Fix UP031: Use format specifiers instead of percent format

pytensor/bin/pytensor_cache.py

@@ -75,9 +75,9 @@ def main():
             if items:
                 _logger.warning(
                     "There remain elements in the cache dir that you may "
-                    "need to erase manually. The cache dir is:\n  %s\n"
+                    f"need to erase manually. The cache dir is:\n  {config.compiledir}\n"
                     'You can also call "pytensor-cache purge" to '
-                    "remove everything from that directory." % config.compiledir
+                    "remove everything from that directory."
                 )
                 _logger.debug(f"Remaining elements ({len(items)}): {', '.join(items)}")
         elif sys.argv[1] == "list":

pytensor/breakpoint.py

@@ -105,7 +105,7 @@ class PdbBreakpoint(Op):
             except Exception:
                 raise ValueError(
                     "Some of the inputs to the PdbBreakpoint op "
-                    "'%s' could not be casted to NumPy arrays" % self.name
+                    f"'{self.name}' could not be casted to NumPy arrays"
                 )
 
             print("\n")

pytensor/compile/builders.py

@@ -709,7 +709,7 @@ class OpFromGraph(Op, HasInnerGraph):
             if not isinstance(roverrides_l, list):
                 raise TypeError(
                     "Rop overriding function should return a list, "
-                    'got "%s"' % type(roverrides_l)
+                    f'got "{type(roverrides_l)}"'
                 )
             all_rops_l, all_rops_ov_l = zip(
                 *[

pytensor/compile/ops.py

@@ -252,7 +252,7 @@ class FromFunctionOp(Op):
         return hash(type(self)) ^ hash(self.__fn)
 
     def __str__(self):
-        return "FromFunctionOp{%s}" % self.__fn.__name__
+        return f"FromFunctionOp{{{self.__fn.__name__}}}"
 
     def perform(self, node, inputs, outputs):
         outs = self.__fn(*inputs)

pytensor/link/c/params_type.py

@@ -262,8 +262,13 @@ class Params(dict):
         self.__dict__.update(__params_type__=params_type, __signatures__=None)
 
     def __repr__(self):
-        return "Params(%s)" % ", ".join(
-            [(f"{k}:{type(self[k]).__name__}:{self[k]}") for k in sorted(self.keys())]
+        return "Params({})".format(
+            ", ".join(
+                [
+                    (f"{k}:{type(self[k]).__name__}:{self[k]}")
+                    for k in sorted(self.keys())
+                ]
+            )
         )
 
     def __getattr__(self, key):
@@ -346,13 +351,11 @@ class ParamsType(CType):
         for attribute_name in kwargs:
             if re.match("^[A-Za-z_][A-Za-z0-9_]*$", attribute_name) is None:
                 raise AttributeError(
-                    'ParamsType: attribute "%s" should be a valid identifier.'
-                    % attribute_name
+                    f'ParamsType: attribute "{attribute_name}" should be a valid identifier.'
                 )
             if attribute_name in c_cpp_keywords:
                 raise SyntaxError(
-                    'ParamsType: "%s" is a potential C/C++ keyword and should not be used as attribute name.'
-                    % attribute_name
+                    f'ParamsType: "{attribute_name}" is a potential C/C++ keyword and should not be used as attribute name.'
                 )
             type_instance = kwargs[attribute_name]
             type_name = type_instance.__class__.__name__
@@ -424,8 +427,10 @@ class ParamsType(CType):
         return super().__getattr__(self, key)
 
     def __repr__(self):
-        return "ParamsType<%s>" % ", ".join(
-            [(f"{self.fields[i]}:{self.types[i]}") for i in range(self.length)]
+        return "ParamsType<{}>".format(
+            ", ".join(
+                [(f"{self.fields[i]}:{self.types[i]}") for i in range(self.length)]
+            )
         )
 
     def __eq__(self, other):
@@ -733,18 +738,18 @@ class ParamsType(CType):
         struct_cleanup = "\n".join(c_cleanup_list)
         struct_extract = "\n\n".join(c_extract_list)
         struct_extract_method = """
-        void extract(PyObject* object, int field_pos) {
-            switch(field_pos) {
+        void extract(PyObject* object, int field_pos) {{
+            switch(field_pos) {{
                 // Extraction cases.
-                %s
+                {}
                 // Default case.
                 default:
-                    PyErr_Format(PyExc_TypeError, "ParamsType: no extraction defined for a field %%d.", field_pos);
+                    PyErr_Format(PyExc_TypeError, "ParamsType: no extraction defined for a field %d.", field_pos);
                     this->setErrorOccurred();
                     break;
-            }
-        }
-        """ % (
+            }}
+        }}
+        """.format(
             "\n".join(
                 [
                     ("case %d: extract_%s(object); break;" % (i, self.fields[i]))
@@ -866,7 +871,7 @@ class ParamsType(CType):
                 struct_name=self.name,
                 length=self.length,
                 fail=sub["fail"],
-                fields_list='"%s"' % '", "'.join(self.fields),
+                fields_list='"{}"'.format('", "'.join(self.fields)),
             )
         )
 

pytensor/link/utils.py

@@ -355,8 +355,13 @@ def raise_with_op(
             + f"\nInputs values: {scalar_values}"
         )
         if verbosity == "high":
-            detailed_err_msg += "\nInputs type_num: %s" % str(
-                [getattr(getattr(i[0], "dtype", ""), "num", "") for i in thunk.inputs]
+            detailed_err_msg += "\nInputs type_num: {}".format(
+                str(
+                    [
+                        getattr(getattr(i[0], "dtype", ""), "num", "")
+                        for i in thunk.inputs
+                    ]
+                )
             )
 
         detailed_err_msg += f"\nOutputs clients: {clients}\n"

pytensor/scalar/basic.py

@@ -475,7 +475,7 @@ class ScalarType(CType, HasDataType, HasShape):
                     sub,
                     name=name,
                     dtype=specs[1],
-                    pyarr_type="Py%sArrType_Type" % specs[2],
+                    pyarr_type=f"Py{specs[2]}ArrType_Type",
                 )
             )
         else:

pytensor/tensor/blas.py

@@ -180,9 +180,9 @@ class Gemv(Op):
 
     def __str__(self):
         if self.inplace:
-            return "%s{inplace}" % self.__class__.__name__
+            return f"{self.__class__.__name__}{{inplace}}"
         else:
-            return "%s{no_inplace}" % self.__class__.__name__
+            return f"{self.__class__.__name__}{{no_inplace}}"
 
     def make_node(self, y, alpha, A, x, beta):
         y = ptb.as_tensor_variable(y)
@@ -279,9 +279,9 @@ class Ger(Op):
 
     def __str__(self):
         if self.destructive:
-            return "%s{destructive}" % self.__class__.__name__
+            return f"{self.__class__.__name__}{{destructive}}"
         else:
-            return "%s{non-destructive}" % self.__class__.__name__
+            return f"{self.__class__.__name__}{{non-destructive}}"
 
     def make_node(self, A, alpha, x, y):
         A = ptb.as_tensor_variable(A)
@@ -1811,9 +1811,10 @@ class BatchedDot(COp):
                         f"{strides}[{i}] > 0 && {strides}[{i}] % type_size == 0"
                         for i in range(1, ndim)
                     ),
-                    "(%s)"
-                    % " || ".join(
-                        f"{strides}[{i}] == type_size" for i in range(1, ndim)
+                    "({})".format(
+                        " || ".join(
+                            f"{strides}[{i}] == type_size" for i in range(1, ndim)
+                        )
                     ),
                 ]
             )

pytensor/tensor/elemwise.py

@@ -1098,14 +1098,14 @@ class Elemwise(OpenMPOp):
                 all_broadcastable = all(s == 1 for s in var.type.shape)
                 cond1 = " && ".join(
                     [
-                        "PyArray_ISCONTIGUOUS(%s)" % arr
+                        f"PyArray_ISCONTIGUOUS({arr})"
                         for arr, var in z
                         if not all_broadcastable
                     ]
                 )
                 cond2 = " && ".join(
                     [
-                        "PyArray_ISFORTRAN(%s)" % arr
+                        f"PyArray_ISFORTRAN({arr})"
                         for arr, var in z
                         if not all_broadcastable
                     ]

pytensor/tensor/extra_ops.py

@@ -652,8 +652,8 @@ class Repeat(Op):
         if repeats.dtype in numpy_unsupported_dtypes:
             raise TypeError(
                 (
-                    "dtypes %s are not supported by numpy.repeat "
-                    "for the 'repeats' parameter, " % str(numpy_unsupported_dtypes)
+                    f"dtypes {numpy_unsupported_dtypes!s} are not supported by numpy.repeat "
+                    "for the 'repeats' parameter, "
                 ),
                 repeats.dtype,
             )
@@ -882,8 +882,8 @@ class FillDiagonal(Op):
         val = ptb.as_tensor_variable(val)
         if a.ndim < 2:
             raise TypeError(
-                "%s: first parameter must have at least"
-                " two dimensions" % self.__class__.__name__
+                f"{self.__class__.__name__}: first parameter must have at least"
+                " two dimensions"
             )
         elif val.ndim != 0:
             raise TypeError(
@@ -892,8 +892,8 @@ class FillDiagonal(Op):
         val = ptb.cast(val, dtype=upcast(a.dtype, val.dtype))
         if val.dtype != a.dtype:
             raise TypeError(
-                "%s: type of second parameter must be the same as"
-                " the first's" % self.__class__.__name__
+                f"{self.__class__.__name__}: type of second parameter must be the same as"
+                " the first's"
             )
         return Apply(self, [a, val], [a.type()])
 
@@ -926,8 +926,8 @@ class FillDiagonal(Op):
             return [None, None]
         elif a.ndim > 2:
             raise NotImplementedError(
-                "%s: gradient is currently implemented"
-                " for matrices only" % self.__class__.__name__
+                f"{self.__class__.__name__}: gradient is currently implemented"
+                " for matrices only"
             )
         wr_a = fill_diagonal(grad, 0)  # valid for any number of dimensions
         # diag is only valid for matrices
@@ -984,8 +984,8 @@ class FillDiagonalOffset(Op):
         offset = ptb.as_tensor_variable(offset)
         if a.ndim != 2:
             raise TypeError(
-                "%s: first parameter must have exactly"
-                " two dimensions" % self.__class__.__name__
+                f"{self.__class__.__name__}: first parameter must have exactly"
+                " two dimensions"
             )
         elif val.ndim != 0:
             raise TypeError(
@@ -998,8 +998,8 @@ class FillDiagonalOffset(Op):
         val = ptb.cast(val, dtype=upcast(a.dtype, val.dtype))
         if val.dtype != a.dtype:
             raise TypeError(
-                "%s: type of second parameter must be the same"
-                " as the first's" % self.__class__.__name__
+                f"{self.__class__.__name__}: type of second parameter must be the same"
+                " as the first's"
             )
         elif offset.dtype not in integer_dtypes:
             raise TypeError(

pytensor/tensor/fft.py

@@ -20,8 +20,7 @@ class RFFTOp(Op):
         a = as_tensor_variable(a)
         if a.ndim < 2:
             raise TypeError(
-                "%s: input must have dimension > 2, with first dimension batches"
-                % self.__class__.__name__
+                f"{self.__class__.__name__}: input must have dimension > 2, with first dimension batches"
             )
 
         if s is None:
@@ -31,8 +30,8 @@ class RFFTOp(Op):
             s = as_tensor_variable(s)
             if s.dtype not in integer_dtypes:
                 raise TypeError(
-                    "%s: length of the transformed axis must be"
-                    " of type integer" % self.__class__.__name__
+                    f"{self.__class__.__name__}: length of the transformed axis must be"
+                    " of type integer"
                 )
         return Apply(self, [a, s], [self.output_type(a)()])
 
@@ -92,8 +91,8 @@ class IRFFTOp(Op):
             s = as_tensor_variable(s)
             if s.dtype not in integer_dtypes:
                 raise TypeError(
-                    "%s: length of the transformed axis must be"
-                    " of type integer" % self.__class__.__name__
+                    f"{self.__class__.__name__}: length of the transformed axis must be"
+                    " of type integer"
                 )
         return Apply(self, [a, s], [self.output_type(a)()])
 

pytensor/tensor/io.py

@@ -28,7 +28,7 @@ class LoadFromDisk(Op):
         if mmap_mode not in (None, "c"):
             raise ValueError(
                 "The only supported values for mmap_mode "
-                "are None and 'c', got %s" % mmap_mode
+                f"are None and 'c', got {mmap_mode}"
             )
         self.mmap_mode = mmap_mode
 

pytensor/tensor/math.py

@@ -1540,7 +1540,7 @@ class Mean(FixedOpCAReduce):
 
     def __str__(self):
         if self.axis is not None:
-            return "Mean{%s}" % (", ".join(str(x) for x in self.axis))
+            return "Mean{{{}}}".format(", ".join(str(x) for x in self.axis))
         else:
             return "Mean"
 

pytensor/tensor/subtensor.py

@@ -2174,7 +2174,7 @@ class AdvancedIncSubtensor1(COp):
         else:
             msg += ",inc"
 
-        return self.__class__.__name__ + "{%s}" % msg
+        return self.__class__.__name__ + f"{{{msg}}}"
 
     def make_node(self, x, y, ilist):
         x_ = as_tensor_variable(x)