4.12 Interface to SAGE

Module: sage.interfaces.sage0

This is an expect interface to another copy of the SAGE interpreter.

Module-level Functions

reduce_load_Sage( )

reduce_load_element( s)

sage0_console( )

sage0_version( )

Class: Sage

class Sage
Expect interface to the SAGE interpreter itself.

INPUT: server - (optional); if specified runs SAGE on a remote machine with address. You must have ssh keys setup so you can login to the remote machine by typing "ssh remote_machine" and no password, e.g.: cd; ssh-keygen -t rsa; scp .ssh/id_rsa.pub remote_machine:.ssh/authorized_keys2

The version of SAGE should be the same as on the local machine, since pickling is used to move data between the two SAGE process.

We create an interface to a copy of SAGE. This copy of SAGE runs as an external process with its own memory space, etc.

sage: s = Sage()

Create the element 2 in our new copy of SAGE, and cubeit.

sage: a = s(2)
sage: a^3
8

Create a vector space of dimension $ 4$ , and compute its generators:

sage: V = s('Q^4')
sage: V.gens()
((1, 0, 0, 0), (0, 1, 0, 0), (0, 0, 1, 0), (0, 0, 0, 1))

Note that V is a not a vector space, it's a wrapper around an object (which happens to be a vector space), in another running instance of SAGE.

sage: type(V)
<class 'sage.interfaces.sage0.SageElement'>
sage: V.parent()
Sage
sage: g = V.0;  g
(1, 0, 0, 0)
sage: g.parent()
Sage

We can still get the actual parent by using the name attribute of g, which is the variable name of the object in the child process.

sage: s('%s.parent()'%g.name())
Vector space of dimension 4 over Rational Field

Note that the memory space is completely different.

sage: x = 10
sage: s('x = 5')
5
sage: x
10
sage: s('x')
5

We can have the child interpreter itself make another child SAGE process, so now three copies of SAGE are running:

sage: s3 = s('Sage()')
sage: a = s3(10)
sage: a
10

This $ a=10$ is in a subprocess of a subprocesses of your original SAGE.

sage: _ = s.eval('%s.eval("x=8")'%s3.name())
sage: s3('"x"')
8
sage: s('x')
5
sage: x
10

The double quotes are needed because the call to s3 first evaluates its arguments using the s interpeter, so the call to s3 is passed s('"x"'), which is the string "x" in the s interpreter.

Sage( self, [maxread=None], [script_subdirectory=True], [logfile=None], [preparse=user], [server=10000])

Functions: console,$  $ eval,$  $ get,$  $ new,$  $ preparse,$  $ quit,$  $ set,$  $ version

eval( self, line)

Send the code x to a second instance of the SAGE interpreter and return the output as a string.

This allows you to run two completely independent copies of SAGE at the same time in a unified way.

get( self, var)

Get the value of the variable var.

set( self, var, value)

Set the variable var to the given value.

Special Functions: __call__,$  $ __reduce__,$  $ _contains,$  $ _get_object_from_server_tmpfile,$  $ _is_true_string,$  $ _object_class,$  $ _quit_string,$  $ _remote_tmpfile,$  $ _send_tmpfile_to_server

Class: SageElement

class SageElement

Special Functions: _sage_

Class: SageFunction

class SageFunction

Special Functions: __call__,$  $ __repr__

See About this document... for information on suggesting changes.