Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

groups atlas error without wget or GAP io #1516

Closed
benlorenz opened this issue Aug 25, 2022 · 18 comments
Closed

groups atlas error without wget or GAP io #1516

benlorenz opened this issue Aug 25, 2022 · 18 comments
Labels
bug Something isn't working topic: groups

Comments

@benlorenz
Copy link
Member

Describe the bug
On julia nightly I got an error in the Atlas groups testgroup. I think I have seen this at least once in the julia nanosoldier tests and I just got this on munk with julia nightly on macos (while investigating the polyhedral geometry timing tests).

Atlas groups: Error During Test at /Users/lorenz/nightly/Oscar.jl/test/Groups/libraries.jl:136
  Test threw exception
  Expression: order(atlas_group(PermGroup, "A5")) == 60
  the group atlas does not provide a representation of type PermGroup for A5
  Stacktrace:
   [1] error(s::String)
     @ Base ./error.jl:35
   [2] atlas_group(#unused#::Type{PermGroup}, name::String)
     @ Oscar ~/nightly/Oscar.jl/src/Groups/libraries/atlasgroups.jl:47
   [3] macro expansion
     @ /Volumes/Julia-c3d500984f/Julia-1.9.app/Contents/Resources/julia/share/julia/stdlib/v1.9/Test/src/Test.jl:477 [inlined]
   [4] macro expansion
     @ ~/nightly/Oscar.jl/test/Groups/libraries.jl:136 [inlined]
   [5] macro expansion
     @ /Volumes/Julia-c3d500984f/Julia-1.9.app/Contents/Resources/julia/share/julia/stdlib/v1.9/Test/src/Test.jl:1492 [inlined]
   [6] top-level scope
     @ ~/nightly/Oscar.jl/test/Groups/libraries.jl:136
Atlas groups: Error During Test at /Users/lorenz/nightly/Oscar.jl/test/Groups/libraries.jl:137
  Test threw exception
  Expression: order(atlas_group(MatrixGroup, "A5")) == 60
  the group atlas does not provide a representation of type MatrixGroup for A5
  Stacktrace:
   [1] error(s::String)
     @ Base ./error.jl:35
   [2] atlas_group(#unused#::Type{MatrixGroup}, name::String)
     @ Oscar ~/nightly/Oscar.jl/src/Groups/libraries/atlasgroups.jl:47
   [3] macro expansion
     @ /Volumes/Julia-c3d500984f/Julia-1.9.app/Contents/Resources/julia/share/julia/stdlib/v1.9/Test/src/Test.jl:477 [inlined]
   [4] macro expansion
     @ ~/nightly/Oscar.jl/test/Groups/libraries.jl:137 [inlined]
   [5] macro expansion
     @ /Volumes/Julia-c3d500984f/Julia-1.9.app/Contents/Resources/julia/share/julia/stdlib/v1.9/Test/src/Test.jl:1492 [inlined]
   [6] top-level scope
     @ ~/nightly/Oscar.jl/test/Groups/libraries.jl:136
Atlas groups: Error During Test at /Users/lorenz/nightly/Oscar.jl/test/Groups/libraries.jl:134
  Got exception outside of a @test
  the group atlas does not provide a representation for Dict{Symbol, Any}(:dim => 4, :repname => "A5G1-f2r4aB0", :name => "A5", :base_ring_iso => Map from
  Galois field with characteristic 2 to GAP: GF(2) defined by a julia-function with inverse)
  Stacktrace:
    [1] error(s::String)
      @ Base ./error.jl:35
    [2] atlas_group(info::Dict{Symbol, Any})
      @ Oscar ~/nightly/Oscar.jl/src/Groups/libraries/atlasgroups.jl:77
    [3] macro expansion
      @ ~/nightly/Oscar.jl/test/Groups/libraries.jl:203 [inlined]
    [4] macro expansion
      @ /Volumes/Julia-c3d500984f/Julia-1.9.app/Contents/Resources/julia/share/julia/stdlib/v1.9/Test/src/Test.jl:1492 [inlined]
    [5] top-level scope
      @ ~/nightly/Oscar.jl/test/Groups/libraries.jl:136
    [6] include(fname::String)
      @ Base.MainInclude ./client.jl:470
    [7] top-level scope
      @ ~/nightly/Oscar.jl/test/Groups/runtests.jl:15
    [8] include(fname::String)
      @ Base.MainInclude ./client.jl:470
    [9] top-level scope
      @ ~/nightly/Oscar.jl/test/runtests.jl:15
   [10] include(fname::String)
      @ Base.MainInclude ./client.jl:470
   [11] top-level scope
      @ none:6
   [12] eval                        
      @ ./boot.jl:370 [inlined]
   [13] exec_options(opts::Base.JLOptions)
      @ Base ./client.jl:272
   [14] _start()
      @ Base ./client.jl:514
Test Summary: | Pass  Error  Total  Time
Atlas groups  |   23      3     26  4.2s
ERROR: LoadError: Some tests did not pass: 23 passed, 0 failed, 3 errored, 0 broken.
in expression starting at /Users/lorenz/nightly/Oscar.jl/test/Groups/libraries.jl:134
in expression starting at /Users/lorenz/nightly/Oscar.jl/test/Groups/runtests.jl:15
in expression starting at /Users/lorenz/nightly/Oscar.jl/test/runtests.jl:15
ERROR: Package Oscar errored during testing

To Reproduce
It seems reproducible on munk right now (at least in my environment) but I don't know what is needed exactly to make this happen as I haven't been able to reproduce it anywhere else.

using Pkg;
Pkg.test("Oscar");

System (please complete the following information):

julia> Oscar.versioninfo(full=true)
OSCAR version 0.10.1 - #bl/testing/timings, 8e8d41c0d -- 2022-08-25 12:11:41 +0200
  combining:
    AbstractAlgebra.jl   v0.27.3
    GAP.jl               v0.8.2
    Hecke.jl             v0.15.0
    Nemo.jl              v0.32.2
    Polymake.jl          v0.8.0
    Singular.jl          v0.12.0
  building on:
    Antic_jll               v0.201.500+0
    Arb_jll                 v200.2300.0+0
    Calcium_jll             v0.401.100+0
    FLINT_jll               v200.900.1+0
    GAP_jll                 v400.1192.2+1
    Singular_jll            v403.100.100+0
    libpolymake_julia_jll   v0.8.1+0
    libsingular_julia_jll   v0.25.0+0
    msolve_jll              v0.4.4+0
    polymake_jll            v400.700.1+1
See `]st -m` for a full list of dependencies.

Julia Version 1.9.0-DEV.1179
Commit c3d500984fd (2022-08-25 06:54 UTC)
Platform Info:
  OS: macOS (x86_64-apple-darwin21.4.0)
  CPU: 8 × Apple M1
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-14.0.5 (ORCJIT, westmere)
  Threads: 1 on 8 virtual cores
Official https://julialang.org/ release
@benlorenz benlorenz added the bug Something isn't working label Aug 25, 2022
@ThomasBreuer
Copy link
Member

atlas_group tries to access remote data at runtime, from http://atlas.math.rwth-aachen.de/Atlas/.
If the data transfer fails then one gets an error.
Would it be enough to add another test that shows a text message about the problem, in order to understand the failures?

And do I have to worry because the problem occurs only in julia nightly?

@benlorenz
Copy link
Member Author

benlorenz commented Sep 1, 2022

I just did some more tries and on munk (Apple M1) it also happens with 1.8.0 (aarch64) and 1.7.3 (x86_64), but I still cannot reproduce it anywhere else. Julia nightly on Linux (openSUSE) seems to work fine.

It would be interesting to get an error string why the network access fails, as I don't see any reason why this would happen, running curl on the base-url works fine.

The corresponding call in GAP is probably something like this?

julia> GAP.prompt()
gap> AtlasGroup( "A5", IsPermGroup, true );                           
fail

@ThomasBreuer
Copy link
Member

Yes, GAP uses fail in order to express that (currently) the requested information cannot be returned.

One could also return nothing instead of an throwing an exception, but the idea was that the various group libraries should behave consistently, and for example small_group(4, 3) also throws an error: There are only 2 groups of order 4, up to isomorphism.
Of course one can argue that it is a different situation whether the desired object does not exist or it is perhaps available but just currently not reachable. With the current setup, we cannot distinguish these two aspects in general.

However, in the tests it is quite probable that data transfer problems are the reason for the failures. The question is then whether we change the test setup such that potential failures of this kind become invisible or whether we make it just easier to detect that these failures are harmless, for example via a text message that is shown in the test protocol.

@benlorenz
Copy link
Member Author

benlorenz commented Sep 1, 2022

It would be great to get a different error if there is a network error. Maybe also for the case where the dependencies of atlasrep are not available.

As it turns out the problem on munk is that there is no wget installed (and the GAP package io is also not available).
We could work around this by adding a wrapper for wget_jll (using BinaryWrappers) to $PATH in the GAP.jl __init__ function, then wget should always work.

@benlorenz benlorenz changed the title groups atlas error on julia nightly groups atlas error without wget or GAP io Sep 1, 2022
@hannes14
Copy link
Member

hannes14 commented Sep 1, 2022

wget on munk is at /opt/homebrew/bin/wget

@benlorenz
Copy link
Member Author

wget on munk is at /opt/homebrew/bin/wget

Ok, thanks, I will add this to my path.
But in this case I think it makes sense to fix this in GAP / GAP.jl as other people might also encounter this without having a wget binary somewhere that is just missing from the path.

@ThomasBreuer
Copy link
Member

We are working on a way to simplify the download issues in GAP (see gap-packages/utils/pull/48), and with this feature we will then be able to use Julia's Download.jl.

@fingolfin
Copy link
Member

@ThomasBreuer I think this can be closed with the new GAP.jl and GAP 4.12.1 in place?

@ThomasBreuer
Copy link
Member

Yes.
Since the changes from oscar-system/GAP.jl/pull/820 are available in GAP.jl and since version 0.77 of GAP's utils package is available via GAP 4.12.1, the download is handled via Julia's Download.jl.

@benlorenz
Copy link
Member Author

It does seem to work at first look:

lorenz@munk Oscar-tmp.jl % julia-18
               _
   _       _ _(_)_     |  Documentation: https://docs.julialang.org
  (_)     | (_) (_)    |
   _ _   _| |_  __ _   |  Type "?" for help, "]?" for Pkg help.
  | | | | | | |/ _` |  |
  | | |_| | | | (_| |  |  Version 1.8.2 (2022-09-29)
 _/ |\__'_|_|_|\__'_|  |  Official https://julialang.org/ release
|__/                   |

(@v1.8) pkg> activate .
  Activating project at `~/Oscar-tmp.jl`

julia> using Oscar
 -----    -----    -----      -      -----   
|     |  |     |  |     |    | |    |     |  
|     |  |        |         |   |   |     |  
|     |   -----   |        |     |  |-----   
|     |        |  |        |-----|  |   |    
|     |  |     |  |     |  |     |  |    |   
 -----    -----    -----   -     -  -     -  

...combining (and extending) ANTIC, GAP, Polymake and Singular
Version 0.11.0 ... 
 ... which comes with absolutely no warranty whatsoever
Type: '?Oscar' for more information
(c) 2019-2022 by The OSCAR Development Team

julia> run(`wget`);
ERROR: IOError: could not spawn `wget`: no such file or directory (ENOENT)
Stacktrace:
<....>

julia> GAP.prompt()
gap> AtlasGroup( "A5", IsPermGroup, true );
Group([ (1,2)(3,4), (1,3,5) ])

But if I don't try to run wget first I get a segfault on munk:

julia> using Oscar
 -----    -----    -----      -      -----   
|     |  |     |  |     |    | |    |     |  
|     |  |        |         |   |   |     |  
|     |   -----   |        |     |  |-----   
|     |        |  |        |-----|  |   |    
|     |  |     |  |     |  |     |  |    |   
 -----    -----    -----   -     -  -     -  

...combining (and extending) ANTIC, GAP, Polymake and Singular
Version 0.11.0 ... 
 ... which comes with absolutely no warranty whatsoever
Type: '?Oscar' for more information
(c) 2019-2022 by The OSCAR Development Team

julia> GAP.prompt()
gap> AtlasGroup( "A5", IsPermGroup, true );

signal (11): Segmentation fault: 11
in expression starting at REPL[3]:1
ReadEvalCommand at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
READ_INNER at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
FuncREAD at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
EvalFunccall1args at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
EvalUnknownBool at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
EvalNot at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
ExecIf at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
ExecSeqStat4 at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
EXEC_CURR_FUNC at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
DoExecFunc1args at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
Allocations: 51633411 (Pool: 51581065; Big: 52346); GC: 18

signal (11): Segmentation fault: 11
in expression starting at REPL[3]:1
DoExecFunc0args at /Users/lorenz/.julia/artifacts/fb4093a2787ae817d48f3ac7c1e1b99712456dbf/lib/libgap.8.dylib (unknown line)
_call_gap_func at /Users/lorenz/.julia/packages/GAP/4OJWm/src/ccalls.jl:251
call_gap_func_nokw at /Users/lorenz/.julia/packages/GAP/4OJWm/src/ccalls.jl:224 [inlined]
GapObj at /Users/lorenz/.julia/packages/GAP/4OJWm/src/ccalls.jl:234
unknown function (ip: 0x2cb7600db)
ijl_apply_generic at /Applications/Julia-1.8.app/Contents/Resources/julia/lib/julia/libjulia-internal.1.8.dylib (unknown line)
#1 at /Users/lorenz/.julia/packages/GAP/4OJWm/src/GAP.jl:234
unknown function (ip: 0x2d37e014f)
ijl_apply_generic at /Applications/Julia-1.8.app/Contents/Resources/julia/lib/julia/libjulia-internal.1.8.dylib (unknown line)
_atexit at ./initdefs.jl:372
unknown function (ip: 0x2d37c81df)
ijl_apply_generic at /Applications/Julia-1.8.app/Contents/Resources/julia/lib/julia/libjulia-internal.1.8.dylib (unknown line)
ijl_atexit_hook at /Applications/Julia-1.8.app/Contents/Resources/julia/lib/julia/libjulia-internal.1.8.dylib (unknown line)
ijl_exit at /Applications/Julia-1.8.app/Contents/Resources/julia/lib/julia/libjulia-internal.1.8.dylib (unknown line)
jl_exit_thread0_cb at /Applications/Julia-1.8.app/Contents/Resources/julia/lib/julia/libjulia-internal.1.8.dylib (unknown line)
Allocations: 51633411 (Pool: 51581065; Big: 52346); GC: 18

Even just calling Oscar.versioninfo() before running the gap commands seems to avoid that crash. Maybe some initialization in the julia IO stuff is not done properly when called from GAP?

@ThomasBreuer
Copy link
Member

Thanks @benlorenz.

Unfortunately I cannot reproduce the problem.
I have installed Julia 1.8.2, updated the packages, tried your input (using Oscar; GAP.prompt() and then AtlasGroup( "A5", IsPermGroup, true ); at the GAP prompt), and got the group in question.

Could you please try the following GAP input after calling GAP.prompt()?

GAPInfo.PackagesLoaded.utils;
GAPInfo.PackagesLoaded.atlasrep;
Download_Methods[1];
SetInfoLevel( InfoUtils, 2 );
AtlasGroup( "A5", IsPermGroup, true );

With this input, I get the following session.

gap> GAPInfo.PackagesLoaded.utils;
[ "/home/sam/.julia/scratchspaces/c863536a-3901-11e9-33e7-d5cd0df7b904/gap_129\
80177691622621448_1.8/pkg/utils", "0.77", "utils", true ]
gap> GAPInfo.PackagesLoaded.atlasrep;
[ "/home/sam/.julia/scratchspaces/c863536a-3901-11e9-33e7-d5cd0df7b904/gap_129\
80177691622621448_1.8/pkg/atlasrep", "2.1.6", "AtlasRep", true ]
gap> Download_Methods[1];
rec( download := <Julia: #2>, isAvailable := function( arg... ) ... end, 
  name := "via Julia's Downloads.download" )
gap> SetInfoLevel( InfoUtils, 2 );
gap> AtlasGroup( "A5", IsPermGroup, true );
#I  calling 'Download' with url 'http://atlas.math.rwth-aachen.de/Atlas/alt/A5/mtx/A5G1-p5B0.m1'
#I  try Download method via Julia's Downloads.download
#I  calling 'Download' with url 'http://atlas.math.rwth-aachen.de/Atlas/alt/A5/mtx/A5G1-p5B0.m2'
#I  try Download method via Julia's Downloads.download
Group([ (1,2)(3,4), (1,3,5) ])

@fingolfin fingolfin reopened this Nov 14, 2022
@fingolfin
Copy link
Member

@ThomasBreuer this is on munk, an ARM 64 Mac (i.e. with Apple's M1 CPU) server we run in Kaiserslauten. If you want you can get SSH access to it, email @hannes14 (with CC to me perhaps) and we can arrange that.

@benlorenz but I also can't reproduce this on munk, though I only tried with Oscar master for now

@benlorenz
Copy link
Member Author

I got the following output:

julia> GAP.prompt()
gap> GAPInfo.PackagesLoaded.utils;
[ "/Users/lorenz/.julia/scratchspaces/c863536a-3901-11e9-33e7-d5cd0df7b904/gap_14782291787692535528_1.8\
/pkg/utils", "0.77", "utils", true ]
gap> GAPInfo.PackagesLoaded.atlasrep;
[ "/Users/lorenz/.julia/scratchspaces/c863536a-3901-11e9-33e7-d5cd0df7b904/gap_14782291787692535528_1.8\
/pkg/atlasrep", "2.1.6", "AtlasRep", true ]
gap> Download_Methods[1];
rec( download := <Julia: #2>, isAvailable := function( arg... ) ... end, 
  name := "via Julia's Downloads.download" )
gap> SetInfoLevel( InfoUtils, 2 );
gap> AtlasGroup( "A5", IsPermGroup, true );
#I  try Download method via Julia's Downloads.download
#I  try Download method via Julia's Downloads.download
Group([ (1,2)(3,4), (1,3,5) ]) 

I triggered the crash once more directly before trying these commands but now I cannot reproduce it anymore (even with a fresh depot). Maybe I had some broken files from my previous experiments in .julia.
This was with the current master and only reproducible on munk, and I needed to remove wget from my path to trigger it (i.e. remove /opt/homebrew/bin from $PATH before starting julia).

@fingolfin
Copy link
Member

@benlorenz I strongly assume you did all this, but to be 100% sure:

  • did you run Pkg.update() ?
  • what does Oscar.versioninfo(full=true) give?
  • what is the output of GAP.Globals.ShowPackageInformation() ?

Here is what I get: This is with the system wide julia on munk:

$ julia --proj=p/1.8
               _
   _       _ _(_)_     |  Documentation: https://docs.julialang.org
  (_)     | (_) (_)    |
   _ _   _| |_  __ _   |  Type "?" for help, "]?" for Pkg help.
  | | | | | | |/ _` |  |
  | | |_| | | | (_| |  |  Version 1.8.2 (2022-09-29)
 _/ |\__'_|_|_|\__'_|  |  Official https://julialang.org/ release
|__/                   |

julia> using Oscar
 -----    -----    -----      -      -----
|     |  |     |  |     |    | |    |     |
|     |  |        |         |   |   |     |
|     |   -----   |        |     |  |-----
|     |        |  |        |-----|  |   |
|     |  |     |  |     |  |     |  |    |
 -----    -----    -----   -     -  -     -

...combining (and extending) ANTIC, GAP, Polymake and Singular
Version 0.11.0 ...
 ... which comes with absolutely no warranty whatsoever
Type: '?Oscar' for more information
(c) 2019-2022 by The OSCAR Development Team

julia> GAP.prompt()
Group([ (1,2)(3,4), (1,3,5) ])

I also verified in a separate session that it download via our new GAP download method, which goes through Julia's download():

julia> GAP.prompt()
gap> SetInfoLevel(InfoUtils, 2);
gap> AtlasGroup( "A5", IsPermGroup, true );
#I  try Download method via Julia's Downloads.download
#I  try Download method via Julia's Downloads.download
Group([ (1,2)(3,4), (1,3,5) ])

@benlorenz
Copy link
Member Author

benlorenz commented Nov 14, 2022

Before posting this I did remove the Manifest.toml in the Oscar directory, ran a git pull (on master) and did ]up using that folder as a project. So apart from stuff in .julia this should have been clean, versioninfo and package info now shows:

julia> Oscar.versioninfo(full=true)
OSCAR version 0.11.0 - #master, 4df1ca5ff -- 2022-11-14 10:19:40 +0100
  combining:
    AbstractAlgebra.jl   v0.27.6
    GAP.jl               v0.9.0
    Hecke.jl             v0.15.8
    Nemo.jl              v0.32.4
    Polymake.jl          v0.8.2
    Singular.jl          v0.14.0
  building on:
    Antic_jll               v0.201.500+0
    Arb_jll                 v200.2300.0+0
    Calcium_jll             v0.401.100+0
    FLINT_jll               v200.900.1+0
    GAP_jll                 v400.1200.101+0
    Singular_jll            v403.101.400+0
    libpolymake_julia_jll   v0.8.3+0
    libsingular_julia_jll   v0.27.0+0
    polymake_jll            v400.700.1+1
See `]st -m` for a full list of dependencies.

Julia Version 1.8.2
Commit 36034abf260 (2022-09-29 15:21 UTC)
Platform Info:
  OS: macOS (arm64-apple-darwin21.3.0)
  CPU: 8 × Apple M1
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-13.0.1 (ORCJIT, apple-m1)
  Threads: 1 on 4 virtual cores
Official https://julialang.org/ release

julia> GAP.Globals.ShowPackageInformation()
 Packages:   AClib 1.3.2, Alnuth 3.2.1, AtlasRep 2.1.6, AutoDoc 2022.10.20, AutPGrp 1.11, 
             Browse 1.8.18, CRISP 1.4.5, Cryst 4.1.25, CrystCat 1.1.10, CTblLib 1.3.4, FactInt 1.6.3, 
             ferret 1.0.9, FGA 1.4.0, Forms 1.2.9, GAPDoc 1.6.6, GUAVA 3.17, IRREDSOL 1.4.3, 
             JuliaInterface 0.9.0, LAGUNA 3.9.5, PackageManager 1.3.2, Polenta 1.3.10, Polycyclic 2.16, 
             PrimGrp 3.4.2, RadiRoot 2.9, ResClasses 4.7.3, SmallGrp 1.5, SONATA 2.9.5, Sophus 1.27, 
             SpinSym 1.5.2, TomLib 1.2.9, TransGrp 3.6.3, utils 0.77, Wedderga 4.10.2
 Try '??help' for help. See also '?copyright', '?cite' and '?authors'

Since I cannot reproduce this I think we can close this until I or someone else can trigger this again.

@ThomasBreuer
Copy link
Member

ThomasBreuer commented Nov 14, 2022

Just one more remark:
In order to make sure that a download is happening at all, the call SetInfoLevel(InfoUtils, 2); in GAP is helpful.
Once one has downloaded the necessary files, they get cached (if the directory in question is writeable), thus a subsequent call AtlasGroup( "A5", IsPermGroup, true ); will not try to download the group generators again. The info message #I try Download method via Julia's Downloads.download expresses that (1) a download is triggered and (2) the intended download method is used for that.

@fingolfin
Copy link
Member

@ThomasBreuer but I also note that in practice, for GAP.jl / Oscar.jl, the caching is always off, because the atlasrep package dir is read-only... This is one of the things which motivated me to open gap-system/gap#5190

@fingolfin
Copy link
Member

@benlorenz thank you. Yeah, I expected that you were thorough; but we all sometimes make mistakes (well, I certainly do, far too often sigh), so I wanted to check (also to see if there was any difference to my setup).

Anyway: since we can't reproduce this right now, let's indeed close, and re-open if anyone sees a similar problem again.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working topic: groups
Projects
None yet
Development

No branches or pull requests

4 participants