use libGAP in MatrixGroup.as_permutation_group()

[dimpase]

Currently, one uses pexpect interface there to convert a libGAP matrix group to strings, feed it to GAP's pexpect interface, etc.

This ticket will streamline this. One still will need one conversion to pexpect GAP, at the end, to feed it to PermutationGroup().

[dimpase]

I am marking this as a defect, as without this change some tests mysteriously fail on a new libgap interface, see #22626 and #26856.

---

New commits:

​db3f63e

more libGAP in as_permutation(), less pexpect mess

[soehms]

The following example which worked with the former version, is causing a string buffer overflow, right now:

sage: S = Sp(6,3)
sage: S.as_permutation_group()
Traceback (most recent call last):
...
TypeError: Gap terminated unexpectedly while reading in a large line:

My suggestion to avoid this can be seen from the following diff:

src/sage/groups/matrix_gps/finitely_generated.py

@@ -613 +613 @@
         iso=self._libgap_().IsomorphismPermGroup()
         if algorithm == "smaller":
             iso=iso.Image().SmallerDegreePermutationRepresentation()
-        PG = PermutationGroup(map(gap, iso.Image().GeneratorsOfGroup()), \
+        PG = PermutationGroup(iso.Image().GeneratorsOfGroup().sage(), \
                        canonicalize=False) # applying gap() - as PermutationGroup is not libGAP
 
         def permutation_group_map(element):
-            return iso.ImageElm(element.gap())._gap_()
+            return PG(iso.ImageElm(element.gap()).sage())
 
         from sage.categories.homset import Hom
         self._permutation_group_morphism = Hom(self, PG)(permutation_group_map)
 

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​2d77eee	follow up pyflake hints (all but one)
​7dd5fa6	reviewer's example and fix

[dimpase]

the last 2 commits add this fix, the corresponding Sp(6,3) example, and also trims the code following pyflake plugin bot hints.

[soehms]

Coming back to my last question in comment 15 of #25706: The aim of that question is the following: What does prevent us to move as_permutation_group to, say GroupMixinLibGAP replacing both implementations?

[dimpase]

Replying to soehms:

Coming back to my last question in comment 15 of #25706: The aim of that question is the following: What does prevent us to move as_permutation_group to, say GroupMixinLibGAP replacing both implementations?

I don't know. If you can get something converging towards removing pexpect gap here, it would be great.

[embray]

This tickets were closed as fixed after the Sage 8.5 release.

[chapoton]

This is breaking some doctests under python3 in the matrix_gps folder (where all tests were passing).

[embray]

There aren't any obvious Python 3-isms in this patch so if that's true that really speaks to the fragility of this code; it looks like it needs quite a bit more cleaning up in general. Please open a new ticket.

[chapoton]

traceback:

File "src/sage/groups/matrix_gps/finitely_generated.py", line 445, in sage.groups.matrix_gps.finitely_generated.FinitelyGeneratedMatrixGroup_generic.__reduce__
Failed example:
    loads(dumps(G)) == G
Expected:
    True
Got:
    RuntimeError: Syntax error: ; expected in stream:1
    Symbolic Ring;
     ^^^^^^^^^^^^
    Error, Variable: 'Complex' must have a value
    <BLANKLINE>
    The above exception was the direct cause of the following exception:
    <BLANKLINE>
    Traceback (most recent call last):
      File "sage/libs/gap/util.pyx", line 506, in sage.libs.gap.util.extract_libgap_errout (build/cythonized/sage/libs/gap/util.c:6644)
    SystemError: <built-in method replace of str object at 0x7f1666da6bb0> returned a result with an error set
    Exception ignored in: 'sage.libs.gap.util.error_handler'
    Traceback (most recent call last):
      File "sage/libs/gap/util.pyx", line 506, in sage.libs.gap.util.extract_libgap_errout (build/cythonized/sage/libs/gap/util.c:6644)
    SystemError: <built-in method replace of str object at 0x7f1666da6bb0> returned a result with an error set
    RuntimeError: Syntax error: ; expected in stream:1
    Complex Field with 53 bits of precision;
     ^^^^^^^^^^^^^^^^^^^^
    Error, Variable: 'bits' must have a value
    <BLANKLINE>
    The above exception was the direct cause of the following exception:
    <BLANKLINE>
    Traceback (most recent call last):
      File "sage/libs/gap/util.pyx", line 506, in sage.libs.gap.util.extract_libgap_errout (build/cythonized/sage/libs/gap/util.c:6644)
    SystemError: <built-in method replace of str object at 0x7f16671a8e00> returned a result with an error set
    Exception ignored in: 'sage.libs.gap.util.error_handler'
    Traceback (most recent call last):
      File "sage/libs/gap/util.pyx", line 506, in sage.libs.gap.util.extract_libgap_errout (build/cythonized/sage/libs/gap/util.c:6644)
    SystemError: <built-in method replace of str object at 0x7f16671a8e00> returned a result with an error set
    True
**********************************************************************
File "src/sage/groups/matrix_gps/finitely_generated.py", line 450, in sage.groups.matrix_gps.finitely_generated.FinitelyGeneratedMatrixGroup_generic.__reduce__
Failed example:
    R = MatrixSpace(SR, 2)
Expected nothing
Got:
    RuntimeError: Error, Variable: 'Complex' must have a value
    <BLANKLINE>
    The above exception was the direct cause of the following exception:
    <BLANKLINE>
    Traceback (most recent call last):
      File "sage/libs/gap/util.pyx", line 506, in sage.libs.gap.util.extract_libgap_errout (build/cythonized/sage/libs/gap/util.c:6644)
    SystemError: <built-in method replace of str object at 0x7f1666da6bb0> returned a result with an error set
    Exception ignored in: 'sage.libs.gap.util.error_handler'
    Traceback (most recent call last):
      File "sage/libs/gap/util.pyx", line 506, in sage.libs.gap.util.extract_libgap_errout (build/cythonized/sage/libs/gap/util.c:6644)
    SystemError: <built-in method replace of str object at 0x7f1666da6bb0> returned a result with an error set
    RuntimeError: Syntax error: ; expected in stream:1
    Complex Field with 53 bits of precision;
     ^^^^^^^^^^^^^^^^^^^^
    Error, Variable: 'bits' must have a value
    <BLANKLINE>
    The above exception was the direct cause of the following exception:
    <BLANKLINE>
    Traceback (most recent call last):
      File "sage/libs/gap/util.pyx", line 506, in sage.libs.gap.util.extract_libgap_errout (build/cythonized/sage/libs/gap/util.c:6644)
    SystemError: <built-in method replace of str object at 0x7f16671a8e00> returned a result with an error set
    Exception ignored in: 'sage.libs.gap.util.error_handler'
    Traceback (most recent call last):
      File "sage/libs/gap/util.pyx", line 506, in sage.libs.gap.util.extract_libgap_errout (build/cythonized/sage/libs/gap/util.c:6644)
    SystemError: <built-in method replace of str object at 0x7f16671a8e00> returned a result with an error set
**********************************************************************
File "src/sage/groups/matrix_gps/finitely_generated.py", line 584, in sage.groups.matrix_gps.finitely_generated.FinitelyGeneratedMatrixGroup_gap.as_permutation_group
Failed example:
    G = MatrixGroup(map(MS, GG.GeneratorsOfGroup()))
Exception raised:
    Traceback (most recent call last):
      File "/home/u1/chapoton/sage3/local/lib/python3.6/site-packages/sage/doctest/forker.py", line 671, in _run
        self.compile_and_execute(example, compiler, test.globs)
      File "/home/u1/chapoton/sage3/local/lib/python3.6/site-packages/sage/doctest/forker.py", line 1086, in compile_and_execute
        exec(compiled, globs)
      File "<doctest sage.groups.matrix_gps.finitely_generated.FinitelyGeneratedMatrixGroup_gap.as_permutation_group[21]>", line 1, in <module>
        G = MatrixGroup(map(MS, GG.GeneratorsOfGroup()))
      File "sage/misc/lazy_import.pyx", line 354, in sage.misc.lazy_import.LazyImport.__call__ (build/cythonized/sage/misc/lazy_import.c:3684)
        return self.get_object()(*args, **kwds)
      File "/home/u1/chapoton/sage3/local/lib/python3.6/site-packages/sage/groups/matrix_gps/finitely_generated.py", line 290, in MatrixGroup
        gens = normalize_square_matrices(gens)
      File "/home/u1/chapoton/sage3/local/lib/python3.6/site-packages/sage/groups/matrix_gps/finitely_generated.py", line 131, in normalize_square_matrices
        raise ValueError('list of plain numbers must have square integer length')
    ValueError: list of plain numbers must have square integer length