thys/notes.tex
changeset 58 1769b702d4dc
parent 57 7093e600ec2c
child 59 434ba43d0430
equal deleted inserted replaced
57:7093e600ec2c 58:1769b702d4dc
   162 (inj\; r_1\;c\;u_1)\cdot u_2 \succ_{r_1 \cdot r_2} 
   162 (inj\; r_1\;c\;u_1)\cdot u_2 \succ_{r_1 \cdot r_2} 
   163 (mkeps\;r_1) \cdot (inj\; r_2\;c\;w_2)
   163 (mkeps\;r_1) \cdot (inj\; r_2\;c\;w_2)
   164 \]
   164 \]
   165 
   165 
   166 \noindent
   166 \noindent
   167 This is what we need to prove.
   167 This is what we need to prove. There are only two rules that
       
   168 can be used to prove this judgement:
       
   169 
       
   170 \begin{center}
       
   171 \begin{tabular}{cc}
       
   172 \begin{tabular}{c}
       
   173 $v_1 = v_1'$\qquad $v_2 \succ_{r_2} v_2'$\\
       
   174 \hline
       
   175 $v_1\cdot v_2 \succ_{r_1\cdot r_2} v_1'\cdot v_2'$
       
   176 \end{tabular} &
       
   177 \begin{tabular}{c}
       
   178 $v_1 \succ_{r_1} v_1'$\\
       
   179 \hline
       
   180 $v_1\cdot v_2 \succ_{r_1\cdot r_2} v_1'\cdot v_2'$
       
   181 \end{tabular}
       
   182 \end{tabular}
       
   183 \end{center}
       
   184 
       
   185 \noindent 
       
   186 Using the left rule would mean we need to show that
       
   187 
       
   188 \[
       
   189 inj\; r_1\;c\;u_1 = mkeps\;r_1
       
   190 \]
       
   191 
       
   192 \noindent 
       
   193 but this can never be the case.\footnote{Actually Isabelle
       
   194 found this out after analysing its argument. ;o)} Lets assume
       
   195 it would be true, then also if we flat each side, it must hold
       
   196 that
       
   197 
       
   198 \[
       
   199 |inj\; r_1\;c\;u_1| = |mkeps\;r_1|
       
   200 \]
       
   201 
       
   202 \noindent 
       
   203 But this leads to a contradiction, because the right-hand side
       
   204 will be equal to the empty list, or empty string. This is 
       
   205 because we assumed $nullable(r_1)$ and there is a lemma
       
   206 called \texttt{mkeps\_flat} which shows this. On the other
       
   207 side we know by assumption (b***) and lemma \texttt{v4} that 
       
   208 the other side needs to be a string starting with $c$ (since
       
   209 we inject $c$ into $u_1$). The empty string can never be equal 
       
   210 to something starting with $c$\ldots therefore there is a 
       
   211 contradiction.
       
   212 
       
   213 That means we can only use the rule on the right-hand side to 
       
   214 prove our goal. This implies we need to prove
       
   215 
       
   216 \[
       
   217 inj\; r_1\;c\;u_1 \succ_{r_1} mkeps\;r_1
       
   218 \]
   168 
   219 
   169 
   220 
   170 \subsubsection*{Case RL}
   221 \subsubsection*{Case RL}
   171 
   222 
   172 The corresponding rule (instantiated) is:
   223 The corresponding rule (instantiated) is: