8000 gh-95504: Fix sign placement in PyUnicode_FromFormat by philg314 · Pull Request #95505 · python/cpython · GitHub
[go: up one dir, main page]

Skip to content

gh-95504: Fix sign placement in PyUnicode_FromFormat #95505

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.

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
wants to merge 8 commits into from
Closed
Show file tree
Hide file tree
Changes from 1 commit
Commits
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
Fix negative numbers in PyUnicode_FromFormat
  • Loading branch information
philg314 committed Jul 31, 2022
commit bd72e2a6f012bea145bef94471f1247492796b35
1 change: 1 addition & 0 deletions Misc/ACKS
Original file line number Diff line number Diff line change
Expand Up @@ -609,6 +609,7 @@ Jan-Philip Gehrcke
Thomas Gellekum
Gabriel Genellina
Christos Georgiou
Philip Georgi
Elazar (אלעזר) Gershuni
Ben Gertzfield
Nadim Ghaznavi
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,2 @@
Fix sign placement when specifying width or precision in
``PyUnicode_FromFormat``. Patch by Philip Georgi.
96 changes: 48 additions & 48 deletions Modules/_testcapimodule.c
CHECK_FORMAT_1("%05.7li", "000-123", (long)-123);
Original file line number Diff line number Diff line change
Expand Up @@ -3331,14 +3331,14 @@ test_string_from_format(PyObject *self, PyObject *Py_UNUSED(ignored))
CHECK_FORMAT_1("%05zu", "00123", (size_t)123);
CHECK_FORMAT_1("%05x", "0007b", (int)123);

CHECK_FORMAT_1("%05d", "0-123", (int)-123);
CHECK_FORMAT_1("%05i", "0-123", (int)-123);
CHECK_FORMAT_1("%05ld", "0-123", (long)-123);
CHECK_FORMAT_1("%05li", "0-123", (long)-123);
CHECK_FORMAT_1("%05lld", "0-123", (long long)-123);
CHECK_FORMAT_1("%05lli", "0-123", (long long)-123);
CHECK_FORMAT_1("%05zd", "0-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%05zi", "0-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%05d", "-0123", (int)-123);
CHECK_FORMAT_1("%05i", "-0123", (int)-123);
CHECK_FORMAT_1("%05ld", "-0123", (long)-123);
CHECK_FORMAT_1("%05li", "-0123", (long)-123);
CHECK_FORMAT_1("%05lld", "-0123", (long long)-123);
CHECK_FORMAT_1("%05lli", "-0123", (long long)-123);
CHECK_FORMAT_1("%05zd", "-0123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%05zi", "-0123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%09x", "0ffffff85", (int)-123);

// Integers: precision < length
Expand Down Expand Up @@ -3381,14 +3381,14 @@ test_string_from_format(PyObject *self, PyObject *Py_UNUSED(ignored))
CHECK_FORMAT_1("%.5zu", "00123", (size_t)123);
CHECK_FORMAT_1("%.5x", "0007b", (int)123);

CHECK_FORMAT_1("%.5d", "0-123", (int)-123);
CHECK_FORMAT_1("%.5i", "0-123", (int)-123);
CHECK_FORMAT_1("%.5ld", "0-123", (long)-123);
CHECK_FORMAT_1("%.5li", "0-123", (long)-123);
CHECK_FORMAT_1("%.5lld", "0-123", (long long)-123);
CHECK_FORMAT_1("%.5lli", "0-123", (long long)-123);
CHECK_FORMAT_1("%.5zd", "0-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%.5zi", "0-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%.5d", "-00123", (int)-123);
CHECK_FORMAT_1("%.5i", "-00123", (int)-123);
CHECK_FORMAT_1("%.5ld", "-00123", (long)-123);
CHECK_FORMAT_1("%.5li", "-00123", (long)-123);
CHECK_FORMAT_1("%.5lld", "-00123", (long long)-123);
CHECK_FORMAT_1("%.5lli", "-00123", (long long)-123);
CHECK_FORMAT_1("%.5zd", "-00123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%.5zi", "-00123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%.9x", "0ffffff85", (int)-123);

// Integers: width > precision > length
Expand All @@ -3406,14 +3406,14 @@ test_string_from_format(PyObject *self, PyObject *Py_UNUSED(ignored))
CHECK_FORMAT_1("%7.5zu", " 00123", (size_t)123);
CHECK_FORMAT_1("%7.5x", " 0007b", (int)123);

CHECK_FORMAT_1("%7.5d", " 0-123", (int)-123);
CHECK_FORMAT_1("%7.5i", " 0-123", (int)-123);
CHECK_FORMAT_1("%7.5ld", " 0-123", (long)-123);
CHECK_FORMAT_1("%7.5li", " 0-123", (long)-123);
CHECK_FORMAT_1("%7.5lld", " 0-123", (long long)-123);
CHECK_FORMAT_1("%7.5lli", " 0-123", (long long)-123);
CHECK_FORMAT_1("%7.5zd", " 0-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%7.5zi", " 0-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%7.5d", " -00123", (int)-123);
CHECK_FORMAT_1("%7.5i", " -00123", (int)-123);
CHECK_FORMAT_1("%7.5ld", " -00123", (long)-123);
CHECK_FORMAT_1("%7.5li", " -00123", (long)-123);
CHECK_FORMAT_1("%7.5lld", " -00123", (long long)-123);
CHECK_FORMAT_1("%7.5lli", " -00123", (long long)-123);
CHECK_FORMAT_1("%7.5zd", " -00123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%7.5zi", " -00123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%10.9x", " 0ffffff85", (int)-123);

// Integers: width > precision > length, 0-flag
Expand All @@ -3431,14 +3431,14 @@ test_string_from_format(PyObject *self, PyObject *Py_UNUSED(ignored))
CHECK_FORMAT_1("%07.5zu", "0000123", (size_t)123);
CHECK_FORMAT_1("%07.5x", "000007b", (int)123);

CHECK_FORMAT_1("%07.5d", "000-123", (int)-123);
CHECK_FORMAT_1("%07.5i", "000-123", (int)-123);
CHECK_FORMAT_1("%07.5ld", "000-123", (long)-123);
CHECK_FORMAT_1("%07.5li", "000-123", (long)-123);
CHECK_FORMAT_1("%07.5lld", "000-123", (long long)-123);
CHECK_FORMAT_1("%07.5lli", "000-123", (long long)-123);
CHECK_FORMAT_1("%07.5zd", "000-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%07.5zi", "000-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%07.5d", "-000123", (int)-123);
CHECK_FORMAT_1("%07.5i", "-000123", (int)-123);
CHECK_FORMAT_1("%07.5ld", "-000123", (long)-123);
CHECK_FORMAT_1("%07.5li", "-000123", (long)-123);
CHECK_FORMAT_1("%07.5lld", "-000123", (long long)-123);
CHECK_FORMAT_1("%07.5lli", "-000123", (long long)-123);
CHECK_FORMAT_1("%07.5zd", "-000123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%07.5zi", "-000123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%010.9x", "00ffffff85", (int)-123);

// Integers: precision > width > length
Expand All @@ -3456,14 +3456,14 @@ test_string_from_format(PyObject *self, PyObject *Py_UNUSED(ignored))
CHECK_FORMAT_1("%5.7zu", "0000123", (size_t)123);
CHECK_FORMAT_1("%5.7x", "000007b", (int)123);

CHECK_FORMAT_1("%5.7d", "000-123", (int)-123);
CHECK_FORMAT_1("%5.7i", "000-123", (int)-123);
CHECK_FORMAT_1("%5.7ld", "000-123", (long)-123);
CHECK_FORMAT_1("%5.7li", "000-123", (long)-123);
CHECK_FORMAT_1("%5.7lld", "000-123", (long long)-123);
CHECK_FORMAT_1("%5.7lli", "000-123", (long long)-123);
CHECK_FORMAT_1("%5.7zd", "000-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%5.7zi", "000-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%5.7d", "-0000123", (int)-123);
CHECK_FORMAT_1("%5.7i", "-0000123", (int)-123);
CHECK_FORMAT_1("%5.7ld", "-0000123", (long)-123);
CHECK_FORMAT_1("%5.7li", "-0000123", (long)-123);
CHECK_FORMAT_1("%5.7lld", "-0000123", (long long)-123);
CHECK_FORMAT_1("%5.7lli", "-0000123", (long long)-123);
CHECK_FORMAT_1("%5.7zd", "-0000123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%5.7zi", "-0000123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%9.10x", "00ffffff85", (int)-123);

// Integers: precision > width > length, 0-flag
Expand All @@ -3481,14 +3481,14 @@ test_string_from_format(PyObject *self, PyObject *Py_UNUSED(ignored))
CHECK_FORMAT_1("%05.7zu", "0000123", (size_t)123);
CHECK_FORMAT_1("%05.7x", "000007b", (int)123);

CHECK_FORMAT_1("%05.7d", "000-123", (int)-123);
CHECK_FORMAT_1("%05.7i", "000-123", (int)-123);
CHECK_FORMAT_1("%05.7ld", "000-123", (long)-123);
CHECK_FORMAT_1("%05.7lld", "000-123", (long long)-123);
CHECK_FORMAT_1("%05.7lli", "000-123", (long long)-123);
CHECK_FORMAT_1("%05.7zd", "000-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%05.7zi", "000-123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%05.7d", "-0000123", (int)-123);
CHECK_FORMAT_1("%05.7i", "-0000123", (int)-123);
CHECK_FORMAT_1("%05.7ld", "-0000123", (long)-123);
CHECK_FORMAT_1("%05.7li", "-0000123", (long)-123);
CHECK_FORMAT_1("%05.7lld", "-0000123", (long long)-123);
CHECK_FORMAT_1("%05.7lli", "-0000123", (long long)-123);
CHECK_FORMAT_1("%05.7zd", "-0000123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%05.7zi", "-0000123", (Py_ssize_t)-123);
CHECK_FORMAT_1("%09.10x", "00ffffff85", (int)-123);

// Integers: precision = 0, arg = 0 (empty string in C)
Expand Down
25 changes: 19 additions & 6 deletions Objects/unicodeobject.c
Original file line number Diff line number Diff line change
Expand Up @@ -2482,29 +2482,42 @@ unicode_fromformat_arg(_PyUnicodeWriter *writer,
}
assert(len >= 0);

if (precision < len)
precision = len;
int negative = buffer[0]=='-'?1:0;
len -= negative;

precision = Py_MAX(precision, len);
width = Py_MAX(width, precision + negative);

arglen = Py_MAX(precision, width);
if (_PyUnicodeWriter_Prepare(writer, arglen, 127) == -1)
return NULL;

if (width > precision) {
Py_UCS4 fillchar;
fill = width - precision;
fillchar = zeropad?'0':' ';
if (negative && zeropad) {
if (_PyUnicodeWriter_WriteChar(writer, '-') == -1)
return NULL;
}

Py_UCS4 fillchar = zeropad?'0':' ';
fill = width - precision - negative;
if (PyUnicode_Fill(writer->buffer, writer->pos, fill, fillchar) == -1)
return NULL;
writer->pos += fill;

if (negative && !zeropad) {
if (_PyUnicodeWriter_WriteChar(writer, '-') == -1)
return NULL;
}
Comment on lines +2507 to +2510
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It can be simpler. Instead of writing minus explicitly, you can write it as a part of the buffer.

You will get the same result if remove this if and pass &buffer[negative && zeropad] instead of &buffer[negative] to _PyUnicodeWriter_WriteASCIIString().

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That wouldn't work with precision: %.5d with -123 would be 00-123.

}

if (precision > len) {
fill = precision - len;
if (PyUnicode_Fill(writer->buffer, writer->pos, fill, '0') == -1)
return NULL;
writer->pos += fill;
}

if (_PyUnicodeWriter_WriteASCIIString(writer, buffer, len) < 0)
if (_PyUnicodeWriter_WriteASCIIString(writer, &buffer[negative], len) < 0)
return NULL;
break;
}
Expand Down
0