>            9 (apply-smob/1 #<catch-closure 20fd8c0>)
>>> In ice-9/eval.scm:
>>>    293:34  8 (_ #(#(#<directory (lily) 21b6f00>) #<variable 2efde6…>))
>>>     619:8  7 (_ #(#(#(#(#(#(#(#<directory (lily) …>) …) …) …) …) …) …))
>>> In srfi/srfi-1.scm:
>>>     640:9  6 (for-each #<procedure 31f7760 at ice-9/eval.scm:333:13…> …)
>>> In ice-9/eval.scm:
>>>     619:8  5 (_ #(#(#(#(#(#<directory (lily) 21b6f00> # …) …) …) …) …))
>>> In ice-9/boot-9.scm:
>>>     829:9  4 (catch _ _ #<procedure 3402640 at ice-9/eval.scm:386:1…> …)
>>> In unknown file:
>>>            3 (ly:parse-file "atest-70.ly")
>>> In ice-9/eval.scm:
>>>    196:35  2 (_ #f)
>>>    223:20  1 (proc #<module (#{ g166}#) 2825f00>)
>>> In unknown file:
>>>            0 (%resolve-variable (7 . bold-red) #<module (#{ g166}#) …>)
>>>
>>> ERROR: In procedure %resolve-variable:
>>> Unbound variable: bold-red
>>>
>>> I doubt it has anything to do with Dan's patches.
>>>
>>> Any hint what's wrong?
>>
>> What happens when you write
>>
>> bold-red-markup = \markup \bold \with-color #red \etc
>> \markup \bold-red "Bold red."
>
> Works without any complains with the minimal.
>
> Changing the whole regtest to:
>
> bold-red-markup = \markup \bold \with-color #red \etc
> italic-color-markup = \markup \italic \with-color \etc
> quarter-markup = \markup \note-by-number #2 \etc
>
> \markup \bold-red "Bold red."
> \markuplist \column-lines \bold-red { Bold red in a list. }
> \markup #(markup #:italic-color green "Italic green.")
> \markuplist \column-lines \italic-color #green { Italic green in a list. }
> \markup { 3/8: #(make-quarter-markup 1 UP) }
>
> returns:
>
> GNU LilyPond 2.21.0
> Processing `atest-70.ly'
> Parsing...
> atest-70.ly:355:10: error: GUILE signaled an error for the expression
> beginning here
> \markup #
>          (markup #:italic-color green "Italic green.")
> Unbound variable: make-italic-color-markup
> atest-70.ly:355:9: error: not a markup
> \markup
>         #(markup #:italic-color green "Italic green.")
> atest-70.ly:357:17: error: GUILE signaled an error for the expression
> beginning here
> \markup { 3/8: #
>                 (make-quarter-markup 1 UP) }
> Unbound variable: make-quarter-markup
> atest-70.ly:357:16: error: not a markup
> \markup { 3/8:
>                #(make-quarter-markup 1 UP) }

That's very much expected.  \markup xxx = ... does more than just
defining xxx-markup and the regtest checks for the other stuff.

What happens if you use only the first line of your mini regtest, namely

\markup bold-red = \markup \bold \with-color #red \etc

and stop without actually using this?

-- 
David Kastrup