%%File fsa.tex \Chapter{Programs for Manipulating Finite State Automata}
In this chapter, we describe the utility functions that are provided for
manipulating finite state automata.
All of these except 'fsafilter', 'nfadeterminize' and 'midfadeterminize'
accept only deterministic automata as input.
For explanation of all of the standard options,
see Section "Exit Status of Programs and Meanings of Some Options".
{\em Acknowledgement}\:\
The code for 'fsagrowth' was written by Laurent Bartholdi
This merely reads in a finite state automaton and prints it out again.
It is used partly for testing, but also if one wants to change the format
of an automaton from dense to sparse or vice-versa.
If the optional <filename> argument is present, then input is from <filename>
and output to <filename>'.filter'. Otherwise, input is from 'stdin' and
output to 'stdout'.
If the '-csn' option is called, then the numbers of the states in the
transition table of the output are included as comments after the lists
of transitions for each state. This can help to increase the readability
of a file defining an automaton with a large number of states.
A finite state automaton is read in, minimized and then printed out again.
If the optional <filename> argument is present, then input is from <filename>
and output to <filename>'.min'. Otherwise, input is from 'stdin' and
output to 'stdout'.
This is a minimization program for finite state automata which may have
more than the standard two state categories (accepting and non-accepting),
and the output automataon has a minimal number of states subject to
the condition that any word read by the initial and output automata
leads to states having the same category.
The input automaton must have state set of type <labeled>, and the labels
of the staes are used to specify the categories. One of these categories
corresponds to label 0, which means no label. Preferably, the accepting states
of the automaton should consist of a union of those states having certain
labels. If this is not the case, then the accepted languages of the input
and output automata will not necessarily be the same.
If the optional <filename> argument is present, then input is from <filename>
and output to <filename>'.labmin'. Otherwise, input is from 'stdin' and
output to 'stdout'.
A finite state automaton is read in, and then its states are permuted into
bfs-order (bfs = breadth-first-search), and it is printed out again.
This means that the states are numbered $1,2, \ldots, ns$, and if one
scans the transition-table, in order of increasing states, then the states
occur in increasing order. 'fsamin' and 'fsabfs' used together can be used
to check whether two deterministic automata with the same alphabet
have the same language.
First apply 'fsamin' and then 'fsabfs'.
If they have the same language, then the resulting automata should be identical.
If the optional <filename> argument is present, then input is from <filename>
and output to <filename>'.bfs'. Otherwise, input is from 'stdin' and
output to 'stdout'.
\Section{fsacount}
'fsacount [-ip d/s] [-silent] [-v] []'
A finite state automaton is read in,
the size of the accepted language is counted, and the answer (which may
of course be infinite) is output to 'stdout'. Input is from <filename> if
the optional argument is present, and otherwise from 'stdin'.
\Section{fsaenumerate}
'fsaenumerate [-ip d/s] [-bfs/-dfs] []'
A finite state automaton is read in from 'stdin' or from the file <filename>
if the optional argument is present.
<min> and <max> should be non-negative integers with <min> $\le$ <max>.
The words in the accepted language having lengths at least <min> and at
most <max> are enumerated, and output as a list of words to 'stdout' or
to the file <filename>'.enumerate'.
If the option '-dfs' is called (depth-first search -- the default),
then the words
in the list will be in lexicographical order, whereas with '[-bfs]'
(breadth-first-search), they will be in order of increasing length, and in
lexicographical order for each individual length (i.e. in shortlex order).
Depth-first-search is marginally quicker.
\Section{fsalequal}
'fsalequal [-ip d/s] [-silent] [-v] '
Two finite state automata are read in from the files <filename1> and
<filename2>. This program tests whether they have the same language.
The exit code is 0 if they do, and 2 if not.
(To have the same language they must have identical alphabets.)
Two finite state automata, which must have the same alphabet,
are read in from the files <filename1> and <filename2>. An automaton that
accepts a word $w$ in the alphabet if and only if both of the input automata
accept $w$ is computed, minimized, and output to the file <outfilename>.
Two finite state automata, which must have the same alphabet, are read in from
the files <filename1> and <filename2>. An automaton that accepts a word $w$ in
the alphabet if and only if at least one of the input automata
accept $w$ is computed, minimized, and output to the file <outfilename>.
Two finite state automata, which must have the same alphabet, are read in from
the files <filename1> and <filename2>. An automaton that accepts a word $w$ in
the alphabet if and only if the first input automaton
accepts $w$ and the second input automaton does not accept $w$
is computed, minimized, and output to the file <outfilename>.
A finite state automaton is read in from the file <filename>, and an automaton
with the same alphabet, which accepts a word $w$ if and only if the input
automaton does not accept $w$, is calculated and output to <filename>'.not'.
Two finite state automata, which must have the same alphabet, are read in from
the files <filename1> and <filename2>. An automaton that accepts a word $w$ in
the alphabet if and only if $w = w_1w_2$, where $w_1$ and $w_2$ are in
the respective languages of the input automata,
is computed, minimized, and output to the file <outfilename>.
\Section{fsastar}
'fsastar [-ip d/s[]] [-op d/s] [-silent] [-v] '
A finite state automaton is read in from the file <filename>, and an automaton
with the same alphabet, which accepts a word $w$ if and only if
$w \in L^\*$, where $L$ is the language of the input automaton,
is calculated, minimized, and output to <filename>'.star'.
A finite state automaton is read in from the file <filename>, and an automaton
with the same alphabet, which accepts a word $w$ if and only if the input
automaton accepts the reversed word $w^R$, is calculated and output to
<filename>'.reverse'.
If '-s' is used, then the states of the output automaton (which will
not be minimized) will be labeled as lists of integers, which specify
the corresponding subsets of the states of the input automaton.
If '-midfa' is used, then the output automaton will be a midfa,
and the initial states will correspond to the accepting states of the input automaton.
A two-variable finite state automaton is read in from the file <filename>,
and a one-variable automaton, which accepts a word $v$ if and only if the input
automaton accepts $(v,w)$ for some word $w$,
is calculated and output to <filename>'.exists'.
\Section{fsaswapcoords}
'fsaswapcoords [-op d/s] [-silent] [-v] [-l] []'
A two-variable finite state automaton is read in from <stdin> or from
the file <filename1> if specified. A new automaton that accepts a pair of
words $(v,w)$ if and only if the original automataon accepts $(w,v)$ is
computed, and output to 'stdout' or to the file <filename2> if specified.
(This could, for example, be used before 'fsaexists', if it was desired to
quantify over the second coordinate of a two-variable automataon, rather
than the first.)
A finite state automaton is read in, and a modified automaton computed
and then printed out. The output automaton accepts a word $w$ if and
only if the input automaton accepts $w$ and infinitely many words having
$w$ as a prefix.
If |-a| is used, then the computation starts by making all states of the
automaton accepting (and the program runs quicker with this option set).
Equivalently, with |-a|, the output automaton accepts a word $w$ if and
only if the input automaton accepts infinitely many words having
$w$ as a prefix.
This program is useful, for example, for finding geodesics to the boundary in
word-hyperbolic groups.
If the optional <filename> argument is present, then input is from <filename>
and output to <filename>'.prune'. Otherwise, input is from 'stdin' and
output to 'stdout'.
'fsagrowth' computes the rational growth function of a finite state automaton.
The input is from <filename> (or from 'stdin' if the optional argument is not
present), which should contain a finite state automaton.
The output is to <filename>'.growth' or to 'stdout', and contains the quotient
of two integral polynomials in the variable $X$. The coefficient of $X^n$ in
the Taylor expansion of this polynomial is equal to the number of accepted
words of the automaton length $n$.
In fact, to avoid the danger of integer overflow, the calculations are not done
over the integers, but modulo a sequence of primes. If, on lifting to the
integers, different primes give different results, then the coefficients of
the polynomials are too large, and the results are likely to be wrong.
A warning message is printed when this happens, and the exit code is 2.
By default, three primes slightly less than $2^{15}$ are used, but the
user can input the primes to be used by using the '-primes x,y,...' option.
The list of primes here should be comma-separate, with no space.
The code for this program was written by Laurent Bartholdi.
'nfadeterminize' read a (possibly) non-deterministic automaton from 'stdin', or from <filename> if the optional argument is present.
It computes and minimizes an equivalent deterministic
automaton with the same language and outputs the result to 'stdout'
or to <filename>'.determinize'. The input automaton, which will
normally be stored in sparse format, may contain $\varepsilon$-transitions,
which should be specified by giving them the generator number 0.
If the '-s' option is called, then the states of the output automaton
(which will not be minimized) will be labeled as lists of integers, which
specify the corresponding subsets of the states of the input automaton.
The other options are standard.
This is in fact a special case of 'nfadeterminize'.
A midfa (multiple initial state deterministic finite state automaton) is read
from 'stdin' or <filename>, and a minimized deterministic automaton that
accepts the same language is output to 'stdout' or to
<filename>'.midfadeterminize'.
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.