Mercurial > hg-stable
changeset 24032:c53bc2e52514
parsers: use k instead of n for PyArg_ParseTuple because python 2.4 is awful
author | Augie Fackler <augie@google.com> |
---|---|
date | Wed, 04 Feb 2015 11:38:30 -0500 |
parents | fe1abb5d92f6 |
children | ed5e8a9598ce |
files | mercurial/parsers.c |
diffstat | 1 files changed, 5 insertions(+), 2 deletions(-) [+] |
line wrap: on
line diff
--- a/mercurial/parsers.c Wed Feb 04 21:37:06 2015 +0900 +++ b/mercurial/parsers.c Wed Feb 04 11:38:30 2015 -0500 @@ -2262,10 +2262,13 @@ static PyObject *fm1readmarkers(PyObject *self, PyObject *args) { const char *data; - Py_ssize_t datalen, offset, stop; + Py_ssize_t datalen; + /* only unsigned long because python 2.4, should be Py_ssize_t */ + unsigned long offset, stop; PyObject *markers = NULL; - if (!PyArg_ParseTuple(args, "s#nn", &data, &datalen, &offset, &stop)) { + /* replace kk with nn when we drop Python 2.4 */ + if (!PyArg_ParseTuple(args, "s#kk", &data, &datalen, &offset, &stop)) { return NULL; } data += offset;