Faster fmpz/int conversion using from_bytes/to_bytes

src/flint/types/fmpz.pxd

@@ -1,20 +1,44 @@
 from cpython.long cimport PyLong_Check
 from flint.flint_base.flint_base cimport flint_scalar
 from flint.utils.conversion cimport chars_from_str
-from flint.flintlib.types.flint cimport slong, pylong_as_slong
+from flint.flintlib.types.flint cimport slong, ulong, pylong_as_slong
 from flint.flintlib.types.flint cimport PyObject
-from flint.flintlib.functions.fmpz cimport fmpz_t, fmpz_set_str, fmpz_set_si
+from flint.flintlib.functions.fmpz cimport fmpz_t, fmpz_set_si, fmpz_set_signed_ui_array
+import sys
 
 cdef int fmpz_set_any_ref(fmpz_t x, obj)
 cdef fmpz_get_intlong(fmpz_t x)
 
+cdef int is_big_endian = int(sys.byteorder == "big")
+
+cdef inline ulong ulong_from_little_endian(unsigned char *ptr):
+    # Read a ulong from little-endian bytes
+    cdef ulong w = 0
+    for i in range(sizeof(ulong) // 8):
+        w = (w << 8) | ptr[i]
+    return w
+
 cdef inline int fmpz_set_pylong(fmpz_t x, obj):
     cdef int overflow
     cdef slong longval
+    cdef slong size
+    cdef bytes b
+    cdef ulong w
+    cdef ulong *words
+    cdef int i
+
     longval = pylong_as_slong(<PyObject*>obj, &overflow)
     if overflow:
-        s = "%x" % obj
-        fmpz_set_str(x, chars_from_str(s), 16)
+        # make sure the sign bit fits
+        # we need 8 * sizeof(ulong) * size > obj.bit_length()
+        size = obj.bit_length() // (8 * sizeof(ulong)) + 1
+        b = obj.to_bytes(sizeof(ulong) * size, "little", signed=True)
+        # b is a local Python object, we access the internal pointer
+        words = <ulong*>(<char *>b)
+        if is_big_endian:
+            for i in range(size):
+                words[i] = ulong_from_little_endian(<unsigned char *>(words + i))
+        fmpz_set_signed_ui_array(x, words, size)
     else:
         fmpz_set_si(x, longval)
 

src/flint/types/fmpz.pyx

@@ -2,7 +2,7 @@ from flint.flint_base.flint_base cimport flint_scalar
 from flint.utils.typecheck cimport typecheck
 from flint.utils.conversion cimport chars_from_str
 from flint.utils.conversion cimport str_from_chars, _str_trunc
-cimport libc.stdlib
+from libc.stdlib cimport malloc, free
 
 from flint.flintlib.types.flint cimport FMPZ_REF, FMPZ_TMP, FMPZ_UNKNOWN, COEFF_IS_MPZ
 from flint.flintlib.functions.flint cimport flint_free
@@ -14,16 +14,33 @@ from flint.flintlib.functions.partitions cimport *
 
 from flint.utils.flint_exceptions import DomainError
 
-
 cdef fmpz_get_intlong(fmpz_t x):
     """
     Convert fmpz_t to a Python int or long.
     """
-    cdef char * s
+    cdef slong size
+    cdef ulong * words
+    cdef int i
+    cdef fmpz_struct xabs[1]
     if COEFF_IS_MPZ(x[0]):
-        s = fmpz_get_str(NULL, 16, x)
-        v = int(str_from_chars(s), 16)
-        flint_free(s)
+        # Python from signed bytes is slow so we convert the absolute value.
+        # we need 8 * sizeof(ulong) * size >= fmpz_bits(x)
+        size = fmpz_bits(x) // (8 * sizeof(ulong)) + 1
+        words = <ulong *>malloc(size * sizeof(ulong))
+        if fmpz_sgn(x) == -1:
+            fmpz_init(xabs)
+            fmpz_abs(xabs, x)
+            fmpz_get_ui_array(words, size, xabs)
+            fmpz_clear(xabs)
+        else:
+            fmpz_get_ui_array(words, size, x)
+        if is_big_endian:
+            for i in range(size):
+                words[i] = ulong_from_little_endian(<unsigned char *>(words + i))
+        v = int.from_bytes((<char *>words)[:size * sizeof(ulong)], "little")
+        if fmpz_sgn(x) == -1:
+            v = -v
+        free(words)
         return v
     else:
         return <slong>x[0]