paolodragone/pymzn

A Python wrapper for the MiniZinc tool pipeline.

pymzn's Issues

Windows support?

Hi,

I was trying to run pymzn on my windows machine and it got stuck on a Linux command. This is the traceback:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\Users\jver\AppData\Local\Programs\Python\Python35\lib\site-packages\pymzn\_mzn\_minizinc.py", line 161, in minizinc stdlib_dir=stdlib_dir)
  File "C:\Users\jver\AppData\Local\Programs\Python\Python35\lib\site-packages\pymzn\_mzn\_minizinc.py", line 295, in mzn2fzn run(args)
  File "C:\Users\jver\AppData\Local\Programs\Python\Python35\lib\site-packages\pymzn\bin.py", line 105, in run preexec_fn=os.setsid) as process:
AttributeError: module 'os' has no attribute 'setsid'

TypeError: stat: path should be string, bytes, os.PathLike or integer, not NoneType

I am running my code on python 3.6 and have installed the pymzn though pip install
When I tried to run below code it threw an error.

```

import pymzn

solution = pymzn.minizinc('seq_disjunctive.mzn','Disjunctive.dzn')

TypeError: stat: path should be string, bytes, os.PathLike or integer, not NoneType

Output error: expecting end of file

I'm new to pymzn. Just running a fairly simple MiniZinc file and I get:

Error: syntax error, unexpected output, expecting end of file

Maybe the solution is too long?

This is the content of the .mzn file:

enum spy = {American, Belgian, Cuban, Dane};
enum place = {Egypt, France, Greece, Haiti};
enum level = {retired, on_run, new, full};

var place: location_American;
var place: location_Belgian;
var place: location_Cuban;
var place: location_Dane;

var level: status_American;
var level: status_Belgian;
var level: status_Cuban;
var level: status_Dane;

constraint location_American != location_Belgian;
constraint location_American != location_Cuban;
constraint location_American != location_Dane;
constraint location_Belgian != location_Cuban;
constraint location_Belgian != location_Dane;
constraint location_Cuban != location_Dane;

constraint status_American != status_Belgian;
constraint status_American != status_Cuban;
constraint status_American != status_Dane;
constraint status_Belgian != status_Cuban;
constraint status_Belgian != status_Dane;
constraint status_Cuban != status_Dane;

% problem constraints

constraint status_Cuban > status_Belgian;
constraint location_Dane = France;
constraint status_American = on_run;

constraint (status_American = retired) = (location_American = Greece);
constraint (status_Belgian = retired) = (location_Belgian = Greece);
constraint (status_Cuban = retired) = (location_Cuban = Greece);
constraint (status_Dane = retired) = (location_Dane = Greece);

constraint (location_American = Haiti) = (status_American = new);
constraint (location_Belgian = Haiti) = (status_Belgian = new);
constraint (location_Cuban = Haiti) = (status_Cuban = new);
constraint (location_Dane = Haiti) = (status_Dane = new);

solve satisfy

Any help would be appreciated.

Incorrect preprocessing of `.mzn` files, instead of `.pmzn` files

G'day @paolodragone,

It seems like the latest version (0.18.1) is incorrectly preprocessing MZN files and treats them like PMZN file.

More specifically, while processing .mzn files, it is removing all characters following the first % character it encounters in a line, even if that character is within a string literal.

For instance, the following sample .mzn file:

% Use a large domain to simulate a long solving process
var 0 .. 100000: x;
solve satisfy;

%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%% Result %%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%
output [
        "\n%%%%%%%%%%%%% RESULTS %%%%%%%%%%%%%\n",
        "filled_x = \(x);\n",
        "\n%%%%%%%%%%%%%% END %%%%%%%%%%%%%%%\n",
       ];

gets processed into following:

var 0 .. 100000: x;
solve satisfy;




output [
        "\n
        "filled_x = \(x);\n",
        "\n
       ];

Note the empty lines above, which are basically lines where the first character was %. Though what is more important is that the processed file is now syntactically incorrectly (note the string literals in output statement); which when executed results in error (of course).

The error then thrown by that call to pymzn.minizinc('rel/path/to/sample.mzn') is:

An error occurred while executing minizinc on file

var 0 .. 100000: x;
solve satisfy;




output [
        "\n
        "filled_x = \(x);\n",
        "\n
       ];
 with command line arguments: ['--model-check-only', '-']

stdin:[line].[col]:
        "\n
           ^
Error: syntax error, unexpected end of line inside string literal

[line] and [col] are just placeholders I have put.

This is likely because pymzn.mzn.minizinc._minizinc_preliminaries() is calling pymzn.mzn.minizinc.preprocess_model() irrespective of the file extension.

System Setup:

• OS: MacOS Mojave 10.14.3
• Python: 3.7.0
• PyMZN: 0.18.1
• MiniZinc: 2.2.3
• Gecode: 6.1.0

What's the correct sequence of dict2dzn

Why the output sequence does not match the original input?

import pymzn

a_string = {'e1', 'e2', 'e3', 'e4'}

pymzn.dict2dzn({'out':a_string}, fout="test.dzn")

##out = {e2, e4, e3, e1};

I am not sure the index of the paramter is valid or not.

pymzn doesnt work with minizinc 2.5.0

Was trying to run pymzn on minizinc 2.5.0. To reproduce, we can use the default example. Looks like the new version of minizinc is adding extra information in the output that pymzn is not able to parse the output properly.

pymzn minizinc knapsack.mzn knapsack.dzn --data '{"capacity": 20}'

Traceback (most recent call last):
  File "/usr/local/bin/pymzn", line 8, in <module>
    sys.exit(main())
  File "/usr/local/lib/python3.8/dist-packages/pymzn/__init__.py", line 222, in main
    args['func'](**args)
  File "/usr/local/lib/python3.8/dist-packages/pymzn/__init__.py", line 43, in _minizinc
    solns = minizinc(
  File "/usr/local/lib/python3.8/dist-packages/pymzn/mzn/minizinc.py", line 687, in minizinc
    solns = parser.parse(proc)
  File "/usr/local/lib/python3.8/dist-packages/pymzn/mzn/output.py", line 245, in parse
    self._collect(proc, solns)
  File "/usr/local/lib/python3.8/dist-packages/pymzn/mzn/output.py", line 249, in _collect
    for soln in self._parse(proc):
  File "/usr/local/lib/python3.8/dist-packages/pymzn/mzn/output.py", line 263, in _parse
    soln = parse_lines.send(line)
  File "/usr/local/lib/python3.8/dist-packages/pymzn/mzn/output.py", line 279, in _parse_lines
    soln = dzn2dict(
  File "/usr/local/lib/python3.8/dist-packages/pymzn/dzn/parse.py", line 552, in dzn2dict
    raise ValueError(
ValueError: Unsupported parsing for statement:
'MiniZinc to FlatZinc converter, version 2.5.0, build 198486215\nCopyright (C) 2014-2020 Monash University, NICTA, Data61\nx = {3,5}'

Added output fails on enum type.

When a model contains variables whose value are of type enum, pymzn includes the enum in the output with a show function which fails on enumerated enums (that are not ranges like 1..10).
Here is a small example coming from the minizinc coursera course (temperature.mzn):

var int: cost;
array[int] of int: readings = [35,35,20,20,20];
int: start = 25;
int: n = length(readings);
set of int: HOUR = 1..n;
set of int: HOUR0 = 0..n;

array[HOUR] of var ACTION: choice;

enum ACTION = {heat, strong_heat, cool, strongly_cool, do_nothing};

array[HOUR0] of var int: temp;

array[ACTION] of int: adjust = [+1, 4, -2, -5, 0];
array[ACTION] of int: adjust_cost = [1, 5, 3, 9, 0];

constraint temp[0] = start;
constraint forall(h in HOUR) (temp[h] = (temp[h-1] + readings[h]) div 2  + adjust[choice[h]]);
constraint cost = sum(t in HOUR)(adjust_cost[choice[t]]);
constraint forall(h in HOUR) (temp[h] >= 25);
constraint forall(h in HOUR) (temp[h] <= 30);

solve minimize cost;

minizinc temperature.mzn leads to a solution while the following script fails with INCOMPLETE:

import pymzn
pymzn.minizinc("temperature.mzn", keep=True)

You can note the show(ACTION) in the output of the generated mzn file. When removing that part of the line, the model works fine.

An issue when using pymzn with gurobi solver

Hi, developers of pymzn.
These is an error occurred when I using pymzn.minizinc with gurobi solver.
s = pymzn.minizinc('placement.mzn', timeout=30, parallel=4, solver=pymzn.gurobi, data=data)
The error is:
`Traceback (most recent call last):
File "D:/Work Space/Pycharm/TensorflowTrainning/vnf_placement_optimization_rl-master/solver.py", line 81, in
env)
File "D:/Work Space/Pycharm/TensorflowTrainning/vnf_placement_optimization_rl-master/solver.py", line 42, in solver
'p_slot': env.p_slot
File "D:\Anaconda3\envs\py36tf180\lib\site-packages\pymzn\mzn\minizinc.py", line 667, in minizinc
**solver_args
File "D:\Anaconda3\envs\py36tf180\lib\site-packages\pymzn\mzn\minizinc.py", line 819, in solve
proc = _run_minizinc_proc(*args, input=input)
File "D:\Anaconda3\envs\py36tf180\lib\site-packages\pymzn\mzn\minizinc.py", line 56, in _run_minizinc_proc
return run_process(*args, input=input)
File "D:\Anaconda3\envs\py36tf180\lib\site-packages\pymzn\mzn\process.py", line 63, in run_process
shell=shell, bufsize=1, universal_newlines=True
File "D:\Anaconda3\envs\py36tf180\lib\subprocess.py", line 423, in run
with Popen(*popenargs, **kwargs) as process:
File "D:\Anaconda3\envs\py36tf180\lib\subprocess.py", line 729, in init
restore_signals, start_new_session)
File "D:\Anaconda3\envs\py36tf180\lib\subprocess.py", line 991, in _execute_child
args = list2cmdline(args)
File "D:\Anaconda3\envs\py36tf180\lib\subprocess.py", line 481, in list2cmdline
needquote = (" " in arg) or ("\t" in arg) or not arg
TypeError: argument of type 'int' is not iterable

Process finished with exit code 1
`

But, It works when I remove solver=pymzn.gurobi
Also, It run correctly on minizinc IDE with gurobi solver.

The Minizinc IDE version is 2.5.5
Gurobi version is 9.1.2
pymzn version is 0.18.3

Hope you can help me
Thanks.

enable gecode multithreading by default

Hi,

I think that it would make sense to enable gecode's parallel execution by default (it is set to single-threaded right now). This way users would get the maximum possible performance automatically, without having to mess with the arguments of fzn_gecode().

Thanks!

pymzn.minizinc timeout leads to type error

While trying to call pymzn.minizinc with a timeout argument, I get a type error:

Traceback (most recent call last):
  File "test.py", line 3, in <module>
    sol = pymzn.minizinc("cumul_sched.mzn", "cumul_sched.dzn", timeout=5)
  File "/usr/local/lib/python3.7/site-packages/pymzn/mzn/minizinc.py", line 649, in minizinc
    **solver_args
  File "/usr/local/lib/python3.7/site-packages/pymzn/mzn/minizinc.py", line 794, in solve
    proc = _run_minizinc_proc(*args, input=input)
  File "/usr/local/lib/python3.7/site-packages/pymzn/mzn/minizinc.py", line 56, in _run_minizinc_proc
    return run_process(*args, input=input)
  File "/usr/local/lib/python3.7/site-packages/pymzn/mzn/process.py", line 63, in run_process
    shell=shell, bufsize=1, universal_newlines=True
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/subprocess.py", line 472, in run
    with Popen(*popenargs, **kwargs) as process:
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/subprocess.py", line 775, in __init__
    restore_signals, start_new_session)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/subprocess.py", line 1453, in _execute_child
    restore_signals, start_new_session, preexec_fn)
TypeError: expected str, bytes or os.PathLike object, not int

This is due to this line in mzn/minizinc.py which should be

args += ['--time-limit', str(timeout * 1000)]

I will shortly send a pull-request.

NameError / typo in master

Hey,

thanks for creating this project!

When running with wait=False, I ran into a "NameError: name 'solver_proces' is not defined". The reason are typos in minizinc.py at lines 340 +341 (see

pymzn/pymzn/mzn/minizinc.py

Line 340 in a5de87f

return solver_process.readlines(), solver_proces.stderr

). In both lines the variable name should read 'solver_process', right now there is an 's' missing at the end.

Best,
-Matthias

Select mzn-g12mip instead of fzn-gecode

I think the subject is self explanatory. You see, the Minizinc tutorial uses always "mzn-g12fd" or "mzn-g12mip" and not "fzn-gecode", the problem is that with "fzn-gecode" is taking a long time to solve the problem, I would like to be able to change it.

Can't convert int object to str implicitly

Hello,

I have been using your package recently (a nice one, btw 👍 ) and just discovered something that looks to like a tiny bug. While passing timeout to pymzn.minizinc() function with the default gecode solver, I get:

Traceback (most recent call last):
  File "main.py", line 39, in <module>
    main()
  File "main.py", line 36, in main
    run_test_cases()
  File "main.py", line 28, in run_test_cases
    schedule = run_solver(**test_case, debug=True)
  File "/Users/me/my_project/minizinc_service.py", line 74, in run_solver
    timeout=1000)
  File "/usr/local/lib/python3.5/site-packages/pymzn/_mzn/_minizinc.py", line 169, in minizinc
    all_solutions=all_solutions, **solver_args)
  File "/usr/local/lib/python3.5/site-packages/pymzn/_mzn/_solvers.py", line 212, in solve
    process = run(args)
  File "/usr/local/lib/python3.5/site-packages/pymzn/bin.py", line 105, in run
    preexec_fn=os.setsid) as process:
  File "/usr/local/Cellar/python3/3.5.2_3/Frameworks/Python.framework/Versions/3.5/lib/python3.5/subprocess.py", line 947, in __init__
    restore_signals, start_new_session)
  File "/usr/local/Cellar/python3/3.5.2_3/Frameworks/Python.framework/Versions/3.5/lib/python3.5/subprocess.py", line 1490, in _execute_child
    restore_signals, start_new_session, preexec_fn)
TypeError: Can't convert 'int' object to str implicitly

As it seemed like there was a typo here, I did a quick fix, and tried it locally. It did the job for me so I'll leave a PR in few minutes in case you find it useful :) Cheers!

ValueError is raised with pymzn but working with MinizincIDE

Hi,

I am trying to use the cbc solver through pymzn. I have two files LP.mzn, Fourier.dzn. These files are running properly when I use the Minizinc IDE but the error ValueError: Value '1' is not a float. is raised with pymzn.

Here is the full error stacktrace:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-4-ab7f17e63e19> in <module>
----> 1 solution=inf555.minizinc('LP.mzn','Fourier.dzn', solver=inf555.cbc, output_node='item')
      2 
      3 solution

~/Documents/lX/2019-2020/s1/INF555-ConDes/TD9/inf555.py in minizinc(mzn, include, *dzn_files, **kwargs)
     14 def minizinc(mzn, *dzn_files, include=".", **kwargs):
     15     """Solve using minizinc."""
---> 16     return minizn(mzn, *dzn_files, include=include, **kwargs)
     17 
     18 

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/mzn/minizinc.py in minizinc(mzn, args, data, include, stdlib_dir, globals_dir, declare_enums, allow_multiple_assignments, keep, output_vars, output_base, output_mode, solver, timeout, two_pass, pre_passes, output_objective, non_unique, all_solutions, num_solutions, free_search, parallel, seed, rebase_arrays, keep_solutions, return_enums, *dzn_files, **kwargs)
    685         types=types, keep_solutions=keep_solutions, return_enums=return_enums
    686     )
--> 687     solns = parser.parse(proc)
    688     return solns
    689 

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/mzn/output.py in parse(self, proc)
    243         logger.info('Started parsing solver output.')
    244         solns = Solutions(Queue(), keep=self.keep_solutions)
--> 245         self._collect(proc, solns)
    246         return solns
    247 

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/mzn/output.py in _collect(self, proc, solns)
    247 
    248     def _collect(self, proc, solns):
--> 249         for soln in self._parse(proc):
    250             solns._queue.put(soln)
    251         logger.info('Solutions parsed: {}'.format(solns._queue.qsize()))

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/mzn/output.py in _parse(self, proc)
    261         parse_lines.send(None)
    262         for line in proc.readlines():
--> 263             soln = parse_lines.send(line)
    264             if soln is not None:
    265                 yield soln

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/mzn/output.py in _parse_lines(self)
    279                     soln = dzn2dict(
    280                         soln, rebase_arrays=self.rebase_arrays,
--> 281                         types=self.types, return_enums=self.return_enums
    282                     )
    283                 line = yield soln

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/dzn/parse.py in dzn2dict(dzn, rebase_arrays, types, return_enums)
    585             var_type = var_types.get(var, None)
    586         assign[var] = parse_value(
--> 587             val, var_type=var_type, enums=enums, rebase_arrays=rebase_arrays
    588         )
    589 

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/dzn/parse.py in parse_value(val, var_type, enums, rebase_arrays)
    433     if 'dims' in var_type:
    434         return _parse_array(
--> 435             val, rebase_arrays=rebase_arrays, var_type=var_type, enums=enums
    436         )
    437 

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/dzn/parse.py in _parse_array(val, rebase_arrays, var_type, enums, raise_errors)
    335             p_val = _parse_array_vals(
    336                 indices, vals, rebase_arrays=rebase_arrays, vals_type=vals_type,
--> 337                 enums=enums, raise_errors=raise_errors
    338             )
    339         return p_val

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/dzn/parse.py in _parse_array_vals(indices, vals, rebase_arrays, vals_type, enums, raise_errors)
    360             indices[1:], vals, rebase_arrays=rebase_arrays, vals_type=vals_type,
    361             enums=enums, raise_errors=raise_errors
--> 362         ) for i in idx_set}
    363 
    364     if rebase_arrays and list(idx_set)[0] == 1:

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/dzn/parse.py in <dictcomp>(.0)
    360             indices[1:], vals, rebase_arrays=rebase_arrays, vals_type=vals_type,
    361             enums=enums, raise_errors=raise_errors
--> 362         ) for i in idx_set}
    363 
    364     if rebase_arrays and list(idx_set)[0] == 1:

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/dzn/parse.py in _parse_array_vals(indices, vals, rebase_arrays, vals_type, enums, raise_errors)
    355             vals.pop(0), var_type=vals_type, enums=enums,
    356             raise_errors=raise_errors
--> 357         ) for i in idx_set}
    358     else:
    359         arr = {i: _parse_array_vals(

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/dzn/parse.py in <dictcomp>(.0)
    355             vals.pop(0), var_type=vals_type, enums=enums,
    356             raise_errors=raise_errors
--> 357         ) for i in idx_set}
    358     else:
    359         arr = {i: _parse_array_vals(

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/dzn/parse.py in _parse_val(val, var_type, enums, raise_errors)
    272 
    273     return _parse_val_basic_type(
--> 274         val, var_type=var_type, enums=enums, raise_errors=raise_errors
    275     )
    276 

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/dzn/parse.py in _parse_val_basic_type(val, var_type, enums, raise_errors)
    154 
    155     if var_type['type'] == 'float':
--> 156         return _parse_float(val, raise_errors=raise_errors)
    157 
    158     if raise_errors:

~/opt/anaconda3/envs/dh/lib/python3.7/site-packages/pymzn/dzn/parse.py in _parse_float(val, raise_errors)
    105         return float(val)
    106     if raise_errors:
--> 107         raise ValueError('Value \'{}\' is not a float.'.format(val))
    108     return None
    109 

ValueError: Value '1' is not a float.

The function inf555.minizinc is just a simple wrapper calling pymzn.minizinc:

def minizinc(mzn, *dzn_files, include=".", **kwargs):
    """Solve using minizinc."""
    return minizn(mzn, *dzn_files, include=include, **kwargs)

I call the solver with:

solution=inf555.minizinc('LP.mzn','Fourier.dzn', solver=inf555.cbc)

My LP.mzn file is:

include "matrix.mzn";

int: m;
int: n;
array[1..1, 1..n] of float: c;
array[1..m, 1..n] of float: A;
array[1..m, 1..1] of float: b;

array[1..n, 1..1] of var float: x;
array[1..m, 1..1] of var float: Ax;

constraint forall(i in 1..n)(x[i,1] >= 0.0);

constraint matrix_mult(A, x, Ax);

constraint matrix_geq(Ax, b);

var float: f = sum(i in 1..n)(c[1,i]*x[i,1]);

solve minimize f;

and finally my Fourier.dzn file is:

m = 9;
n = 5;

A = [|-1.0, 0.0, 0.0, 0.0, 0.0
     |0.0, -1.0, 0.0, 0.0, 0.0
     |0.0, 0.0, -1.0, 0.0, 0.0
     |1.0, 1.0, 1.0, 0.0, 0.0
     |-1.0, -1.0, -1.0, 0.0, 0.0
     |0.0, 20.0, 0.0, -2.0, 0.0
     |0.0, -20.0, 0.0, 2.0, 0.0
     |0.0, 0.0, 20.0, 0.0, -2.0
     |0.0, 0.0, -20.0, 0.0, 2.0
     |];

b = [|-1.0
     |-1.0
     |-1.0
     |2.0
     |-2.0
     |0.0
     |0.0
     |0.0
     |0.0
     |];
     
c = [|0, 0, 0, -1, 0|];

Thank you for your help!

FileNotFoundError: [Errno 2] No such file or directory: 'mzn2fzn': 'mzn2fzn'

Hello Paolo and other pymzn experts,
there is a problem that I hope someone could help me
when I use pymzn.minizinc() this function,the Traceback would tell me FileNotFoundError: [Errno 2] No such file or directory: 'mzn2fzn': 'mzn2fzn'
but in minizinc ide,there is not any problem

thank u!!!!

'CalledProcessError' object has no attribute 'stderr'

Hello,

I'm on ubuntu 14, and use python2.7 (yeah I know, we have an old-ish system). I installed pymzn by using "pip install git+https://github.com/paolodragone/pymzn.git@python2", and for minizinc, I cloned the official github repo and ran the instructions in INSTALL.txt. If I run the following code:

import pymzn
import pandas as pd 
import os

MINIZINC_PATH = "/home/hristo/libminizinc/build"
TIMEOUT = 300


skills = pd.read_csv('data/skills.csv')
people = skills.name
skills = skills.loc[:, skills.columns != 'name']
demand_forecast = pd.read_csv('data/demand_forecast.csv')
hours_num = 24
days_num = 3
people_num = 4

in_data = {
    'days_num': days_num,
    'D' : list(range(days_num)),
    'hours_num': hours_num,
    'H': list(range(hours_num)),
    'skills_num': skills.shape[1],
    'positions_num': skills.shape[1], # same as skills,
    'Pskl': skills.values,
    'Ppattern': list(range(people_num)),
    'people_pattern_num': people_num,
    'Fct_reduced': demand_forecast.values,
    'skill_pattern_counts': [15,15,15,15],
    'P_salary': list(range(people_num)) + [3,2,3,2]
}

pymzn.dict2dzn(in_data, fout='data/in_data.dzn')


solver = pymzn.Chuffed(os.path.join(MINIZINC_PATH, 'mzn-chuffed'),
                   os.path.join(MINIZINC_PATH, 'fzn-chuffed'))
res = pymzn.minizinc('simple_model_warehouse_no_crews.mzn',
                     data=in_data,
                     solver=solver, timeout=TIMEOUT)

I get the following error:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-1-13a2d44180ca> in <module>()
     38 res = pymzn.minizinc('simple_model_warehouse_no_crews.mzn',
     39                      data=in_data,
---> 40                      solver=solver, timeout=TIMEOUT)

/home/hristo/mlenv/local/lib/python2.7/site-packages/pymzn/mzn/minizinc.pyc in minizinc(mzn, *dzn_files, **solver_args)
    230             out = solver.solve(mzn_file, *dzn_files, data=data,
    231                         include=include, timeout=timeout, output_mode=u'item',
--> 232                         all_solutions=all_solutions, **solver_args)
    233             solns, complete = split_solns(out)
    234             solns = list(imap(dzn2dict, solns))

/home/hristo/mlenv/local/lib/python2.7/site-packages/pymzn/mzn/solvers.pyc in solve(self, mzn_file, *dzn_files, **kwargs)
    486                 raise RuntimeError(process.stderr)
    487         except CalledProcessError, err:
--> 488             log.exception(err.stderr)
    489             raise RuntimeError(err.stderr)
    490         return out

AttributeError: 'CalledProcessError' object has no attribute 'stderr'

i cant enable the addon

I have lilys installed (1.7.4 version) but when i try to enable the addon i get this error:
"TypeError: stat: path should be string, bytes, os.PathLike or integer, not NoneType"

i dont know anything about coding and cant find anything on google can you help me

Add an option to obtain the minizinc stats

The command line version of minizinc has an option (-s / --statistics) that outputs solver related statistics for every solution found. I didn't find any way to obtain this data when using pymzn.minizinc. Would it be possible to add an option and return this data with the resulting dictionary?

Cannot import templates

When attempting to use pyconstruct, the following line failed

pymzn.templates.add_package('pyconstruct.domains', 'share')
with error
AttributeError: module 'pymzn' has no attribute 'templates

It seems
from .templates import *
is missing from this file
https://github.com/paolodragone/pymzn/blob/master/pymzn/__init__.py

It might have gone missing in this commit:
df5aca6#diff-21aa6139d6400b1aa70e461a45c6cf63

Question: output of objective value using gurobi solver with pymzn.minizinc() function

Hello Paolo and other pymzn experts,

this is not an issue but more a question on usability of arguments in pymzn.minizinc() function.
As I could not find a forum, I hope this post here is okay.

When I use following code, I get the objective value printed out:
pymzn.minizinc(mzn_file,dzn_file,solver=pymzn.gecode,output_mode='item')
When I use Gurobi solver instead, I only get informed that the solving is "COMPLETE":
pymzn.minizinc(mzn_file,dzn_file,solver=pymzn.gurobi,output_mode='item')

I have tried several things, but do not get the desired output. For example, I failed to hand over arguments:
pymzn.minizinc(mzn_file,dzn_file,args={'arg1':'-s'},solver=pymzn.gurobi)

In fact, I really just want to have the objective value, ideally including the already outputted items of the optimal solution.
Below you can find my mzn and dzn file.
Furthermore, I might want to make use of --relGap argument. If this is not a hassle, could you please present an example usage of pymzn.minizinc() with the use of gurobi and --relGap argument?
Just equivilant to the windows command line, e.g.:
minizinc --solver gurobi -a --statistics knapsack_1.mzn knapsack_9.dzn

I would be so thankful for a hint as I could not figure out a solution since last thursday.
Info: MiniZinc Version 2.4.3., pymzn Version 18.3, Visual Studio Code Version 1.49.2, Python 3.8.5 x64, Windows 10.0.18363 x64,
Best regards
Dominic

---

mzn file:
`enum ITEM;
int: wcapacity;
int: vcapacity;

array[ITEM] of int: profits;
array[ITEM] of int: weights;
array[ITEM] of int: volumes;

var set of ITEM: knapsack;

var int: weight = sum (i in knapsack) (weights[i]);
constraint weight <= wcapacity;

var int: volume = sum (i in knapsack) (volumes[i]);
constraint volume <= vcapacity;

var int: profit = sum (i in knapsack) (profits[i]);
solve maximize profit ;

%regardless of using below output line or not
output [show(profit)];`

---

dzn file:
`wcapacity = 3500;
vcapacity = 20;

ITEM = { AAA, AAB, AAC, AAD, AAE};

profits = [195,150,63,183,20];

weights = [982,124,1144,1507,1009];

volumes = [5,1,8,8,6];`

Model Flattening generates no fzn and ozn file

Hi,
when I try running any model, for instance the 'knapsack' model, I get the error " 'NoneType' object has no attribute 'endswith' " which appears for the mzn_file variable in the args function being called from the _solve function.
I could backtrack the problem to the mzn2fzn function where the process is run but during that no fzn or ozn file is generated. This leads to the fzn_file and ozn_file variables being assigned the None object which then is handed over to the _solve function and causing the problem.

Has anyone ever had the same problem or knows how to deal with it?
Best regards,
Julian

FlatZinc::AST::TypeError with chuffed and output_vars

When generating a MiniZinc model with output variables, pymzn adds output annotations, such as output_array([1..52]). When using Chuffed as a solver, this leads to an error FlatZinc::AST::TypeError.

Example

Using the attached files, running chuffed fails, while mzn-gecode works:

$ mzn-chuffed pymzn_lo34fh_5.mzn pymzn_lo34fh_5.dzn
terminate called after throwing an instance of 'FlatZinc::AST::TypeError'

The error is issued by the annotations in lines 5-6 in the mzn file:

array[1..52] of var 1..52: x :: output_array([1..52]);
array[1..52] of var 1..52: y :: output_array([1..52]);

In pymzn, they are added in _redefine_output_vars in model.py.

pymzn was invoked via:
pymzn.minizinc('black-hole.mzn' , data={'layout': [[10, 11, 12], [30, 7, 4], [20, 36, 22], [37, 18, 9], [33, 19, 29], [26, 2, 51], [31, 44, 25], [50, 52, 43], [38, 48, 41], [27, 49, 42], [34, 3, 46], [5, 47, 24], [28, 21, 16], [45, 14, 15], [32, 23, 40], [39, 6, 35], [8, 17, 13]]}, all_solutions=False, solver=pymzn.chuffed, output_vars=['x', 'y'], output_mode='dict', force_flatten=True, keep=False)

force_flatten=True is not relevant for the problem, but a workaround for this issue.

I'm not entirely sure in which cases the output annotations might be necessary. For now, I've removed the call to this method locally, but that's only a workaround.

pymzn_lo34fh_5.dzn.txt
pymzn_lo34fh_5.mzn.txt
black-hole.mzn.txt

typo in minizinc.py, line 294

In minizinc.py, lien 294, there is:
args =+ ['-D', data]
I guess it should be
args += ['-D', data]

Loading pymzn throws syntax error in __init__.py

Hi,
I am using Anaconda Python 2.7.12 in Windows 8.1. I installed pymzn with pip as instructed. When I tried to import the library, it throws me this error:

Traceback (most recent call last):
  File "<pyshell#3>", line 1, in <module>
    import pymzn
  File "C:\Users\Indrajit\Anaconda2\lib\site-packages\pymzn\__init__.py", line 114
    args.func(**{**vars(args), **args.solver_args})
                  ^
SyntaxError: invalid syntax

Please look into it.

Regards,
Indrajit

Can you please cut a new release? Current pip version isn't compatible with 3.7

The subject says it all :)

Is there any way to set flattener two-pass options?

I want to pass to the compiler the option '-O5' to set a higher optimization level. I can't find a way to do it using pymzn. Am I missing something?

parallel argument leads to type error

I've run in a fairly similar bug like #35 - this time for solver argument parallel. Might probably also be an issue for seed and num_solutions as they are not converted to a string.
Solved it for the time being by setting

        if num_solutions is not None:
            args += ['-n', str(num_solutions)]
        if free_search:
            args.append('-f')
        if parallel is not None:
            args += ['-p', str(parallel)]
        if seed is not None:
            args += ['-r', str(seed)]

in the Solver class.
If I got time this weekend I could try to propose a PR.

async mod with windows 10

When trying the async mode on windows 10 i get the following error

stderr=stderr, **kwds)
File "C:\Users\tamirg\Anaconda3\lib\asyncio\base_events.py", line 1540, in subprocess_exec
bufsize, **kwargs)
File "C:\Users\tamirg\Anaconda3\lib\asyncio\base_events.py", line 462, in _make_subprocess_transport
raise NotImplementedError
NotImplementedError

my understanding is the windows 10 is using another event loop called
ProactorEventLoop

dict2dzn / arrays indexed by enums not possible?

Hey again,

In my mzn I have a type definition like this:

enum JOB;
array[JOB] of int: job_durations;

Now when I feed my data to minizinc(data={"job_durations": [1,2,3], ...) then it will result in a data file like this:

task_durations = array1d(1..3, [1, 2, 3]);

This results in an error in MiniZinc:

MiniZinc: type error: initialisation value for 'job_durations' has invalid type-inst: expected 'array[JOB] of int', actual 'array[int] of int'

Looking at the code, especially _dzn_array_nd and _index_set in marsh.py it seems that this use case is not supported.

Is there a way to make this work?

When I do things manually I do it like this:

job_durations = [ 1, 2, 3 ];
job_flags = 
[| false, false, false,  
 | false, false, false,
 | false, false, false
 |];

for 1- and 2d-arrays respectively. That is, I just omit the dimensions and arrayNd stuff.

Error "FileNotFound", Python 3.5 on Windows and macOS

Hi,
I tried to launch a simple .py file like the example proposed, but i got this error on macOS High Sierra:

Traceback (most recent call last):
 File "/Users/Username/Desktop/Test.py", line 3, in <module>
  pymzn.minizinc("test.mzn", "test.dzn")
 File "/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/site-packages/pymzn/mzn/minizinc.py", line 202, in minizinc output_mode=_output_mode)
 File "/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/site-packages/pymzn/mzn/minizinc.py", line 308, in mzn2fzn run(args)
 File "/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/site-
packages/pymzn/mzn/utils.py", line 50, in run stderr=subprocess.PIPE) as process:
 File "/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/subprocess.py",
line 676, in __init__
  restore_signals, start_new_session)
 File "/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/subprocess.py",
line 1289, in _execute_child
  raise child_exception_type(errno_num, err_msg)
FileNotFoundError: [Errno 2] No such file or directory: "mzn2fzn"

and this one on Windows 10

Traceback (most recent call last):
  File "C:/Users/Username/Desktop/pymzn/Test.py", line 3, in <module>
    pymzn.minizinc('test.mzn', 'test.dzn')
  File "C:\Program Files\Python35\lib\site-packages\pymzn\mzn\minizinc.py", line 202, in minizinc
    output_mode=_output_mode)
  File "C:\Program Files\Python35\lib\site-packages\pymzn\mzn\minizinc.py", line 308, in mzn2fzn
    run(args)
  File "C:\Program Files\Python35\lib\site-packages\pymzn\utils.py", line 50, in run
    stderr=subprocess.PIPE) as process:
  File "C:\Program Files\Python35\lib\subprocess.py", line 947, in __init__
    restore_signals, start_new_session)
  File "C:\Program Files\Python35\lib\subprocess.py", line 1224, in _execute_child
    startupinfo)
FileNotFoundError: [WinError 2] Impossibile trovare il file specificato

I've already installed MiniZinc. How can I solve this?

Temporary files not deleted

Hi,

The named temporary files that are created never seem to get deleted (delete=False). This causes the files to pile up in my /tmp dir. I've found the _cleanup function, but it doesn't seem to get invoked. One reason could be exception raising and the stream never being iterated at all. I think there should be some kind of context manager to prevent this. I'm also not sure how this should work with the wait=False mode.

set mzn_globals to "gecode" by default

Returning "INCOMPLETE" instead a "ERROR"

Hi,

When the model or the data has a problem, pymzn.minizinc() returns INCOMPLETE instead a ERROR. Let's see in the terminal:

$ cat model.mzn 
set of 0..10: values;
array[values] of var 0..1: x;
solve minimize sum(x);
values = {1, 3, 5};

$ minizinc --solver COIN-BC model.mzn 
MiniZinc: evaluation error: 
 ...../model.mzn:2:
  in variable declaration for 'x'
  array index set must be contiguous range

Which is OK calling minizinc directly. Now, let's see using pymzn:

In [1]: import pymzn                                                                                                                                    
In [2]: result = pymzn.minizinc(mzn="model.mzn", solver=pymzn.cbc)                                                                                      
In [3]: result                                                                                                                                          
Out[3]: <Solutions: INCOMPLETE>
In [4]: result.status == pymzn.Status.ERROR                                                                                                             
Out[4]: False

The documentation is a little bit vague, but it says that status INCOMPLETE represents an incomplete solution stream. I don't know what this means precisely. Assuming that we have status COMPLETE (that assumes that an optimum solution is found) and an ERROR saying that a "Generic error in the execution of the solver," I would tell that incomplete means that the search is incomplete, i.e., we have no errors (since erros are supposed to be reported by ERROR flag), but only a situation where we could not complete the search (maybe do not find a feasible solution).

However, note that models errors should be treated as ERROR since we have an actual error, not an incomplete search, or special condition in the search.

Lists containing empty sets are not translated correctly to dzn.

In pymzn 0.18.3, Python sets are correctly translated to the dzn format, even when empty, except when they are empty and in a list. For example:

pymzn.dict2dzn({'non_empty_set': set([1,2,3,8]), 
                'empty_set': set(), 
                'list_with_empty_set': [set()],
                'list_with_mixed_sets': [set([1,3]), set(), set([4,5])]})

returns

['non_empty_set = {8, 1, 2, 3};',
 'empty_set = {};',
 'list_with_empty_set = array1d(1..1, [set()]);',
 'list_with_mixed_sets = array1d(1..3, [{1, 3}, set(), {4, 5}]);']