Function: algmodprinit
Section: algebras
C-Name: algmodprinit
Prototype: GGDn
Help: algmodprinit(al,pr,{v}): initialises a reduction mod pr map.
Doc: Consider a central simple algebra \var{al} output by \tet{alginit} and a
 \kbd{prid} \var{pr} representing a prime ideal $\goth{p}$ at which the order
 ${\cal O}$ stored in \var{al} is maximal (not checked; by default, this is true
 for all primes, see \kbd{alginit}). This function returns a data structure
 \var{modP} representing a reduction mod $\goth{p}$ map
 $$
  {\cal O} / \goth{p}{\cal O} \longrightarrow {\cal O} / \goth{P}
  \simeq M_k(\F_{p^m})
 $$
 where~$\goth{P}$ is the unique maximal two-sided ${\cal O}$-ideal above~$\goth{p}$,
 $m = e \times pr.f$ with~$e$ the ramification index of~\var{al}
 at~$\goth{p}$ ($\goth{P}^e = \goth{p}{\cal O}$) and~$d = ke$ is the degree of~\var{al}.
 In particular, if $\goth{p}$ is unramified in~\var{al}, then~$k=d$ and~$\F_{p^m}$
 is isomorphic to the residue field of~$\goth{p}$.
 The variable \var{v} is used to display the finite field elements.
 The function is nondeterministic and does not always return the same map, but
 the target ring is deterministic (the finite field is always represented by the
 \tet{ffinit} polynomial).
 The structure \var{modP} contains:

 \item \var{modP}[1]: the \kbd{prid} \var{pr};

 \item \var{modP}[2]: the \typ{VECSMALL}~$[k,m]$;

 \item \var{modP}[3]: the \typ{FFELT}~$1\in \F_{q^m}$;

 \item other technical data.

 \bprog
 ? nf = nfinit(y^2-5);
 ? al = alginit(nf, [-1,-1]); \\ quaternion algebra unramified at all primes
 ? pr = idealprimedec(nf, 3)[1]; pr.f
 % = 2
 ? modP = algmodprinit(al, pr, 'a);
 ? modP[2] \\ image in M_2(F_{3^2})
 % = Vecsmall([2,2])
 ? type(modP[3])
 % = "t_FFELT"
 ? algmodpr(al,algrandom(al,3),modP)
 % =
 [2*a + 1 2*a + 1]

 [      a     2*a]
 ? nf = nfinit(y); \\ Q
 ? al = alginit(nf, [-1,-1]); \\ quaternion algebra ramified at 2 and oo
 ? pr = idealprimedec(nf, 2)[1];
 ? modP = algmodprinit(al, pr, 'a);
 ? modP[2] \\ image in M_1(F_{2^2})
 % = Vecsmall([1,2])
 ? algmodpr(al,algrandom(al,4),modP)
 % =
 [a + 1]
 @eprog
