Journal/Paper.thy
changeset 179 edacc141060f
parent 178 c6ebfe052109
child 180 b755090d0f3d
equal deleted inserted replaced
178:c6ebfe052109 179:edacc141060f
   777   This allows us to prove a version of Arden's Lemma on the level of equations.
   777   This allows us to prove a version of Arden's Lemma on the level of equations.
   778 
   778 
   779   \begin{lmm}\label{ardenable}
   779   \begin{lmm}\label{ardenable}
   780   Given an equation @{text "X = rhs"}.
   780   Given an equation @{text "X = rhs"}.
   781   If @{text "X = \<Union>\<calL> ` rhs"},
   781   If @{text "X = \<Union>\<calL> ` rhs"},
   782   @{thm (prem 2) Arden_keeps_eq}, and
   782   @{thm (prem 2) Arden_preserves_soundness}, and
   783   @{thm (prem 3) Arden_keeps_eq}, then
   783   @{thm (prem 3) Arden_preserves_soundness}, then
   784   @{text "X = \<Union>\<calL> ` (Arden X rhs)"}.
   784   @{text "X = \<Union>\<calL> ` (Arden X rhs)"}.
   785   \end{lmm}
   785   \end{lmm}
   786   
   786   
   787   \noindent
   787   \noindent
   788   Our @{text ardenable} condition is slightly stronger than needed for applying Arden's Lemma,
   788   Our @{text ardenable} condition is slightly stronger than needed for applying Arden's Lemma,
  1096 
  1096 
  1097   \noindent
  1097   \noindent
  1098   Much more interesting, however, are the inductive cases. They seem hard to solve 
  1098   Much more interesting, however, are the inductive cases. They seem hard to solve 
  1099   directly. The reader is invited to try. 
  1099   directly. The reader is invited to try. 
  1100 
  1100 
       
  1101   Constable et al \cite{Constable00} give the following suggestive picture about
       
  1102   equivalence classes:
       
  1103 
  1101   Our first proof will rely on some
  1104   Our first proof will rely on some
  1102   \emph{tagging-functions} defined over strings. Given the inductive hypothesis, it will 
  1105   \emph{tagging-functions} defined over strings. Given the inductive hypothesis, it will 
  1103   be easy to prove that the \emph{range} of these tagging-functions is finite. 
  1106   be easy to prove that the \emph{range} of these tagging-functions is finite. 
  1104   The range of a function @{text f} is defined as 
  1107   The range of a function @{text f} is defined as 
  1105   
  1108