WIP: gh-106593: PyResource API

vstinner · vstinner · commit 8af4606d89f8 · 2023-07-17T20:03:02.000+02:00
* Add PyUnicode_AsUTF8Resource()
* Add PyBytes_AsStringResource()
* Add PySequence_AsObjectArray()
* Add Include/pyresource.h
* Add PyResource_Close() to the stable ABI
* compute_abstract_methods(): Replace PySequence_Fast()
  with PySequence_AsObjectArray()
diff --git a/Doc/data/stable_abi.dat b/Doc/data/stable_abi.dat
diff --git a/Include/Python.h b/Include/Python.h
@@ -39,6 +39,7 @@
 #include "pymacro.h"
 #include "pymath.h"
 #include "pymem.h"
+#include "pyresource.h"
 #include "pytypedefs.h"
 #include "pybuffer.h"
 #include "object.h"
diff --git a/Include/abstract.h b/Include/abstract.h
@@ -706,6 +706,12 @@ PyAPI_FUNC(PyObject *) PySequence_List(PyObject *o);
    TypeError exception with 'm' as the message text. */
 PyAPI_FUNC(PyObject *) PySequence_Fast(PyObject *o, const char* m);
 
+PyAPI_FUNC(int) PySequence_AsObjectArray(
+    PyObject *,
+    PyResource *res,
+    PyObject ***parray,
+    Py_ssize_t *psize);
+
 /* Return the size of the sequence 'o', assuming that 'o' was returned by
    PySequence_Fast and is not NULL. */
 #define PySequence_Fast_GET_SIZE(o) \
diff --git a/Include/bytesobject.h b/Include/bytesobject.h
@@ -40,6 +40,7 @@ PyAPI_FUNC(PyObject *) PyBytes_FromFormat(const char*, ...)
                                 Py_GCC_ATTRIBUTE((format(printf, 1, 2)));
 PyAPI_FUNC(Py_ssize_t) PyBytes_Size(PyObject *);
 PyAPI_FUNC(char *) PyBytes_AsString(PyObject *);
+PyAPI_FUNC(char *) PyBytes_AsStringResource(PyObject *, PyResource *res);
 PyAPI_FUNC(PyObject *) PyBytes_Repr(PyObject *, int);
 PyAPI_FUNC(void) PyBytes_Concat(PyObject **, PyObject *);
 PyAPI_FUNC(void) PyBytes_ConcatAndDel(PyObject **, PyObject *);
diff --git a/Include/cpython/unicodeobject.h b/Include/cpython/unicodeobject.h
@@ -455,6 +455,10 @@ PyAPI_FUNC(PyObject*) PyUnicode_FromKindAndData(
 
 PyAPI_FUNC(const char *) PyUnicode_AsUTF8(PyObject *unicode);
 
+PyAPI_FUNC(const char *) PyUnicode_AsUTF8(PyObject *unicode);
+
+PyAPI_FUNC(const char *) PyUnicode_AsUTF8Resource(PyObject *unicode, PyResource *res);
+
 #define _PyUnicode_AsString PyUnicode_AsUTF8
 
 /* === Characters Type APIs =============================================== */
diff --git a/Include/pyresource.h b/Include/pyresource.h
@@ -0,0 +1,21 @@
+#ifndef Py_RESOURCE_H
+#define Py_RESOURCE_H
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef struct {
+    void (*close_func) (void *data);
+    void *data;
+} PyResource;
+
+PyAPI_FUNC(void) PyResource_Close(PyResource *res);
+
+#ifdef Py_BUILD_CORE
+PyAPI_FUNC(void) _PyResource_DECREF(void *data);
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+#endif  // !Py_RESOURCE_H
diff --git a/Lib/test/test_stable_abi_ctypes.py b/Lib/test/test_stable_abi_ctypes.py
diff --git a/Makefile.pre.in b/Makefile.pre.in
@@ -1657,6 +1657,7 @@ PYTHON_HEADERS= \
 		$(srcdir)/Include/pymath.h \
 		$(srcdir)/Include/pymem.h \
 		$(srcdir)/Include/pyport.h \
+		$(srcdir)/Include/pyresource.h \
 		$(srcdir)/Include/pystate.h \
 		$(srcdir)/Include/pystats.h \
 		$(srcdir)/Include/pystrcmp.h \
diff --git a/Misc/stable_abi.toml b/Misc/stable_abi.toml
@@ -2444,3 +2444,8 @@
     added = '3.13'
 [function.PyMapping_GetOptionalItemString]
     added = '3.13'
+[struct.PyResource]
+    added = '3.13'
+    struct_abi_kind = 'full-abi'
+[function.PyResource_Close]
+    added = '3.13'
diff --git a/Modules/_abc.c b/Modules/_abc.c
@@ -317,34 +317,46 @@ compute_abstract_methods(PyObject *self)
     }
     assert(PyList_Check(items));
     for (Py_ssize_t pos = 0; pos < PyList_GET_SIZE(items); pos++) {
-        PyObject *it = PySequence_Fast(
-                PyList_GET_ITEM(items, pos),
-                "items() returned non-iterable");
-        if (!it) {
+        PyObject *seq = PyList_GET_ITEM(items, pos);
+        PyResource res;
+        PyObject **items;
+        Py_ssize_t nitem;
+        if (PySequence_AsObjectArray(seq, &res, &items, &nitem) < 0) {
+            if (PyErr_ExceptionMatches(PyExc_TypeError)) {
+                PyErr_SetString(PyExc_TypeError, "items() returned non-iterable");
+            }
             goto error;
         }
-        if (PySequence_Fast_GET_SIZE(it) != 2) {
+        if (nitem != 2) {
             PyErr_SetString(PyExc_TypeError,
                             "items() returned item which size is not 2");
-            Py_DECREF(it);
+            PyResource_Close(&res);
             goto error;
         }
 
-        // borrowed
-        PyObject *key = PySequence_Fast_GET_ITEM(it, 0);
-        PyObject *value = PySequence_Fast_GET_ITEM(it, 1);
         // items or it may be cleared while accessing __abstractmethod__
         // So we need to keep strong reference for key
-        Py_INCREF(key);
+        PyObject *key = Py_NewRef(items[0]);
+        PyObject *value = Py_NewRef(items[1]);
+        PyResource_Close(&res);
+
         int is_abstract = _PyObject_IsAbstract(value);
-        if (is_abstract < 0 ||
-                (is_abstract && PySet_Add(abstracts, key) < 0)) {
-            Py_DECREF(it);
-            Py_DECREF(key);
-            goto error;
+        int err;
+        if (is_abstract < 0) {
+            err = 1;
+        }
+        else if (is_abstract) {
+            err = PySet_Add(abstracts, key) < 0;
+        }
+        else {
+            err = 0;
         }
         Py_DECREF(key);
-        Py_DECREF(it);
+        Py_DECREF(value);
+
+        if (err) {
+            goto error;
+        }
     }
 
     /* Stage 2: inherited abstract methods. */
diff --git a/Objects/abstract.c b/Objects/abstract.c
@@ -2205,6 +2205,72 @@ PySequence_Fast(PyObject *v, const char *m)
     return v;
 }
 
+static void pysequence_decref_track(void *data)
+{
+    PyObject *obj = _PyObject_CAST(data);
+    _PyObject_GC_TRACK(obj);
+    Py_DECREF(obj);
+}
+
+int
+PySequence_AsObjectArray(PyObject *v, PyResource *res,
+                         PyObject ***parray, Py_ssize_t *psize)
+{
+    if (v == NULL) {
+        null_error();
+        goto error;
+    }
+
+    if (PyTuple_CheckExact(v)) {
+        if (_PyObject_GC_IS_TRACKED(v)) {
+            // Don't track the tuple while it's being used
+            _PyObject_GC_UNTRACK(v);
+            res->close_func = pysequence_decref_track;
+        }
+        else {
+            res->close_func = _PyResource_DECREF;
+        }
+        res->data = Py_NewRef(v);
+        *parray = _PyTuple_ITEMS(v);
+        *psize = PyTuple_GET_SIZE(v);
+        return 0;
+    }
+
+    if (PyList_CheckExact(v)) {
+        // Don't track the list while it's being used
+        _PyObject_GC_UNTRACK(v);
+        res->close_func = pysequence_decref_track;
+        res->data = Py_NewRef(v);
+        *parray = _PyList_ITEMS(v);
+        *psize = PyList_GET_SIZE(v);
+        return 0;
+    }
+
+    PyObject *it = PyObject_GetIter(v);
+    if (it == NULL) {
+        goto error;
+    }
+
+    v = PySequence_List(it);
+    Py_DECREF(it);
+    if (v == NULL) {
+        goto error;
+    }
+
+    // Don't track the list while it's being used
+    _PyObject_GC_UNTRACK(v);
+    res->close_func = pysequence_decref_track;
+    res->data = v;
+    *parray = _PyList_ITEMS(v);
+    *psize = PyList_GET_SIZE(v);
+    return 0;
+
+error:
+    *parray = NULL;
+    *psize = 0;
+    return -1;
+}
+
 /* Iterate over seq.  Result depends on the operation:
    PY_ITERSEARCH_COUNT:  -1 if error, else # of times obj appears in seq.
    PY_ITERSEARCH_INDEX:  0-based index of first occurrence of obj in seq;
diff --git a/Objects/bytesobject.c b/Objects/bytesobject.c
@@ -1220,6 +1220,19 @@ PyBytes_AsString(PyObject *op)
     return ((PyBytesObject *)op)->ob_sval;
 }
 
+char *
+PyBytes_AsStringResource(PyObject *op, PyResource *res)
+{
+    if (!PyBytes_Check(op)) {
+        PyErr_Format(PyExc_TypeError,
+             "expected bytes, %.200s found", Py_TYPE(op)->tp_name);
+        return NULL;
+    }
+    res->close_func = _PyResource_DECREF;
+    res->data = Py_NewRef(op);
+    return ((PyBytesObject *)op)->ob_sval;
+}
+
 int
 PyBytes_AsStringAndSize(PyObject *obj,
                          char **s,
diff --git a/Objects/object.c b/Objects/object.c
@@ -2678,6 +2678,20 @@ int Py_IsFalse(PyObject *x)
     return Py_Is(x, Py_False);
 }
 
+void _PyResource_DECREF(void *data)
+{
+    PyObject *obj = _PyObject_CAST(data);
+    Py_DECREF(obj);
+}
+
+void PyResource_Close(PyResource *res)
+{
+    if (res->close_func == NULL) {
+        return;
+    }
+    res->close_func(res->data);
+}
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
@@ -3806,6 +3806,14 @@ PyUnicode_AsUTF8(PyObject *unicode)
     return PyUnicode_AsUTF8AndSize(unicode, NULL);
 }
 
+const char *
+PyUnicode_AsUTF8Resource(PyObject *unicode, PyResource *res)
+{
+    res->close_func = _PyResource_DECREF;
+    res->data = Py_NewRef(unicode);
+    return PyUnicode_AsUTF8AndSize(unicode, NULL);
+}
+
 /*
 PyUnicode_GetSize() has been deprecated since Python 3.3
 because it returned length of Py_UNICODE.
diff --git a/PC/python3dll.c b/PC/python3dll.c
diff --git a/PCbuild/pythoncore.vcxproj b/PCbuild/pythoncore.vcxproj
@@ -308,6 +308,7 @@
     <ClInclude Include="..\Include\pymath.h" />
     <ClInclude Include="..\Include\pymem.h" />
     <ClInclude Include="..\Include\pyport.h" />
+    <ClInclude Include="..\Include\pyresource.h" />
     <ClInclude Include="..\Include\pystate.h" />
     <ClInclude Include="..\Include\pystats.h" />
     <ClInclude Include="..\Include\pystrcmp.h" />
diff --git a/PCbuild/pythoncore.vcxproj.filters b/PCbuild/pythoncore.vcxproj.filters
@@ -171,6 +171,9 @@
     <ClInclude Include="..\Include\pyport.h">
       <Filter>Include</Filter>
     </ClInclude>
+    <ClInclude Include="..\Include\pyresource.h">
+      <Filter>Include</Filter>
+    </ClInclude>
     <ClInclude Include="..\Include\pystate.h">
       <Filter>Include</Filter>
     </ClInclude>
