added variable names with direct solver and mps. CBC and GLPK are incompatible with variables names. MindOpt untested since my license is not functioning

Author: olivierjuan

linopy/io.py

@@ -69,48 +69,45 @@ def print_coord(coord):
     return coord
 
 
-def get_printers(m: Model, io_api: str, anonymously: bool = True):
-    if anonymously:
-        if io_api == "lp":
-            def print_variable_anonymous(var):
-                return f"x{var}"
-
-            def print_constraint_anonymous(cons):
-                return f"c{cons}"
-
-        elif io_api == "lp-polars":
-            def print_variable_anonymous(series):
-                return pl.lit(" x").alias("x"), series.cast(pl.String)
-
-            def print_constraint_anonymous(series):
-                return pl.lit("c").alias("c"), series.cast(pl.String)
-        else:
-            return None
-
-        return print_variable_anonymous, print_constraint_anonymous
-    else:
-        def print_variable_lp(var):
+def get_printers(m: Model, for_polars: bool = False, with_names: bool = False):
+    if with_names:
+        def print_variable(var):
             name, coord = m.variables.get_label_position(var)
             name = clean_name(name)
             return f"{name}{print_coord(coord)}#{var}"
 
-        def print_constraint_lp(cons):
+        def print_constraint(cons):
             name, coord = m.constraints.get_label_position(cons)
             name = clean_name(name)
             return f"{name}{print_coord(coord)}#{cons}"
 
         def print_variable_polars(series):
-            return pl.lit(" "), series.map_elements(print_variable_lp, return_dtype=pl.String)
+            return pl.lit(" "), series.map_elements(print_variable, return_dtype=pl.String)
+
+        def print_constraint_polars(series):
+            return pl.lit(None), series.map_elements(print_constraint, return_dtype=pl.String)
+
+        if for_polars:
+            return print_variable_polars, print_constraint_polars
+        else:
+            return print_variable, print_constraint
+    else:
+        def print_variable(var):
+            return f"x{var}"
+
+        def print_constraint(cons):
+            return f"c{cons}"
+
+        def print_variable_polars(series):
+            return pl.lit(" x").alias("x"), series.cast(pl.String)
 
         def print_constraint_polars(series):
-            return pl.lit(None), series.map_elements(print_constraint_lp, return_dtype=pl.String)
+            return pl.lit("c").alias("c"), series.cast(pl.String)
 
-        if io_api == "lp":
-            return print_variable_lp, print_constraint_lp
-        elif io_api == "lp-polars":
+        if for_polars:
             return print_variable_polars, print_constraint_polars
         else:
-            return None
+            return print_variable, print_constraint
 
 
 def objective_write_linear_terms(
@@ -739,9 +736,7 @@ def to_file(
     if progress is None:
         progress = m._xCounter > 10_000
 
-    printers = get_printers(m, io_api, anonymously=not with_names)
-    if with_names and io_api not in ["lp", "lp-polars"]:
-        logger.warning("Debug names are only supported for LP files.")
+    printers = get_printers(m, for_polars=io_api=="lp-polars", with_names=with_names)
 
     if io_api == "lp":
         to_lp_file(m, fn, integer_label, slice_size=slice_size, progress=progress, printers=printers)
@@ -758,7 +753,7 @@ def to_file(
 
         # Use very fast highspy implementation
         # Might be replaced by custom writer, however needs C/Rust bindings for performance
-        h = m.to_highspy()
+        h = m.to_highspy(with_names=with_names)
         h.writeModel(str(fn))
     else:
         raise ValueError(
@@ -768,7 +763,7 @@ def to_file(
     return fn
 
 
-def to_mosek(m: Model, task: Any | None = None) -> Any:
+def to_mosek(m: Model, task: Any | None = None, with_names: bool = False) -> Any:
     """
     Export model to MOSEK.
 
@@ -786,6 +781,8 @@ def to_mosek(m: Model, task: Any | None = None) -> Any:
 
     import mosek
 
+    print_variable, print_constraint = get_printers(m, with_names=with_names)
+
     if task is None:
         task = mosek.Task()
 
@@ -796,9 +793,9 @@ def to_mosek(m: Model, task: Any | None = None) -> Any:
     # for j, n in enumerate(("x" + M.vlabels.astype(str).astype(object))):
     #    task.putvarname(j, n)
 
-    labels = M.vlabels.astype(str).astype(object)
+    labels = np.vectorize(print_variable)(M.vlabels).astype(object)
     task.generatevarnames(
-        np.arange(0, len(labels)), "x%0", [len(labels)], None, [0], list(labels)
+        np.arange(0, len(labels)), "%0", [len(labels)], None, [0], list(labels)
     )
 
     ## Variables
@@ -835,7 +832,7 @@ def to_mosek(m: Model, task: Any | None = None) -> Any:
     ## Constraints
 
     if len(m.constraints) > 0:
-        names = "c" + M.clabels.astype(str).astype(object)
+        names = np.vectorize(print_constraint)(M.clabels).astype(object)
         for i, n in enumerate(names):
             task.putconname(i, n)
         bkc = [
@@ -874,7 +871,7 @@ def to_mosek(m: Model, task: Any | None = None) -> Any:
     return task
 
 
-def to_gurobipy(m: Model, env: Any | None = None) -> Any:
+def to_gurobipy(m: Model, env: Any | None = None, with_names: bool = False) -> Any:
     """
     Export the model to gurobipy.
 
@@ -893,12 +890,14 @@ def to_gurobipy(m: Model, env: Any | None = None) -> Any:
     """
     import gurobipy
 
+    print_variable, print_constraint = get_printers(m, with_names=with_names)
+
     m.constraints.sanitize_missings()
     model = gurobipy.Model(env=env)
 
     M = m.matrices
 
-    names = "x" + M.vlabels.astype(str).astype(object)
+    names = np.vectorize(print_variable)(M.vlabels).astype(object)
     kwargs = {}
     if len(m.binaries.labels) + len(m.integers.labels):
         kwargs["vtype"] = M.vtypes
@@ -913,15 +912,15 @@ def to_gurobipy(m: Model, env: Any | None = None) -> Any:
         model.ModelSense = -1
 
     if len(m.constraints):
-        names = "c" + M.clabels.astype(str).astype(object)
+        names = np.vectorize(print_constraint)(M.clabels).astype(object)
         c = model.addMConstr(M.A, x, M.sense, M.b)  # type: ignore
         c.setAttr("ConstrName", list(names))  # type: ignore
 
     model.update()
     return model
 
 
-def to_highspy(m: Model) -> Highs:
+def to_highspy(m: Model, with_names: bool = False) -> Highs:
     """
     Export the model to highspy.
 
@@ -940,6 +939,8 @@ def to_highspy(m: Model) -> Highs:
     """
     import highspy
 
+    print_variable, print_constraint = get_printers(m, with_names=with_names)
+
     M = m.matrices
     h = highspy.Highs()
     h.addVars(len(M.vlabels), M.lb, M.ub)
@@ -966,9 +967,9 @@ def to_highspy(m: Model) -> Highs:
         h.addRows(num_cons, lower, upper, A.nnz, A.indptr, A.indices, A.data)
 
     lp = h.getLp()
-    lp.col_names_ = "x" + M.vlabels.astype(str).astype(object)
+    lp.col_names_ = np.vectorize(print_variable)(M.vlabels).astype(object)
     if len(M.clabels):
-        lp.row_names_ = "c" + M.clabels.astype(str).astype(object)
+        lp.row_names_ = np.vectorize(print_constraint)(M.clabels).astype(object)
     h.passModel(lp)
 
     # quadrative objective

linopy/model.py

@@ -1125,8 +1125,6 @@ def solve(
                 **solver_options,
             )
             if io_api == "direct":
-                if with_names:
-                    logger.warning("Passing variable and constraint names is only supported with lp files")
                 # no problem file written and direct model is set for solver
                 result = solver.solve_problem_from_model(
                     model=self,
@@ -1135,8 +1133,12 @@ def solve(
                     warmstart_fn=to_path(warmstart_fn),
                     basis_fn=to_path(basis_fn),
                     env=env,
+                    with_names=with_names,
                 )
             else:
+                if solver_name in ["glpk", "cbc"] and with_names:
+                    logger.warning(f"{solver_name} does not support writing names to lp files, disabling it.")
+                    with_names = False
                 problem_fn = self.to_file(
                     to_path(problem_fn),
                     io_api=io_api,

linopy/solvers.py

@@ -240,6 +240,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         """
         Abstract method to solve a linear problem from a model.
@@ -278,6 +279,7 @@ def solve_problem(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         """
         Solve a linear problem either from a model or a problem file.
@@ -297,6 +299,7 @@ def solve_problem(
                 warmstart_fn=warmstart_fn,
                 basis_fn=basis_fn,
                 env=env,
+                with_names=with_names,
             )
         elif problem_fn is not None:
             return self.solve_problem_from_file(
@@ -340,6 +343,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ):
         msg = "Direct API not implemented for CBC"
         raise NotImplementedError(msg)
@@ -497,6 +501,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         msg = "Direct API not implemented for GLPK"
         raise NotImplementedError(msg)
@@ -679,6 +684,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         """
         Solve a linear problem directly from a linopy model using the Highs solver.
@@ -700,6 +706,8 @@ def solve_problem_from_model(
             Path to the basis file.
         env : None, optional
             Environment for the solver
+        with_names : bool, optional
+            Transfer variable and constraint names to the solver (default: False)
 
         Returns
         -------
@@ -719,7 +727,7 @@ def solve_problem_from_model(
                 "Drop the solver option or use 'choose' to enable quadratic terms / integrality."
             )
 
-        h = model.to_highspy()
+        h = model.to_highspy(with_names=with_names)
 
         if log_fn is None and model is not None:
             log_fn = model.solver_dir / "highs.log"
@@ -896,6 +904,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         """
         Solve a linear problem directly from a linopy model using the Gurobi solver.
@@ -916,6 +925,8 @@ def solve_problem_from_model(
             Path to the basis file.
         env : None, optional
             Gurobi environment for the solver
+        with_names : bool, optional
+            Transfer variable and constraint names to the solver (default: False)
 
         Returns
         -------
@@ -927,7 +938,7 @@ def solve_problem_from_model(
             else:
                 env_ = env
 
-            m = model.to_gurobipy(env=env_)
+            m = model.to_gurobipy(env=env_, with_names=with_names)
 
             return self._solve(
                 m,
@@ -1128,6 +1139,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         msg = "Direct API not implemented for Cplex"
         raise NotImplementedError(msg)
@@ -1270,6 +1282,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         msg = "Direct API not implemented for SCIP"
         raise NotImplementedError(msg)
@@ -1411,6 +1424,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         msg = "Direct API not implemented for Xpress"
         raise NotImplementedError(msg)
@@ -1556,6 +1570,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         """
         Solve a linear problem directly from a linopy model using the MOSEK solver.
@@ -1573,7 +1588,9 @@ def solve_problem_from_model(
         basis_fn : Path, optional
             Path to the basis file.
         env : None, optional
-            Gurobi environment for the solver
+            Mosek environment for the solver
+        with_names : bool, optional
+            Transfer variable and constraint names to the solver (default: False)
 
         Returns
         -------
@@ -1584,7 +1601,7 @@ def solve_problem_from_model(
                 env_ = stack.enter_context(mosek.Env())
 
             with env_.Task() as m:
-                m = model.to_mosek(m)
+                m = model.to_mosek(m, with_names=with_names)
 
                 return self._solve(
                     m,
@@ -1878,6 +1895,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         msg = "Direct API not implemented for COPT"
         raise NotImplementedError(msg)
@@ -2020,6 +2038,7 @@ def solve_problem_from_model(
         warmstart_fn: Path | None = None,
         basis_fn: Path | None = None,
         env: None = None,
+        with_names: bool = False,
     ) -> Result:
         msg = "Direct API not implemented for MindOpt"
         raise NotImplementedError(msg)