Visitar URL original
gh-119609: Add PyUnicode_Export() and PyUnicode_Import() functions by vstinner · Pull Request #119610 · python/cpython · GitHub
Skip to content

gh-119609: Add PyUnicode_Export() and PyUnicode_Import() functions #119610

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
vstinner wants to merge 13 commits into from
+521 −0
Closed

gh-119609: Add PyUnicode_Export() and PyUnicode_Import() functions #119610

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
32a373a
gh-119609: Add PyUnicode_Export() function
vstinner May 27, 2024
c42cebc
stable_abi.toml: Add constants
vstinner Jun 21, 2024
1310b6d
Export UCS1 as UCS2
vstinner Jun 21, 2024
1c78103
Add test_unicode_export_import_roundtrip()
vstinner Jun 21, 2024
62f4598
Rename size parameter to nbytes
vstinner Jun 21, 2024
53cd937
Update doc
vstinner Jun 21, 2024
3a08491
Make sure that exported string ends with NUL character
vstinner Jun 21, 2024
a9a9083
tests: rename size to nbytes
vstinner Jun 21, 2024
1297c52
Update Doc/c-api/unicode.rst
vstinner Jun 22, 2024
d8e3d5d
Update Doc/c-api/unicode.rst
vstinner Jun 22, 2024
da67c89
Address Petr's review
vstinner Jun 22, 2024
894921a
Merge branch 'main' into unicode_native
vstinner Jun 22, 2024
f3857d8
Fix doc: add uint32_t type
vstinner Jun 26, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Export UCS1 as UCS2 
Co-Authored-By: Petr Viktorin <encukou@gmail.com>
  • Loading branch information
@vstinner @encukou
vstinner and encukou committed Jun 21, 2024
commit 1310b6d80ede2eb3aa3f15bdd299bdfde411889b
6 changes: 6 additions & 0 deletions Lib/test/test_capi/test_unicode.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1713,6 +1713,12 @@ def test_unicode_export(self):
self.assertEqual(unicode_export("abc", PyUnicode_FORMAT_UCS1),
(b'abc', PyUnicode_FORMAT_UCS1))

# export ASCII and UCS1 to UCS2
self.assertEqual(unicode_export("abc", PyUnicode_FORMAT_UCS2),
('abc'.encode(ucs2_enc), PyUnicode_FORMAT_UCS2))
self.assertEqual(unicode_export("latin1:\xe9", PyUnicode_FORMAT_UCS2),
('latin1:\xe9'.encode(ucs2_enc), PyUnicode_FORMAT_UCS2))

# always export to UCS4
self.assertEqual(unicode_export("abc", PyUnicode_FORMAT_UCS4),
('abc'.encode(ucs4_enc), PyUnicode_FORMAT_UCS4))
Expand Down
30 changes: 27 additions & 3 deletions Objects/unicodeobject.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2105,9 +2105,9 @@ PyUnicode_Export(PyObject *unicode, uint32_t requested_formats,
PyErr_Format(PyExc_TypeError, "must be str, not %T", unicode);
goto error;
}

Py_ssize_t len = PyUnicode_GET_LENGTH(unicode);

// Native ASCII
if (PyUnicode_IS_ASCII(unicode)
&& (requested_formats & PyUnicode_FORMAT_ASCII))
{
Expand All @@ -2116,6 +2116,7 @@ PyUnicode_Export(PyObject *unicode, uint32_t requested_formats,
return PyUnicode_1BYTE_DATA(unicode);
}

// Native UCS1
int kind = PyUnicode_KIND(unicode);
if (kind == PyUnicode_1BYTE_KIND
&& (requested_formats & PyUnicode_FORMAT_UCS1))
Expand All @@ -2125,6 +2126,7 @@ PyUnicode_Export(PyObject *unicode, uint32_t requested_formats,
return PyUnicode_1BYTE_DATA(unicode);
}

// Native UCS2
if (kind == PyUnicode_2BYTE_KIND
&& (requested_formats & PyUnicode_FORMAT_UCS2))
{
Expand All @@ -2133,6 +2135,28 @@ PyUnicode_Export(PyObject *unicode, uint32_t requested_formats,
return PyUnicode_2BYTE_DATA(unicode);
}

// Convert ASCII or UCS1 to UCS2
if (kind == PyUnicode_1BYTE_KIND
&& requested_formats & PyUnicode_FORMAT_UCS2)
{
Py_UCS2 *ucs2 = PyMem_Malloc((len + 1) * sizeof(Py_UCS2));
if (!ucs2) {
PyErr_NoMemory();
goto error;
}

_PyUnicode_CONVERT_BYTES(Py_UCS1, Py_UCS2,
PyUnicode_1BYTE_DATA(unicode),
PyUnicode_1BYTE_DATA(unicode) + len,
ucs2);
ucs2[len] = 0;

*format = PyUnicode_FORMAT_UCS2;
*size = len * 2;
return ucs2;
}

// Native UCS4
if (kind == PyUnicode_4BYTE_KIND
&& (requested_formats & PyUnicode_FORMAT_UCS4))
{
Expand All @@ -2141,8 +2165,8 @@ PyUnicode_Export(PyObject *unicode, uint32_t requested_formats,
return PyUnicode_4BYTE_DATA(unicode);
}

// Convert ASCII, UCS1 or UCS2 to UCS4
if (requested_formats & PyUnicode_FORMAT_UCS4) {
// Convert UCS1 or UCS2 to UCS4
Py_UCS4 *ucs4 = PyUnicode_AsUCS4Copy(unicode);
if (ucs4 == NULL) {
goto error;
Expand All @@ -2152,6 +2176,7 @@ PyUnicode_Export(PyObject *unicode, uint32_t requested_formats,
return ucs4;
}

// Convert to UTF-8
if (requested_formats & PyUnicode_FORMAT_UTF8) {
// Encode UCS1, UCS2 or UCS4 to UTF-8
const char *utf8 = PyUnicode_AsUTF8AndSize(unicode, size);
Expand All @@ -2164,7 +2189,6 @@ PyUnicode_Export(PyObject *unicode, uint32_t requested_formats,

PyErr_Format(PyExc_ValueError, "unable to find a matching export format");


error:
*size = 0;
*format = 0;
Expand Down