# Difference between revisions of "ApCoCoA-1:NC.SetOrdering"

Line 2: | Line 2: | ||

<title>NC.SetOrdering</title> | <title>NC.SetOrdering</title> | ||

<short_description> | <short_description> | ||

− | Set | + | Set a word ordering on the monoid of all words in a non-commutative polynomial ring. |

</short_description> | </short_description> | ||

<syntax> | <syntax> | ||

Line 31: | Line 31: | ||

</example> | </example> | ||

</description> | </description> | ||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

− | |||

<types> | <types> | ||

<type>non_commutative</type> | <type>non_commutative</type> | ||

</types> | </types> | ||

− | <key> | + | <key>ncpoly.SetOrdering</key> |

<key>NC.SetOrdering</key> | <key>NC.SetOrdering</key> | ||

<key>SetOrdering</key> | <key>SetOrdering</key> | ||

− | <wiki-category> | + | <wiki-category>Package_ncpoly</wiki-category> |

</command> | </command> |

## Revision as of 14:49, 25 April 2013

## NC.SetOrdering

Set a word ordering on the monoid of all words in a non-commutative polynomial ring.

### Syntax

NC.SetOrdering(Ordering:STRING)

### Description

Note that the default ordering is "LLEX" (length-lexicographic ordering).

@param

*Ordering*: a string which indicates an (admissible) ordering. For the time being, the package supports "LLEX" (length-lexicographic ordering), "ELIM" (elimination ordering) and "LRLEX" (length-reverse-lexicographic ordering).

Let `X=x_{1}x_{2}...x_{n}`. We define the (left-to-right) lexicographic ordering "LEX" on `<X>` as follows. For two words `W1, W2` in `<X>`, we say `W1>_{Lex}W2` if we have `W1=W2*W` for some non-empty word `W` in `<X>`, or if we have `W1=W*x_{i}*W3, W2=W*x_{j}*W4` for some words `W,W3,W4` in `<X>` and some letters `x_{i},x_{j}` in `X` such that `i<j`. Thus, we have `x_{1}>_{LEX}x_{2}>_{LEX}...>_{LEX}x_{n}`. Note that "LEX" is not an admissible ordering on `<X>`. We define admissible orderings "LLEX", "ELIM" and "LRLEX" on `<X>` as follows.

"LLEX": for two words

`W1, W2`in`<X>`, we say`W1>_{LLEX}W2`if`len(W1)>len(W2)`, or`len(W1)=len(W2)`and`W1`is lexicographically larger than`W2`."ELIM": it first compares the associated commutative terms lexicographically and then breaks ties using the non-commutative lexicographic orderin with respect to

`x_{1}>...>x_{n}`. That is, for two words`W1, W2`in`<X>`, we say`W1>_{ELIM}W2`if`W1`is lexicographically larger than`W2`by considering`W1, W2`as two terms in the commutative case, or`W1=W2`by considering`W1, W2`as two terms in the commutative case and`W1>_{Lex}W2`(`W1`is left-to-right lexicographically larger than`W2`by considering`W1, W2`as two words in the non-commutative case). Thus, the elimination ordering "ELIM" first eliminates the letter`x_{1}`, and then`x_{2}`, and then`x_{3}`, and so on and so forth."LRLEX": for two words

`W1, W2`in`<X>`, we say`W1>_{LRLEX}W2`if`len(W1)>len(W2)`, or`len(W1)=len(W2)`and`W1`is larger than`W2`by right-to-left lexicographic ordering.

An admissible ordering on is called *length compatible* if `len(W1)>len(W2)` implies `W1` is larger than `W2` for all `W1, W2` in `<X>`. For instance, "LLEX" and "LRLEX" are length compatible while "ELIM" is not.

#### Example

NC.RingEnv(); Coefficient ring : Q Ordering : LLEX ------------------------------- NC.SetOrdering(<quotes>ELIM</quotes>); NC.RingEnv(); Coefficient ring : Q Ordering : ELIM -------------------------------