Merge pull request #409 from oscarbenjamin/pr_bump_flint36

Bump to FLINT 3.6

Author: oscarbenjamin

.github/workflows/buildwheel.yml

@@ -302,6 +302,7 @@ jobs:
           'v3.2.2',
           'v3.3.1',
           'v3.4.0',
+          'v3.5.0',
         ]
     steps:
       - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0

README.md

@@ -144,7 +144,7 @@ Compatible versions:
 
 | python-flint | Release date  | CPython     | FLINT      | Cython           |
 |--------------|---------------|-------------|------------|------------------|
-| `0.9.0`      |      ???      | `3.11-3.14` | `3.0-3.5`  | `3.1-3.2?`       |
+| `0.9.0`      |      ???      | `3.11-3.14` | `3.0-3.6`  | `3.1-3.2?`       |
 | `0.8.0`      | 29th Aug 2025 | `3.11-3.14` | `3.0-3.3`  | `3.1` only       |
 | `0.7.0`      | 16th Mar 2025 | `3.11-3.13` | `3.0-3.2`  | `3.0.11-3.1.0a1` |
 | `0.6.0`      |  1st Feb 2024 | `3.9-3.12`  | `3.0` only | `3.0` only       |
@@ -197,10 +197,41 @@ Changes (0.9.0):
   work. (OB)
 - [gh-324](https://github.com/flintlib/python-flint/pull/324),
   Faster conversion from `int` to `fmpz` and back. (RO).
+- [gh-326](https://github.com/flintlib/python-flint/pull/326),
+  Add support for building with the Python stable ABI. (OB)
+- [gh-341](https://github.com/flintlib/python-flint/pull/341),
+  Disable hashing for inexact `arb` values. (DSM)
+- [gh-347](https://github.com/flintlib/python-flint/pull/347),
+  Fix storing the variable name for `fq_default`. (OB)
+- [gh-350](https://github.com/flintlib/python-flint/pull/350),
+  Add support for Cython 3.2. (OB)
+- [gh-352](https://github.com/flintlib/python-flint/pull/352),
+  Fix a crash in `fmpz_mat.lll()` for empty matrices. (OB)
 - [gh-359](https://github.com/flintlib/python-flint/pull/359),
   Sort factorisations of all mpoly types. (OB)
-- [gh-374](https://github.com/flintlib/python-flint/pull/374), Fixed a bug in `nmod.__hash__`. (FH)
-- [gh-392](https://github.com/flintlib/python-flint/pull/392), Add interface to `acb_theta_jet` (JH, SS)
+- [gh-367](https://github.com/flintlib/python-flint/pull/367),
+  Add type stubs and tests for many more classes. (OB)
+- [gh-370](https://github.com/flintlib/python-flint/pull/370),
+  Fix `unused_gens()` for the zero polynomial. (OB)
+- [gh-374](https://github.com/flintlib/python-flint/pull/374),
+  Fixed a bug in `nmod.__hash__`. (FH)
+- [gh-378](https://github.com/flintlib/python-flint/pull/378),
+  Build Windows extension modules with MSVC. (OB)
+- [gh-379](https://github.com/flintlib/python-flint/pull/379),
+  Add Windows ARM64 wheels. (OB)
+- [gh-380](https://github.com/flintlib/python-flint/pull/380),
+  Support building against FLINT 3.4.0. (OB)
+- [gh-393](https://github.com/flintlib/python-flint/pull/393),
+  Use `fmpz_mod_mat_det()` when building against FLINT 3.1 or newer. (OB)
+- [gh-401](https://github.com/flintlib/python-flint/pull/401),
+  Upload Pyodide wheels to PyPI. (OB)
+- [gh-389](https://github.com/flintlib/python-flint/pull/389),
+  [gh-402](https://github.com/flintlib/python-flint/pull/402),
+  Support building against FLINT 3.5.0. (OB)
+- [gh-392](https://github.com/flintlib/python-flint/pull/392),
+  Add interface to `acb_theta_jet` (JH, SS)
+- [gh-409](https://github.com/flintlib/python-flint/pull/409),
+  Support building against FLINT 3.6.0. (OB)
 
 0.8.0
 -----

bin/build_variables.sh

@@ -19,4 +19,4 @@ MPIRVER=3.0.0 # MPIR build no longer works (not clear where to download from)
 # These are the actual dependencies used (at least by default):
 GMPVER=6.3.0
 MPFRVER=4.2.2
-FLINTVER=3.5.0
+FLINTVER=3.6.0

bin/patch-flint-windows-arm64-link-3.6.0.diff

@@ -0,0 +1,32 @@
+diff --git a/Makefile.in b/Makefile.in
+index 8185fa609..a3822f53b 100644
+--- a/Makefile.in
++++ b/Makefile.in
+@@ -424,22 +424,16 @@ endif
+ 
+ ifneq ($(SHARED), 0)
+ shared: $(FLINT_DIR)/$(FLINT_LIB_FULL)
+-# The following is to avoid reaching the maximum length of command line
+-# arguments, mainly present on MinGW.
+-define xxx_merged_lobj_rule
+-$(BUILD_DIR)/$(1)_merged.lo: $($(1)_LOBJS) | $(BUILD_DIR)
+-	@$(LD) -r $($(1)_LOBJS) -o $(BUILD_DIR)/$(1)_merged.lo
+-endef
+-$(foreach dir, $(DIRS), $(eval $(call xxx_merged_lobj_rule,$(dir))))
+-MERGED_LOBJS:=$(foreach dir, $(DIRS),$(BUILD_DIR)/$(dir)_merged.lo)
+ ifeq ($(WANT_LTO),1)
+ SHARED_LIB_DEPS:=$(LOBJS)
+ else
+-SHARED_LIB_DEPS:=$(MERGED_LOBJS)
++SHARED_LINK_RSP := $(BUILD_DIR)/libflint-shared.rsp
++SHARED_LIB_DEPS:=$(LOBJS)
+ endif
+-$(FLINT_DIR)/$(FLINT_LIB_FULL): $(SHARED_LIB_DEPS)
++$(FLINT_DIR)/$(FLINT_LIB_FULL): $(SHARED_LIB_DEPS) | $(BUILD_DIR)
+ 	@echo "Building $(FLINT_LIB_FULL)"
+-	$(CMD) $(CC) $(CFLAGS) -shared $(EXTRA_SHARED_FLAGS) $(SHARED_LIB_DEPS) -o $(FLINT_LIB_FULL) $(LDFLAGS) $(LIBS)
++	@: $(file >$(SHARED_LINK_RSP))$(foreach obj,$(LOBJS),$(file >>$(SHARED_LINK_RSP),$(obj)))
++	$(CMD) $(CC) $(CFLAGS) -shared $(EXTRA_SHARED_FLAGS) @$(SHARED_LINK_RSP) -o $(FLINT_LIB_FULL) $(LDFLAGS) $(LIBS)
+ 	@$(RM_F) $(FLINT_LIB)
+ 	@$(RM_F) $(FLINT_LIB_MAJOR)
+ 	@$(LN_S) $(FLINT_LIB_FULL) $(FLINT_LIB)

doc/source/build.rst

@@ -87,7 +87,7 @@ Compatible versions:
    * - 0.9.0
      - Not yet
      - 3.11-3.14
-     - 3.0-3.5
+     - 3.0-3.6
      - 3.1-3.2?
    * - 0.8.0
      - 29th Aug 2025

meson.build

@@ -16,7 +16,7 @@ project(
 # then we can consider not using a speculative upper version cap here.
 #
 flint_lower = '>=3.0'
-flint_upper = '<3.6'
+flint_upper = '<3.7'
 cython_lower = '>=3.0.11'
 cython_upper = '<3.3'
 

src/flint/flintlib/functions/compat.pxd

@@ -1,6 +1,6 @@
 from flint.flintlib.types.flint cimport slong
 from flint.flintlib.types.fmpz_mod cimport fmpz_mod_mpoly_ctx_t, fmpz_mod_mpoly_t
-from flint.flintlib.types.gr cimport gr_ctx_t
+from flint.flintlib.types.gr cimport gr_ctx_t, gr_ptr, gr_srcptr, gr_vec_t
 
 
 cdef extern from *:
@@ -30,8 +30,19 @@ cdef extern from *:
 
     #define compat_fmpz_mod_mpoly_compose_fmpz_mod_mpoly_gen(...) fmpz_mod_mpoly_compose_fmpz_mod_mpoly_gen(__VA_ARGS__)
 
+    #endif
+
+    #if __FLINT_RELEASE >= 30600 /* Flint 3.6.0 or later */
+
+    #define compat_gr_factor(c, factors, exponents, x, flags, ctx) gr_factor(c, factors, (fmpz_vec_struct *) exponents, x, flags, ctx)
+
+    #else
+
+    #define compat_gr_factor(c, factors, exponents, x, flags, ctx) gr_factor(c, factors, exponents, x, flags, ctx)
+
     #endif
     """
     void compat_gr_ctx_init_gr_series(gr_ctx_t ctx, gr_ctx_t base_ring, slong prec)
     void compat_gr_ctx_init_series_mod_gr_poly(gr_ctx_t ctx, gr_ctx_t base_ring, slong n)
     void compat_fmpz_mod_mpoly_compose_fmpz_mod_mpoly_gen(fmpz_mod_mpoly_t A, const fmpz_mod_mpoly_t B, const slong * c, const fmpz_mod_mpoly_ctx_t ctxB, const fmpz_mod_mpoly_ctx_t ctxAC)
+    int compat_gr_factor(gr_ptr c, gr_vec_t factors, gr_vec_t exponents, gr_srcptr x, int flags, gr_ctx_t ctx)

src/flint/types/_gr.pxd

@@ -33,6 +33,7 @@ from flint.flintlib.functions.fmpq_poly cimport (
     fmpq_poly_set,
 )
 from flint.flintlib.functions.compat cimport (
+    compat_gr_factor,
     compat_gr_ctx_init_gr_series,
     # compat_gr_ctx_init_series_mod_gr_poly,
 )
@@ -159,7 +160,6 @@ from flint.flintlib.functions.gr cimport (
 
     gr_gcd,
     gr_lcm,
-    gr_factor,
 
     gr_floor,
     gr_ceil,
@@ -782,7 +782,7 @@ cdef class gr_ctx(flint_ctx):
         c = self.new_gr()
         gr_vec_init(factors, 0, self.ctx_t)
         gr_vec_init(exponents, 0, gr_fmpz_ctx_c.ctx_t)
-        err = gr_factor(c.pval, factors, exponents, x.pval, flags, self.ctx_t)
+        err = compat_gr_factor(c.pval, factors, exponents, x.pval, flags, self.ctx_t)
         if err != GR_SUCCESS:
             raise self._error(err, "Failed to factor gr object")
         length = gr_vec_length(factors, self.ctx_t)
@@ -1716,7 +1716,7 @@ cdef class gr(flint_scalar):
         c = self.ctx.new_gr()
         gr_vec_init(factors, 0, self.ctx.ctx_t)
         gr_vec_init(exponents, 0, gr_fmpz_ctx_c.ctx_t)
-        err = gr_factor(c.pval, factors, exponents, self.pval, flags, self.ctx.ctx_t)
+        err = compat_gr_factor(c.pval, factors, exponents, self.pval, flags, self.ctx.ctx_t)
         if err != GR_SUCCESS:
             raise self._error(err, "Failed to factor gr object")
         length = gr_vec_length(factors, self.ctx.ctx_t)