Open kohlhase opened 2 years ago
Also note the justification construction in the second frame (third line):
\begin{sproof}[for=eta-sound,id=eta-sound-pf]{by calculation}
\spfstep*{\begin{eqnarray*}
\appo{\Evaluation\phi{\lamcdot{X}{\apply[lambda]\bA{X}}}}\sema
& = & \Evaluation{\sbstExtend\phi{\sbst{\sema}{X}}}{\apply[lambda]\bA{X}}\\
& = & \appo{\Evaluation{\sbstExtend\phi{\sbst{\sema}{X}}}\bA}{\Evaluation{\sbstExtend\phi{\sbst{\sema}{X}}}{X}}\\
& = & \appo{\Evaluation\phi\bA}{\Evaluation{\sbstExtend\phi{\sbst{\sema}{X}}}{X}} \qquad
\text{as $\ninset{X}{\freevars\bA}$.}\\
& = & \appo{\Evaluation\phi\bA}\sema
\end{eqnarray*}}
\end{sproof}
If you actually have a new line for every new derivation step, it seems to me like \eqstep perfectly subsumes that (I think it's not well documented yet). Apart from that, \begin{spfblock} is the most "neutral" variant of putting arbitrary text in a proof (e.g. eqnarray), and should still allo for proofsteps, yield etc, just making them presentationally neutral
It seems currently impossible to have an
eqnarry*
directly inside ansproof
environment. SeeMathHub/MiKoMH/CompLog/source/stlc/mod/abe-sound.en.tex
, where I have fixed this via an additional\spfstep
.But this is not very nice, we should have something better (yes, I know that there is
\spfstep*
that is a bit less intrusive, but that does not really help the underlying problem).Either a replacement for the
eqnarray*
(that I only had to introduce since the old{spfeq}
environment got deleted BTW), that allows to mark up the "steps" in the equational proof somehow semantically, ... or ... don't really know.Actually,
abe-sound.en.tex
is the only place where I had{spfeq}
in the sources.