We write tokens (hence LP files) Gurobi cannot read if variable names are long enough

[emma58]

Summary

I'm not actually sure that this really is a bug because the use case is ridiculous, but at least for the sake of documentation, Gurobi has a 255 character limit on tokens. With symbolic_solver_labels set to True we will happily violate this limit.

(Please don't ask me how I know. :P)

Steps to reproduce the issue

In the following script, the first solve is fine because we make sane names. The second solve gives a Gurobi error, recorded below.

from pyomo.environ import *

m = ConcreteModel()
m.s = Set(initialize=["apologies",
                      "This_is_a_name_with_more_than_255_characters_which_I_agree_is_a_totally_ridiculous_thing_to_do_so_why_did_I_do_it_you_might_ask_but_naturally_I_would_dodge_the_question_so_you_will_have_to_imagine_the_reason_which_honestly_should_be_way_more_fun_than_knowing_the_unfortunate_reality"])
m.x = Var(m.s, bounds=(0, 10))

m.obj = Objective(expr=sum(m.x[i] for i in m.s), sense=maximize)

SolverFactory('gurobi').solve(m, tee=True)

# uh oh
SolverFactory('gurobi').solve(m, symbolic_solver_labels=True, tee=True)

Yields the output:

Read LP format model from file /tmp/tmpagj3adcl.pyomo.lp
Reading time = 0.00 seconds
x1: 1 rows, 3 columns, 1 nonzeros
Gurobi Optimizer version 12.0.0 build v12.0.0rc1 (linux64 - "CentOS Linux 7 (Core)")

CPU model: 13th Gen Intel(R) Core(TM) i7-1365U, instruction set [SSE2|AVX|AVX2]
Thread count: 4 physical cores, 4 logical processors, using up to 4 threads

Optimize a model with 1 rows, 3 columns and 1 nonzeros
Model fingerprint: 0x08b875f2
Coefficient statistics:
  Matrix range     [1e+00, 1e+00]
  Objective range  [1e+00, 1e+00]
  Bounds range     [1e+00, 1e+01]
  RHS range        [1e+00, 1e+00]
Presolve removed 1 rows and 3 columns
Presolve time: 0.00s
Presolve: All rows and columns removed
Iteration    Objective       Primal Inf.    Dual Inf.      Time
       0    2.0000000e+01   0.000000e+00   0.000000e+00      0s

Solved in 0 iterations and 0.00 seconds (0.00 work units)
Optimal objective  2.000000000e+01
Error reading LP format file /tmp/tmp7mef0crh.pyomo.lp at line 5
Token contains more than 255 characters
Neighboring tokens: " max obj: +1 x(apologies) +1 x(This_is_a_name_with_more_than_255_characters_which_I_agree_is_a_totally_ridiculous_thing_to_do_so_why_did_I_do_it_you_might_ask_but_naturally_I_would_dodge_the_question_so_you_will_have_to_imagine_the_reason_which_honestly_should_be_way_more_fun_than_kno "

Traceback (most recent call last):
  File "/home/esjohn/src/capsiid/capsiid-interdiction/nihba/ha.py", line 13, in <module>
    SolverFactory('gurobi').solve(m, symbolic_solver_labels=True, tee=True)
  File "/home/esjohn/src/pyomo/pyomo/opt/base/solvers.py", line 610, in solve
    _status = self._apply_solver()
              ^^^^^^^^^^^^^^^^^^^^
  File "/home/esjohn/src/pyomo/pyomo/solvers/plugins/solvers/GUROBI.py", line 689, in _apply_solver
    self._soln = GUROBI_RUN.gurobi_run(
                 ^^^^^^^^^^^^^^^^^^^^^^
  File "/home/esjohn/src/pyomo/pyomo/solvers/plugins/solvers/GUROBI_RUN.py", line 66, in gurobi_run
    model = read(model_file)
            ^^^^^^^^^^^^^^^^
  File "src/gurobipy/gurobi.pxi", line 336, in gurobipy._core.read
  File "src/gurobipy/gurobi.pxi", line 121, in gurobipy._core.gurobi.read
gurobipy._exception.GurobiError: Unable to read file

Information on your system

Pyomo version: main
Python version: 3.11
Operating system: linux
How Pyomo was installed (PyPI, conda, source): source
Solver (if applicable): Gurobi

Additional information

For the LP interface the error message is clear and it's easy enough to fix on the user side that this probably doesn't matter. But I thought I might note it because it is a bit more opaque in a direct interface (though the Gurobi error in that case too at least cites the problem, though it does not give the name that is causing it.)

[michaelbynum]

I'm not sure I see a way to fix this...