Merge pull request hpyproject#411 from hpyproject/fa/unicode_fromecod…

…ed_substring

Introduce HPyUnicode_FromEncodedObjec, HPyUnicode_Substring, HPyDict_Keys, and HPyDict_Copy.

docs/api-reference/function-index.rst

@@ -26,6 +26,8 @@ HPy Core API Function Index
 * :c:func:`HPyContextVar_New`
 * :c:func:`HPyContextVar_Set`
 * :c:func:`HPyDict_Check`
+* :c:func:`HPyDict_Copy`
+* :c:func:`HPyDict_Keys`
 * :c:func:`HPyDict_New`
 * :c:func:`HPyErr_Clear`
 * :c:func:`HPyErr_ExceptionMatches`
@@ -94,9 +96,11 @@ HPy Core API Function Index
 * :c:func:`HPyUnicode_DecodeFSDefaultAndSize`
 * :c:func:`HPyUnicode_DecodeLatin1`
 * :c:func:`HPyUnicode_EncodeFSDefault`
+* :c:func:`HPyUnicode_FromEncodedObject`
 * :c:func:`HPyUnicode_FromString`
 * :c:func:`HPyUnicode_FromWideChar`
 * :c:func:`HPyUnicode_ReadChar`
+* :c:func:`HPyUnicode_Substring`
 * :c:func:`HPy_ASCII`
 * :c:func:`HPy_Absolute`
 * :c:func:`HPy_Add`

docs/porting-guide.rst

@@ -170,6 +170,8 @@ with the code for the :term:`CPython ABI` mode, so it is guaranteed to be correc
     `PyContextVar_New <https://docs.python.org/3/c-api/contextvars.html#c.PyContextVar_New>`_                                          :c:func:`HPyContextVar_New`
     `PyContextVar_Set <https://docs.python.org/3/c-api/contextvars.html#c.PyContextVar_Set>`_                                          :c:func:`HPyContextVar_Set`
     `PyDict_Check <https://docs.python.org/3/c-api/dict.html#c.PyDict_Check>`_                                                         :c:func:`HPyDict_Check`
+    `PyDict_Copy <https://docs.python.org/3/c-api/dict.html#c.PyDict_Copy>`_                                                           :c:func:`HPyDict_Copy`
+    `PyDict_Keys <https://docs.python.org/3/c-api/dict.html#c.PyDict_Keys>`_                                                           :c:func:`HPyDict_Keys`
     `PyDict_New <https://docs.python.org/3/c-api/dict.html#c.PyDict_New>`_                                                             :c:func:`HPyDict_New`
     `PyErr_Clear <https://docs.python.org/3/c-api/exceptions.html#c.PyErr_Clear>`_                                                     :c:func:`HPyErr_Clear`
     `PyErr_ExceptionMatches <https://docs.python.org/3/c-api/exceptions.html#c.PyErr_ExceptionMatches>`_                               :c:func:`HPyErr_ExceptionMatches`
@@ -269,9 +271,11 @@ with the code for the :term:`CPython ABI` mode, so it is guaranteed to be correc
     `PyUnicode_DecodeFSDefaultAndSize <https://docs.python.org/3/c-api/unicode.html#c.PyUnicode_DecodeFSDefaultAndSize>`_              :c:func:`HPyUnicode_DecodeFSDefaultAndSize`
     `PyUnicode_DecodeLatin1 <https://docs.python.org/3/c-api/unicode.html#c.PyUnicode_DecodeLatin1>`_                                  :c:func:`HPyUnicode_DecodeLatin1`
     `PyUnicode_EncodeFSDefault <https://docs.python.org/3/c-api/unicode.html#c.PyUnicode_EncodeFSDefault>`_                            :c:func:`HPyUnicode_EncodeFSDefault`
+    `PyUnicode_FromEncodedObject <https://docs.python.org/3/c-api/unicode.html#c.PyUnicode_FromEncodedObject>`_                        :c:func:`HPyUnicode_FromEncodedObject`
     `PyUnicode_FromString <https://docs.python.org/3/c-api/unicode.html#c.PyUnicode_FromString>`_                                      :c:func:`HPyUnicode_FromString`
     `PyUnicode_FromWideChar <https://docs.python.org/3/c-api/unicode.html#c.PyUnicode_FromWideChar>`_                                  :c:func:`HPyUnicode_FromWideChar`
     `PyUnicode_ReadChar <https://docs.python.org/3/c-api/unicode.html#c.PyUnicode_ReadChar>`_                                          :c:func:`HPyUnicode_ReadChar`
+    `PyUnicode_Substring <https://docs.python.org/3/c-api/unicode.html#c.PyUnicode_Substring>`_                                        :c:func:`HPyUnicode_Substring`
     ================================================================================================================================== ================================================
 
 

hpy/debug/src/debug_ctx.c

@@ -589,3 +589,24 @@ int debug_ctx_Type_IsSubtype(HPyContext *dctx, DHPy sub, DHPy type)
     ctx_info->is_valid = true;
     return res;
 }
+
+DHPy debug_ctx_Unicode_Substring(HPyContext *dctx, DHPy str, HPy_ssize_t start, HPy_ssize_t end)
+{
+    HPyDebugCtxInfo *ctx_info;
+    HPyContext *uctx;
+
+    ctx_info = get_ctx_info(dctx);
+    if (!ctx_info->is_valid) {
+        report_invalid_debug_context();
+    }
+
+    HPy uh_str = DHPy_unwrap(dctx, str);
+    uctx = ctx_info->info->uctx;
+    if (!HPy_TypeCheck(uctx, uh_str, uctx->h_UnicodeType)) {
+        HPy_FatalError(uctx, "HPyUnicode_Substring arg 1 must be a Unicode object");
+    }
+    ctx_info->is_valid = false;
+    HPy universal_result = HPyUnicode_Substring(uctx, uh_str, start, end);
+    ctx_info->is_valid = true;
+    return DHPy_open(dctx, universal_result);
+}

hpy/tools/autogen/debug.py

@@ -92,6 +92,7 @@ class autogen_debug_wrappers(AutoGenFile):
         'HPyContextVar_Get',
         'HPyType_GetName',
         'HPyType_IsSubtype',
+        'HPyUnicode_Substring',
     }
 
     def generate(self):

hpy/tools/autogen/public_api.h

@@ -406,7 +406,7 @@ int HPy_TypeCheck(HPyContext *ctx, HPy obj, HPy type);
  * Return the type's name.
  *
  * Equivalent to getting the type's ``__name__`` attribute. If you want to
- * retrieve the type's name as Python unicode object, then just use
+ * retrieve the type's name as a handle that refers to a ``str``, then just use
  * ``HPy_GetAttr_s(ctx, type, "__name__")``.
  *
  * :param ctx:
@@ -536,6 +536,62 @@ HPy HPyUnicode_DecodeASCII(HPyContext *ctx, const char *ascii, HPy_ssize_t size,
 HPy_ID(197)
 HPy HPyUnicode_DecodeLatin1(HPyContext *ctx, const char *latin1, HPy_ssize_t size, const char *errors);
 
+/**
+ * Decode a bytes-like object to a Unicode object.
+ *
+ * The bytes of the bytes-like object are decoded according to the given
+ * encoding and using the error handling defined by ``errors``.
+ *
+ * :param ctx:
+ *     The execution context.
+ * :param obj:
+ *     A bytes-like object. This can be, for example, Python *bytes*,
+ *     *bytearray*, *memoryview*, *array.array* and objects that support the
+ *     Buffer protocol. If this argument is `HPy_NULL``, a ``SystemError`` will
+ *     be raised. If the argument is not a bytes-like object, a ``TypeError``
+ *     will be raised.
+ * :param encoding:
+ *     The name (UTF-8 encoded C string) of the encoding to use. If the encoding
+ *     does not exist, a ``LookupError`` will be raised. If this argument is
+ *     ``NULL``, the default encoding ``UTF-8`` will be used.
+ * :param errors:
+ *     The error handling (UTF-8 encoded C string) to use when decoding. The
+ *     possible values depend on the used encoding. This argument may be
+ *     ``NULL`` in which case it will default to ``"strict"``.
+ *
+ * :returns:
+ *     A handle to a ``str`` object created from the decoded bytes or
+ *     ``HPy_NULL`` in case of errors.
+ */
+HPy_ID(255)
+HPy HPyUnicode_FromEncodedObject(HPyContext *ctx, HPy obj, const char *encoding, const char *errors);
+
+/**
+ * Return a substring of ``str``, from character index ``start`` (included) to
+ * character index ``end`` (excluded).
+ *
+ * Indices ``start`` and ``end`` must not be negative, otherwise an
+ * ``IndexError`` will be raised. If ``start >= len(str)`` or if
+ * ``end < start``, an empty string will be returned. If ``end > len(str)`` then
+ * ``end == len(str)`` will be assumed.
+ *
+ * :param ctx:
+ *     The execution context.
+ * :param str:
+ *     A Python Unicode object (must not be ``HPy_NULL``). Otherwise, the
+ *     behavior is undefined (verification of the argument is only done in
+ *     debug mode).
+ * :param start:
+ *     The non-negative start index (inclusive).
+ * :param end:
+ *    The non-negative end index (exclusive).
+ *
+ * :returns:
+ *     The requested substring or ``HPy_NULL`` in case of an error.
+ */
+HPy_ID(256)
+HPy HPyUnicode_Substring(HPyContext *ctx, HPy str, HPy_ssize_t start, HPy_ssize_t end);
+
 /* listobject.h */
 HPy_ID(198)
 int HPyList_Check(HPyContext *ctx, HPy h);
@@ -550,6 +606,41 @@ int HPyDict_Check(HPyContext *ctx, HPy h);
 HPy_ID(202)
 HPy HPyDict_New(HPyContext *ctx);
 
+/**
+ * Returns a list of all keys from the dictionary.
+ *
+ * Note: This function will directly access the storage of the dict object and
+ * therefore ignores if method ``keys`` was overwritten.
+ *
+ * :param ctx:
+ *     The execution context.
+ * :param h:
+ *     A Python dict object. If this argument is ``HPy_NULL`` or not an
+ *     instance of a Python dict, a ``SystemError`` will be raised.
+ *
+ * :returns:
+ *     A Python list object containing all keys of the given dictionary or
+ *     ``HPy_NULL`` in case of an error.
+ */
+HPy_ID(257)
+HPy HPyDict_Keys(HPyContext *ctx, HPy h);
+
+/**
+ * Creates a copy of the provided Python dict object.
+ *
+ * :param ctx:
+ *     The execution context.
+ * :param h:
+ *     A Python dict object. If this argument is ``HPy_NULL`` or not an
+ *     instance of a Python dict, a ``SystemError`` will be raised.
+ *
+ * :returns:
+ *     Return a new dictionary that contains the same key-value pairs as ``h``
+ *     or ``HPy_NULL`` in case of an error.
+ */
+HPy_ID(258)
+HPy HPyDict_Copy(HPyContext *ctx, HPy h);
+
 /* tupleobject.h */
 HPy_ID(203)
 int HPyTuple_Check(HPyContext *ctx, HPy h);