static int
join_append(WriterObj *self, PyObject *field, int quoted)
{
unsigned int field_kind = -1;
void *field_data = NULL;
Py_ssize_t field_len = 0;
Py_ssize_t rec_len;
if (field != NULL) {
if (PyUnicode_READY(field) == -1)
return 0;
field_kind = PyUnicode_KIND(field);
field_data = PyUnicode_DATA(field);
field_len = PyUnicode_GET_LENGTH(field);
}
rec_len = join_append_data(self, field_kind, field_data, field_len,
"ed, 0);
if (rec_len < 0)
return 0;
/* grow record buffer if necessary */
if (!join_check_rec_size(self, rec_len))
return 0;
self->rec_len = join_append_data(self, field_kind, field_data, field_len,
"ed, 1);
self->num_fields++;
return 1;
}
static PyObject *
csv_register_dialect(PyObject *module, PyObject *args, PyObject *kwargs)
{
PyObject *name_obj, *dialect_obj = NULL;
PyObject *dialect;
if (!PyArg_UnpackTuple(args, "", 1, 2, &name_obj, &dialect_obj))
return NULL;
if (!PyUnicode_Check(name_obj)) {
PyErr_SetString(PyExc_TypeError,
"dialect name must be a string");
return NULL;
}
if (PyUnicode_READY(name_obj) == -1)
return NULL;
dialect = _call_dialect(dialect_obj, kwargs);
if (dialect == NULL)
return NULL;
if (PyDict_SetItem(_csvstate_global->dialects, name_obj, dialect) < 0) {
Py_DECREF(dialect);
return NULL;
}
Py_DECREF(dialect);
Py_RETURN_NONE;
}
/* Given a path to a .pyc file in the archive, return the
modification time of the matching .py file, or 0 if no source
is available. */
static time_t
get_mtime_of_source(ZipImporter *self, PyObject *path)
{
PyObject *toc_entry, *stripped;
time_t mtime;
/* strip 'c' or 'o' from *.py[co] */
if (PyUnicode_READY(path) == -1)
return (time_t)-1;
stripped = PyUnicode_FromKindAndData(PyUnicode_KIND(path),
PyUnicode_DATA(path),
PyUnicode_GET_LENGTH(path) - 1);
if (stripped == NULL)
return (time_t)-1;
toc_entry = PyDict_GetItem(self->files, stripped);
Py_DECREF(stripped);
if (toc_entry != NULL && PyTuple_Check(toc_entry) &&
PyTuple_Size(toc_entry) == 8) {
/* fetch the time stamp of the .py file for comparison
with an embedded pyc time stamp */
int time, date;
time = PyLong_AsLong(PyTuple_GetItem(toc_entry, 5));
date = PyLong_AsLong(PyTuple_GetItem(toc_entry, 6));
mtime = parse_dostime(time, date);
} else
mtime = 0;
return mtime;
}
/* return fullname.split(".")[-1] */
static PyObject *
get_subname(PyObject *fullname)
{
Py_ssize_t len, dot;
if (PyUnicode_READY(fullname) < 0)
return NULL;
len = PyUnicode_GET_LENGTH(fullname);
dot = PyUnicode_FindChar(fullname, '.', 0, len, -1);
if (dot == -1) {
Py_INCREF(fullname);
return fullname;
} else
return PyUnicode_Substring(fullname, dot+1, len);
}
static PyObject *
zipimport_zipimporter_get_data_impl(ZipImporter *self, PyObject *path)
/*[clinic end generated code: output=65dc506aaa268436 input=fa6428b74843c4ae]*/
{
PyObject *key;
PyObject *toc_entry;
Py_ssize_t path_start, path_len, len;
if (self->archive == NULL) {
PyErr_SetString(PyExc_ValueError,
"zipimporter.__init__() wasn't called");
return NULL;
}
#ifdef ALTSEP
path = _PyObject_CallMethodId((PyObject *)&PyUnicode_Type, &PyId_replace,
"OCC", path, ALTSEP, SEP);
if (!path)
return NULL;
#else
Py_INCREF(path);
#endif
if (PyUnicode_READY(path) == -1)
goto error;
path_len = PyUnicode_GET_LENGTH(path);
len = PyUnicode_GET_LENGTH(self->archive);
path_start = 0;
if (PyUnicode_Tailmatch(path, self->archive, 0, len, -1)
&& PyUnicode_READ_CHAR(path, len) == SEP) {
path_start = len + 1;
}
key = PyUnicode_Substring(path, path_start, path_len);
if (key == NULL)
goto error;
toc_entry = PyDict_GetItem(self->files, key);
if (toc_entry == NULL) {
PyErr_SetFromErrnoWithFilenameObject(PyExc_OSError, key);
Py_DECREF(key);
goto error;
}
Py_DECREF(key);
Py_DECREF(path);
return get_data(self->archive, toc_entry);
error:
Py_DECREF(path);
return NULL;
}
static PyObject *
zipimporter_get_data(PyObject *obj, PyObject *args)
{
ZipImporter *self = (ZipImporter *)obj;
PyObject *path, *key;
#ifdef ALTSEP
_Py_IDENTIFIER(replace);
#endif
PyObject *toc_entry;
Py_ssize_t path_start, path_len, len;
if (!PyArg_ParseTuple(args, "U:zipimporter.get_data", &path))
return NULL;
#ifdef ALTSEP
path = _PyObject_CallMethodId(path, &PyId_replace, "CC", ALTSEP, SEP);
if (!path)
return NULL;
#else
Py_INCREF(path);
#endif
if (PyUnicode_READY(path) == -1)
goto error;
path_len = PyUnicode_GET_LENGTH(path);
len = PyUnicode_GET_LENGTH(self->archive);
path_start = 0;
if (PyUnicode_Tailmatch(path, self->archive, 0, len, -1)
&& PyUnicode_READ_CHAR(path, len) == SEP) {
path_start = len + 1;
}
key = PyUnicode_Substring(path, path_start, path_len);
if (key == NULL)
goto error;
toc_entry = PyDict_GetItem(self->files, key);
if (toc_entry == NULL) {
PyErr_SetFromErrnoWithFilenameObject(PyExc_IOError, key);
Py_DECREF(key);
goto error;
}
Py_DECREF(key);
Py_DECREF(path);
return get_data(self->archive, toc_entry);
error:
Py_DECREF(path);
return NULL;
}
/* Intern selected string constants */
static int
intern_string_constants(PyObject *tuple)
{
int modified = 0;
Py_ssize_t i;
for (i = PyTuple_GET_SIZE(tuple); --i >= 0; ) {
PyObject *v = PyTuple_GET_ITEM(tuple, i);
if (PyUnicode_CheckExact(v)) {
if (PyUnicode_READY(v) == -1) {
PyErr_Clear();
continue;
}
if (all_name_chars(v)) {
PyObject *w = v;
PyUnicode_InternInPlace(&v);
if (w != v) {
PyTuple_SET_ITEM(tuple, i, v);
modified = 1;
}
}
}
else if (PyTuple_CheckExact(v)) {
intern_string_constants(v);
}
else if (PyFrozenSet_CheckExact(v)) {
PyObject *w = v;
PyObject *tmp = PySequence_Tuple(v);
if (tmp == NULL) {
PyErr_Clear();
continue;
}
if (intern_string_constants(tmp)) {
v = PyFrozenSet_New(tmp);
if (v == NULL) {
PyErr_Clear();
}
else {
PyTuple_SET_ITEM(tuple, i, v);
Py_DECREF(w);
modified = 1;
}
}
Py_DECREF(tmp);
}
}
return modified;
}
static int
_set_str(const char *name, PyObject **target, PyObject *src, const char *dflt)
{
if (src == NULL)
*target = PyUnicode_DecodeASCII(dflt, strlen(dflt), NULL);
else {
if (src == Py_None)
*target = NULL;
else if (!PyUnicode_Check(src)) {
PyErr_Format(PyExc_TypeError,
"\"%s\" must be a string", name);
return -1;
}
else {
if (PyUnicode_READY(src) == -1)
return -1;
Py_INCREF(src);
Py_XSETREF(*target, src);
}
}
return 0;
}
/* Verify that the identifier follows PEP 3131.
All identifier strings are guaranteed to be "ready" unicode objects.
*/
static int
verify_identifier(struct tok_state *tok)
{
PyObject *s;
int result;
if (tok->decoding_erred)
return 0;
s = PyUnicode_DecodeUTF8(tok->start, tok->cur - tok->start, NULL);
if (s == NULL || PyUnicode_READY(s) == -1) {
if (PyErr_ExceptionMatches(PyExc_UnicodeDecodeError)) {
PyErr_Clear();
tok->done = E_IDENTIFIER;
} else {
tok->done = E_ERROR;
}
return 0;
}
result = PyUnicode_IsIdentifier(s);
Py_DECREF(s);
if (result == 0)
tok->done = E_IDENTIFIER;
return result;
}
static int
all_name_chars(PyObject *o)
{
static char ok_name_char[256];
static unsigned char *name_chars = (unsigned char *)NAME_CHARS;
PyUnicodeObject *u = (PyUnicodeObject *)o;
const unsigned char *s;
if (!PyUnicode_Check(o) || PyUnicode_READY(u) == -1 ||
PyUnicode_MAX_CHAR_VALUE(u) >= 128)
return 0;
if (ok_name_char[*name_chars] == 0) {
unsigned char *p;
for (p = name_chars; *p; p++)
ok_name_char[*p] = 1;
}
s = PyUnicode_1BYTE_DATA(u);
while (*s) {
if (ok_name_char[*s++] == 0)
return 0;
}
return 1;
}
static int
format_long_internal(PyObject *value, const InternalFormatSpec *format,
_PyUnicodeWriter *writer)
{
int result = -1;
Py_UCS4 maxchar = 127;
PyObject *tmp = NULL;
Py_ssize_t inumeric_chars;
Py_UCS4 sign_char = '\0';
Py_ssize_t n_digits; /* count of digits need from the computed
string */
Py_ssize_t n_remainder = 0; /* Used only for 'c' formatting, which
produces non-digits */
Py_ssize_t n_prefix = 0; /* Count of prefix chars, (e.g., '0x') */
Py_ssize_t n_total;
Py_ssize_t prefix = 0;
NumberFieldWidths spec;
long x;
/* Locale settings, either from the actual locale or
from a hard-code pseudo-locale */
LocaleInfo locale = STATIC_LOCALE_INFO_INIT;
/* no precision allowed on integers */
if (format->precision != -1) {
PyErr_SetString(PyExc_ValueError,
"Precision not allowed in integer format specifier");
goto done;
}
/* special case for character formatting */
if (format->type == 'c') {
/* error to specify a sign */
if (format->sign != '\0') {
PyErr_SetString(PyExc_ValueError,
"Sign not allowed with integer"
" format specifier 'c'");
goto done;
}
/* taken from unicodeobject.c formatchar() */
/* Integer input truncated to a character */
x = PyLong_AsLong(value);
if (x == -1 && PyErr_Occurred())
goto done;
if (x < 0 || x > 0x10ffff) {
PyErr_SetString(PyExc_OverflowError,
"%c arg not in range(0x110000)");
goto done;
}
tmp = PyUnicode_FromOrdinal(x);
inumeric_chars = 0;
n_digits = 1;
maxchar = Py_MAX(maxchar, (Py_UCS4)x);
/* As a sort-of hack, we tell calc_number_widths that we only
have "remainder" characters. calc_number_widths thinks
these are characters that don't get formatted, only copied
into the output string. We do this for 'c' formatting,
because the characters are likely to be non-digits. */
n_remainder = 1;
}
else {
int base;
int leading_chars_to_skip = 0; /* Number of characters added by
PyNumber_ToBase that we want to
skip over. */
/* Compute the base and how many characters will be added by
PyNumber_ToBase */
switch (format->type) {
case 'b':
base = 2;
leading_chars_to_skip = 2; /* 0b */
break;
case 'o':
base = 8;
leading_chars_to_skip = 2; /* 0o */
break;
case 'x':
case 'X':
base = 16;
leading_chars_to_skip = 2; /* 0x */
break;
default: /* shouldn't be needed, but stops a compiler warning */
case 'd':
case 'n':
base = 10;
break;
}
if (format->sign != '+' && format->sign != ' '
&& format->width == -1
&& format->type != 'X' && format->type != 'n'
&& !format->thousands_separators
&& PyLong_CheckExact(value))
{
/* Fast path */
return _PyLong_FormatWriter(writer, value, base, format->alternate);
}
/* The number of prefix chars is the same as the leading
chars to skip */
if (format->alternate)
n_prefix = leading_chars_to_skip;
/* Do the hard part, converting to a string in a given base */
tmp = _PyLong_Format(value, base);
if (tmp == NULL || PyUnicode_READY(tmp) == -1)
goto done;
inumeric_chars = 0;
n_digits = PyUnicode_GET_LENGTH(tmp);
prefix = inumeric_chars;
/* Is a sign character present in the output? If so, remember it
and skip it */
if (PyUnicode_READ_CHAR(tmp, inumeric_chars) == '-') {
sign_char = '-';
++prefix;
++leading_chars_to_skip;
}
/* Skip over the leading chars (0x, 0b, etc.) */
n_digits -= leading_chars_to_skip;
inumeric_chars += leading_chars_to_skip;
}
/* Determine the grouping, separator, and decimal point, if any. */
if (get_locale_info(format->type == 'n' ? LT_CURRENT_LOCALE :
(format->thousands_separators ?
LT_DEFAULT_LOCALE :
LT_NO_LOCALE),
&locale) == -1)
goto done;
/* Calculate how much memory we'll need. */
n_total = calc_number_widths(&spec, n_prefix, sign_char, tmp, inumeric_chars,
inumeric_chars + n_digits, n_remainder, 0,
&locale, format, &maxchar);
/* Allocate the memory. */
if (_PyUnicodeWriter_Prepare(writer, n_total, maxchar) == -1)
goto done;
/* Populate the memory. */
result = fill_number(writer, &spec,
tmp, inumeric_chars, inumeric_chars + n_digits,
tmp, prefix, format->fill_char,
&locale, format->type == 'X');
done:
Py_XDECREF(tmp);
free_locale_info(&locale);
return result;
}
/* Fill in the digit parts of a numbers's string representation,
as determined in calc_number_widths().
Return -1 on error, or 0 on success. */
static int
fill_number(_PyUnicodeWriter *writer, const NumberFieldWidths *spec,
PyObject *digits, Py_ssize_t d_start, Py_ssize_t d_end,
PyObject *prefix, Py_ssize_t p_start,
Py_UCS4 fill_char,
LocaleInfo *locale, int toupper)
{
/* Used to keep track of digits, decimal, and remainder. */
Py_ssize_t d_pos = d_start;
const unsigned int kind = writer->kind;
const void *data = writer->data;
Py_ssize_t r;
if (spec->n_lpadding) {
_PyUnicode_FastFill(writer->buffer,
writer->pos, spec->n_lpadding, fill_char);
writer->pos += spec->n_lpadding;
}
if (spec->n_sign == 1) {
PyUnicode_WRITE(kind, data, writer->pos, spec->sign);
writer->pos++;
}
if (spec->n_prefix) {
_PyUnicode_FastCopyCharacters(writer->buffer, writer->pos,
prefix, p_start,
spec->n_prefix);
if (toupper) {
Py_ssize_t t;
for (t = 0; t < spec->n_prefix; t++) {
Py_UCS4 c = PyUnicode_READ(kind, data, writer->pos + t);
c = Py_TOUPPER(c);
assert (c <= 127);
PyUnicode_WRITE(kind, data, writer->pos + t, c);
}
}
writer->pos += spec->n_prefix;
}
if (spec->n_spadding) {
_PyUnicode_FastFill(writer->buffer,
writer->pos, spec->n_spadding, fill_char);
writer->pos += spec->n_spadding;
}
/* Only for type 'c' special case, it has no digits. */
if (spec->n_digits != 0) {
/* Fill the digits with InsertThousandsGrouping. */
char *pdigits;
if (PyUnicode_READY(digits))
return -1;
pdigits = PyUnicode_DATA(digits);
if (PyUnicode_KIND(digits) < kind) {
pdigits = _PyUnicode_AsKind(digits, kind);
if (pdigits == NULL)
return -1;
}
r = _PyUnicode_InsertThousandsGrouping(
writer->buffer, writer->pos,
spec->n_grouped_digits,
pdigits + kind * d_pos,
spec->n_digits, spec->n_min_width,
locale->grouping, locale->thousands_sep, NULL);
if (r == -1)
return -1;
assert(r == spec->n_grouped_digits);
if (PyUnicode_KIND(digits) < kind)
PyMem_Free(pdigits);
d_pos += spec->n_digits;
}
if (toupper) {
Py_ssize_t t;
for (t = 0; t < spec->n_grouped_digits; t++) {
Py_UCS4 c = PyUnicode_READ(kind, data, writer->pos + t);
c = Py_TOUPPER(c);
if (c > 127) {
PyErr_SetString(PyExc_SystemError, "non-ascii grouped digit");
return -1;
}
PyUnicode_WRITE(kind, data, writer->pos + t, c);
}
}
writer->pos += spec->n_grouped_digits;
if (spec->n_decimal) {
_PyUnicode_FastCopyCharacters(
writer->buffer, writer->pos,
locale->decimal_point, 0, spec->n_decimal);
writer->pos += spec->n_decimal;
d_pos += 1;
}
if (spec->n_remainder) {
_PyUnicode_FastCopyCharacters(
writer->buffer, writer->pos,
digits, d_pos, spec->n_remainder);
writer->pos += spec->n_remainder;
/* d_pos += spec->n_remainder; */
}
if (spec->n_rpadding) {
_PyUnicode_FastFill(writer->buffer,
writer->pos, spec->n_rpadding,
fill_char);
writer->pos += spec->n_rpadding;
}
return 0;
}
PyCodeObject *
PyCode_New(int argcount, int kwonlyargcount,
int nlocals, int stacksize, int flags,
PyObject *code, PyObject *consts, PyObject *names,
PyObject *varnames, PyObject *freevars, PyObject *cellvars,
PyObject *filename, PyObject *name, int firstlineno,
PyObject *lnotab)
{
PyCodeObject *co;
unsigned char *cell2arg = NULL;
Py_ssize_t i, n_cellvars;
/* Check argument types */
if (argcount < 0 || kwonlyargcount < 0 || nlocals < 0 ||
code == NULL ||
consts == NULL || !PyTuple_Check(consts) ||
names == NULL || !PyTuple_Check(names) ||
varnames == NULL || !PyTuple_Check(varnames) ||
freevars == NULL || !PyTuple_Check(freevars) ||
cellvars == NULL || !PyTuple_Check(cellvars) ||
name == NULL || !PyUnicode_Check(name) ||
filename == NULL || !PyUnicode_Check(filename) ||
lnotab == NULL || !PyBytes_Check(lnotab) ||
!PyObject_CheckReadBuffer(code)) {
PyErr_BadInternalCall();
return NULL;
}
/* Ensure that the filename is a ready Unicode string */
if (PyUnicode_READY(filename) < 0)
return NULL;
n_cellvars = PyTuple_GET_SIZE(cellvars);
intern_strings(names);
intern_strings(varnames);
intern_strings(freevars);
intern_strings(cellvars);
/* Intern selected string constants */
for (i = PyTuple_GET_SIZE(consts); --i >= 0; ) {
PyObject *v = PyTuple_GetItem(consts, i);
if (!all_name_chars(v))
continue;
PyUnicode_InternInPlace(&PyTuple_GET_ITEM(consts, i));
}
/* Create mapping between cells and arguments if needed. */
if (n_cellvars) {
Py_ssize_t total_args = argcount + kwonlyargcount +
((flags & CO_VARARGS) != 0) + ((flags & CO_VARKEYWORDS) != 0);
Py_ssize_t alloc_size = sizeof(unsigned char) * n_cellvars;
int used_cell2arg = 0;
cell2arg = PyMem_MALLOC(alloc_size);
if (cell2arg == NULL)
return NULL;
memset(cell2arg, CO_CELL_NOT_AN_ARG, alloc_size);
/* Find cells which are also arguments. */
for (i = 0; i < n_cellvars; i++) {
Py_ssize_t j;
PyObject *cell = PyTuple_GET_ITEM(cellvars, i);
for (j = 0; j < total_args; j++) {
PyObject *arg = PyTuple_GET_ITEM(varnames, j);
if (!PyUnicode_Compare(cell, arg)) {
cell2arg[i] = j;
used_cell2arg = 1;
break;
}
}
}
if (!used_cell2arg) {
PyMem_FREE(cell2arg);
cell2arg = NULL;
}
}
co = PyObject_NEW(PyCodeObject, &PyCode_Type);
if (co == NULL) {
if (cell2arg)
PyMem_FREE(cell2arg);
return NULL;
}
co->co_argcount = argcount;
co->co_kwonlyargcount = kwonlyargcount;
co->co_nlocals = nlocals;
co->co_stacksize = stacksize;
co->co_flags = flags;
Py_INCREF(code);
co->co_code = code;
Py_INCREF(consts);
co->co_consts = consts;
Py_INCREF(names);
co->co_names = names;
Py_INCREF(varnames);
co->co_varnames = varnames;
Py_INCREF(freevars);
co->co_freevars = freevars;
Py_INCREF(cellvars);
co->co_cellvars = cellvars;
co->co_cell2arg = cell2arg;
Py_INCREF(filename);
co->co_filename = filename;
Py_INCREF(name);
co->co_name = name;
co->co_firstlineno = firstlineno;
Py_INCREF(lnotab);
co->co_lnotab = lnotab;
co->co_zombieframe = NULL;
co->co_weakreflist = NULL;
return co;
}
static PyObject *
zipimport_zipimporter_load_module_impl(ZipImporter *self, PyObject *fullname)
/*[clinic end generated code: output=7303cebf88d47953 input=c236e2e8621f04ef]*/
{
PyObject *code = NULL, *mod, *dict;
PyObject *modpath = NULL;
int ispackage;
if (PyUnicode_READY(fullname) == -1)
return NULL;
code = get_module_code(self, fullname, &ispackage, &modpath);
if (code == NULL)
goto error;
mod = PyImport_AddModuleObject(fullname);
if (mod == NULL)
goto error;
dict = PyModule_GetDict(mod);
/* mod.__loader__ = self */
if (PyDict_SetItemString(dict, "__loader__", (PyObject *)self) != 0)
goto error;
if (ispackage) {
/* add __path__ to the module *before* the code gets
executed */
PyObject *pkgpath, *fullpath, *subname;
int err;
subname = get_subname(fullname);
if (subname == NULL)
goto error;
fullpath = PyUnicode_FromFormat("%U%c%U%U",
self->archive, SEP,
self->prefix, subname);
Py_DECREF(subname);
if (fullpath == NULL)
goto error;
pkgpath = Py_BuildValue("[N]", fullpath);
if (pkgpath == NULL)
goto error;
err = PyDict_SetItemString(dict, "__path__", pkgpath);
Py_DECREF(pkgpath);
if (err != 0)
goto error;
}
mod = PyImport_ExecCodeModuleObject(fullname, code, modpath, NULL);
Py_CLEAR(code);
if (mod == NULL)
goto error;
if (Py_VerboseFlag)
PySys_FormatStderr("import %U # loaded from Zip %U\n",
fullname, modpath);
Py_DECREF(modpath);
return mod;
error:
Py_XDECREF(code);
Py_XDECREF(modpath);
return NULL;
}
static PyObject *
Reader_iternext(ReaderObj *self)
{
PyObject *fields = NULL;
Py_UCS4 c;
Py_ssize_t pos, linelen;
unsigned int kind;
void *data;
PyObject *lineobj;
if (parse_reset(self) < 0)
return NULL;
do {
lineobj = PyIter_Next(self->input_iter);
if (lineobj == NULL) {
/* End of input OR exception */
if (!PyErr_Occurred() && (self->field_len != 0 ||
self->state == IN_QUOTED_FIELD)) {
if (self->dialect->strict)
PyErr_SetString(_csvstate_global->error_obj,
"unexpected end of data");
else if (parse_save_field(self) >= 0)
break;
}
return NULL;
}
if (!PyUnicode_Check(lineobj)) {
PyErr_Format(_csvstate_global->error_obj,
"iterator should return strings, "
"not %.200s "
"(did you open the file in text mode?)",
lineobj->ob_type->tp_name
);
Py_DECREF(lineobj);
return NULL;
}
if (PyUnicode_READY(lineobj) == -1) {
Py_DECREF(lineobj);
return NULL;
}
++self->line_num;
kind = PyUnicode_KIND(lineobj);
data = PyUnicode_DATA(lineobj);
pos = 0;
linelen = PyUnicode_GET_LENGTH(lineobj);
while (linelen--) {
c = PyUnicode_READ(kind, data, pos);
if (c == '\0') {
Py_DECREF(lineobj);
PyErr_Format(_csvstate_global->error_obj,
"line contains NULL byte");
goto err;
}
if (parse_process_char(self, c) < 0) {
Py_DECREF(lineobj);
goto err;
}
pos++;
}
Py_DECREF(lineobj);
if (parse_process_char(self, 0) < 0)
goto err;
} while (self->state != START_RECORD);
fields = self->fields;
self->fields = NULL;
err:
return fields;
}
/* zipimporter.__init__
Split the "subdirectory" from the Zip archive path, lookup a matching
entry in sys.path_importer_cache, fetch the file directory from there
if found, or else read it from the archive. */
static int
zipimporter_init(ZipImporter *self, PyObject *args, PyObject *kwds)
{
PyObject *path, *files, *tmp;
PyObject *filename = NULL;
Py_ssize_t len, flen;
if (!_PyArg_NoKeywords("zipimporter()", kwds))
return -1;
if (!PyArg_ParseTuple(args, "O&:zipimporter",
PyUnicode_FSDecoder, &path))
return -1;
if (PyUnicode_READY(path) == -1)
return -1;
len = PyUnicode_GET_LENGTH(path);
if (len == 0) {
PyErr_SetString(ZipImportError, "archive path is empty");
goto error;
}
#ifdef ALTSEP
tmp = _PyObject_CallMethodId(path, &PyId_replace, "CC", ALTSEP, SEP);
if (!tmp)
goto error;
Py_DECREF(path);
path = tmp;
#endif
filename = path;
Py_INCREF(filename);
flen = len;
for (;;) {
struct stat statbuf;
int rv;
rv = _Py_stat(filename, &statbuf);
if (rv == -2)
goto error;
if (rv == 0) {
/* it exists */
if (!S_ISREG(statbuf.st_mode))
/* it's a not file */
Py_CLEAR(filename);
break;
}
Py_CLEAR(filename);
/* back up one path element */
flen = PyUnicode_FindChar(path, SEP, 0, flen, -1);
if (flen == -1)
break;
filename = PyUnicode_Substring(path, 0, flen);
if (filename == NULL)
goto error;
}
if (filename == NULL) {
PyErr_SetString(ZipImportError, "not a Zip file");
goto error;
}
if (PyUnicode_READY(filename) < 0)
goto error;
files = PyDict_GetItem(zip_directory_cache, filename);
if (files == NULL) {
files = read_directory(filename);
if (files == NULL)
goto error;
if (PyDict_SetItem(zip_directory_cache, filename, files) != 0)
goto error;
}
else
Py_INCREF(files);
self->files = files;
/* Transfer reference */
self->archive = filename;
filename = NULL;
/* Check if there is a prefix directory following the filename. */
if (flen != len) {
tmp = PyUnicode_Substring(path, flen+1,
PyUnicode_GET_LENGTH(path));
if (tmp == NULL)
goto error;
self->prefix = tmp;
if (PyUnicode_READ_CHAR(path, len-1) != SEP) {
/* add trailing SEP */
tmp = PyUnicode_FromFormat("%U%c", self->prefix, SEP);
if (tmp == NULL)
goto error;
Py_SETREF(self->prefix, tmp);
}
}
else
self->prefix = PyUnicode_New(0, 0);
Py_DECREF(path);
return 0;
error:
Py_DECREF(path);
Py_XDECREF(filename);
return -1;
}
static PyObject *
parse(PyObject *self, PyObject *args, PyObject *kwargs)
{
PyObject *string;
if (!PyArg_ParseTuple(args, "U", &string))
return NULL;
if (PyUnicode_READY(string) == -1)
return NULL;
Py_ssize_t string_length = PyUnicode_GET_LENGTH(string);
int string_kind = PyUnicode_KIND(string);
void *string_data = PyUnicode_DATA(string);
Py_UCS4 c;
int i;
Symbol token = {.kind = UNDECIDED, .data=NULL};
int escape = 0;
for (i = 0; i < string_length; i++) {
c = PyUnicode_READ(string_kind, string_data, i);
switch (c) {
case 0x22: // "
if (escape) {
} else {
}
break;
case 0x27: // '
if (escape) {
} else {
}
break;
case 0x30:
case 0x31:
case 0x32:
case 0x33:
case 0x34:
case 0x35:
case 0x36:
case 0x37:
case 0x38:
case 0x39:
switch (token.kind) {
case UNDECIDED:
token.kind = INTEGER;
token.data = c - 0x30;
break;
case INTEGER:
token.data += (c - 0x30);
break;
}
case 0x5c: // backslash
if (escape) {
printf("todo literal backslash\n");
escape = 0;
} else {
escape = 1;
}
break;
CASE__WHITE_SPACE:
switch (token.kind) {
case INTEGER:
case IDENTIFIER:
printf("whitespace new token\n");
break;
}
break;
CASE__XID_START:
switch (token.kind) {
case UNDECIDED:
token.kind = IDENTIFIER;
break;
case INTEGER:
printf("todo\n");
break;
}
break;
CASE__XID_CONTINUE__EXCLUDING__XID_START:
switch (token.kind) {
case UNDECIDED:
printf("continue start?\n");
break;
case INTEGER:
printf("continue integer\n");
break;
}
break;
default:
printf("unclassified");
}
}
PyErr_SetString(PyExc_NotImplementedError, "parse");
return NULL;
}
static int
zipimport_zipimporter___init___impl(ZipImporter *self, PyObject *path)
/*[clinic end generated code: output=141558fefdb46dc8 input=92b9ebeed1f6a704]*/
{
PyObject *files, *tmp;
PyObject *filename = NULL;
Py_ssize_t len, flen;
if (PyUnicode_READY(path) == -1)
return -1;
len = PyUnicode_GET_LENGTH(path);
if (len == 0) {
PyErr_SetString(ZipImportError, "archive path is empty");
goto error;
}
#ifdef ALTSEP
tmp = _PyObject_CallMethodId(path, &PyId_replace, "CC", ALTSEP, SEP);
if (!tmp)
goto error;
Py_DECREF(path);
path = tmp;
#endif
filename = path;
Py_INCREF(filename);
flen = len;
for (;;) {
struct stat statbuf;
int rv;
rv = _Py_stat(filename, &statbuf);
if (rv == -2)
goto error;
if (rv == 0) {
/* it exists */
if (!S_ISREG(statbuf.st_mode))
/* it's a not file */
Py_CLEAR(filename);
break;
}
Py_CLEAR(filename);
/* back up one path element */
flen = PyUnicode_FindChar(path, SEP, 0, flen, -1);
if (flen == -1)
break;
filename = PyUnicode_Substring(path, 0, flen);
if (filename == NULL)
goto error;
}
if (filename == NULL) {
PyErr_SetString(ZipImportError, "not a Zip file");
goto error;
}
if (PyUnicode_READY(filename) < 0)
goto error;
files = PyDict_GetItem(zip_directory_cache, filename);
if (files == NULL) {
files = read_directory(filename);
if (files == NULL)
goto error;
if (PyDict_SetItem(zip_directory_cache, filename, files) != 0)
goto error;
}
else
Py_INCREF(files);
Py_XSETREF(self->files, files);
/* Transfer reference */
Py_XSETREF(self->archive, filename);
filename = NULL;
/* Check if there is a prefix directory following the filename. */
if (flen != len) {
tmp = PyUnicode_Substring(path, flen+1,
PyUnicode_GET_LENGTH(path));
if (tmp == NULL)
goto error;
Py_XSETREF(self->prefix, tmp);
if (PyUnicode_READ_CHAR(path, len-1) != SEP) {
/* add trailing SEP */
tmp = PyUnicode_FromFormat("%U%c", self->prefix, SEP);
if (tmp == NULL)
goto error;
Py_SETREF(self->prefix, tmp);
}
}
else {
Py_XSETREF(self->prefix, PyUnicode_New(0, 0));
}
Py_DECREF(path);
return 0;
error:
Py_DECREF(path);
Py_XDECREF(filename);
return -1;
}
/* Load and return the module named by 'fullname'. */
static PyObject *
zipimporter_load_module(PyObject *obj, PyObject *args)
{
ZipImporter *self = (ZipImporter *)obj;
PyObject *code = NULL, *mod, *dict;
PyObject *fullname;
PyObject *modpath = NULL;
int ispackage;
if (!PyArg_ParseTuple(args, "U:zipimporter.load_module",
&fullname))
return NULL;
if (PyUnicode_READY(fullname) == -1)
return NULL;
code = get_module_code(self, fullname, &ispackage, &modpath);
if (code == NULL)
goto error;
mod = PyImport_AddModuleObject(fullname);
if (mod == NULL)
goto error;
dict = PyModule_GetDict(mod);
/* mod.__loader__ = self */
if (PyDict_SetItemString(dict, "__loader__", (PyObject *)self) != 0)
goto error;
if (ispackage) {
/* add __path__ to the module *before* the code gets
executed */
PyObject *pkgpath, *fullpath, *subname;
int err;
subname = get_subname(fullname);
if (subname == NULL)
goto error;
fullpath = PyUnicode_FromFormat("%U%c%U%U",
self->archive, SEP,
self->prefix, subname);
Py_DECREF(subname);
if (fullpath == NULL)
goto error;
pkgpath = Py_BuildValue("[N]", fullpath);
if (pkgpath == NULL)
goto error;
err = PyDict_SetItemString(dict, "__path__", pkgpath);
Py_DECREF(pkgpath);
if (err != 0)
goto error;
}
mod = PyImport_ExecCodeModuleObject(fullname, code, modpath, NULL);
Py_CLEAR(code);
if (mod == NULL)
goto error;
if (Py_VerboseFlag)
PySys_FormatStderr("import %U # loaded from Zip %U\n",
fullname, modpath);
Py_DECREF(modpath);
return mod;
error:
Py_XDECREF(code);
Py_XDECREF(modpath);
return NULL;
}
PyCodeObject *
PyCode_New(int argcount, int kwonlyargcount,
int nlocals, int stacksize, int flags,
PyObject *code, PyObject *consts, PyObject *names,
PyObject *varnames, PyObject *freevars, PyObject *cellvars,
PyObject *filename, PyObject *name, int firstlineno,
PyObject *lnotab)
{
PyCodeObject *co;
unsigned char *cell2arg = NULL;
Py_ssize_t i, n_cellvars;
// We originally had a Py_GUARD here, and all was well. It was never hit
// by normal parallel contexts. Then, after misusing datrie, it was
// suddenly being hit -- turns out Cython calls PyCode_New() when an
// exception occurs in order to do stuff. I've since perused the body and
// nothing immediately stands out as being unsafe for calling in a
// parallel context (i.e. no random mallocs or static storage), so, let's
// remove the guard and see what happens.
//Py_GUARD();
/* Check argument types */
if (argcount < 0 || kwonlyargcount < 0 || nlocals < 0 ||
code == NULL ||
consts == NULL || !PyTuple_Check(consts) ||
names == NULL || !PyTuple_Check(names) ||
varnames == NULL || !PyTuple_Check(varnames) ||
freevars == NULL || !PyTuple_Check(freevars) ||
cellvars == NULL || !PyTuple_Check(cellvars) ||
name == NULL || !PyUnicode_Check(name) ||
filename == NULL || !PyUnicode_Check(filename) ||
lnotab == NULL || !PyBytes_Check(lnotab) ||
!PyObject_CheckReadBuffer(code)) {
PyErr_BadInternalCall();
return NULL;
}
/* Ensure that the filename is a ready Unicode string */
if (PyUnicode_READY(filename) < 0)
return NULL;
n_cellvars = PyTuple_GET_SIZE(cellvars);
intern_strings(names);
intern_strings(varnames);
intern_strings(freevars);
intern_strings(cellvars);
/* Intern selected string constants */
for (i = PyTuple_GET_SIZE(consts); --i >= 0; ) {
PyObject *v = PyTuple_GetItem(consts, i);
if (!all_name_chars(v))
continue;
PyUnicode_InternInPlace(&PyTuple_GET_ITEM(consts, i));
}
/* Create mapping between cells and arguments if needed. */
if (n_cellvars) {
Py_ssize_t total_args = argcount + kwonlyargcount +
((flags & CO_VARARGS) != 0) + ((flags & CO_VARKEYWORDS) != 0);
Py_ssize_t alloc_size = sizeof(unsigned char) * n_cellvars;
int used_cell2arg = 0;
cell2arg = PyMem_MALLOC(alloc_size);
if (cell2arg == NULL)
return NULL;
memset(cell2arg, CO_CELL_NOT_AN_ARG, alloc_size);
/* Find cells which are also arguments. */
for (i = 0; i < n_cellvars; i++) {
Py_ssize_t j;
PyObject *cell = PyTuple_GET_ITEM(cellvars, i);
for (j = 0; j < total_args; j++) {
PyObject *arg = PyTuple_GET_ITEM(varnames, j);
if (!PyUnicode_Compare(cell, arg)) {
cell2arg[i] = j;
used_cell2arg = 1;
break;
}
}
}
if (!used_cell2arg) {
PyMem_FREE(cell2arg);
cell2arg = NULL;
}
}
co = PyObject_NEW(PyCodeObject, &PyCode_Type);
if (co == NULL) {
if (cell2arg)
PyMem_FREE(cell2arg);
return NULL;
}
co->co_argcount = argcount;
co->co_kwonlyargcount = kwonlyargcount;
co->co_nlocals = nlocals;
co->co_stacksize = stacksize;
co->co_flags = flags;
Py_INCREF(code);
co->co_code = code;
Py_INCREF(consts);
co->co_consts = consts;
Py_INCREF(names);
co->co_names = names;
Py_INCREF(varnames);
co->co_varnames = varnames;
Py_INCREF(freevars);
co->co_freevars = freevars;
Py_INCREF(cellvars);
co->co_cellvars = cellvars;
co->co_cell2arg = cell2arg;
Py_INCREF(filename);
co->co_filename = filename;
Py_INCREF(name);
co->co_name = name;
co->co_firstlineno = firstlineno;
Py_INCREF(lnotab);
co->co_lnotab = lnotab;
co->co_zombieframe = NULL;
co->co_weakreflist = NULL;
return co;
}
PyCodeObject *
PyCode_New(int argcount, int kwonlyargcount,
int nlocals, int stacksize, int flags,
PyObject *code, PyObject *consts, PyObject *names,
PyObject *varnames, PyObject *freevars, PyObject *cellvars,
PyObject *filename, PyObject *name, int firstlineno,
PyObject *lnotab)
{
PyCodeObject *co;
Py_ssize_t *cell2arg = NULL;
Py_ssize_t i, n_cellvars, n_varnames, total_args;
/* Check argument types */
if (argcount < 0 || kwonlyargcount < 0 || nlocals < 0 ||
code == NULL || !PyBytes_Check(code) ||
consts == NULL || !PyTuple_Check(consts) ||
names == NULL || !PyTuple_Check(names) ||
varnames == NULL || !PyTuple_Check(varnames) ||
freevars == NULL || !PyTuple_Check(freevars) ||
cellvars == NULL || !PyTuple_Check(cellvars) ||
name == NULL || !PyUnicode_Check(name) ||
filename == NULL || !PyUnicode_Check(filename) ||
lnotab == NULL || !PyBytes_Check(lnotab)) {
PyErr_BadInternalCall();
return NULL;
}
/* Ensure that the filename is a ready Unicode string */
if (PyUnicode_READY(filename) < 0)
return NULL;
intern_strings(names);
intern_strings(varnames);
intern_strings(freevars);
intern_strings(cellvars);
intern_string_constants(consts);
/* Check for any inner or outer closure references */
n_cellvars = PyTuple_GET_SIZE(cellvars);
if (!n_cellvars && !PyTuple_GET_SIZE(freevars)) {
flags |= CO_NOFREE;
} else {
flags &= ~CO_NOFREE;
}
n_varnames = PyTuple_GET_SIZE(varnames);
if (argcount <= n_varnames && kwonlyargcount <= n_varnames) {
/* Never overflows. */
total_args = (Py_ssize_t)argcount + (Py_ssize_t)kwonlyargcount +
((flags & CO_VARARGS) != 0) + ((flags & CO_VARKEYWORDS) != 0);
}
else {
total_args = n_varnames + 1;
}
if (total_args > n_varnames) {
PyErr_SetString(PyExc_ValueError, "code: varnames is too small");
return NULL;
}
/* Create mapping between cells and arguments if needed. */
if (n_cellvars) {
bool used_cell2arg = false;
cell2arg = PyMem_NEW(Py_ssize_t, n_cellvars);
if (cell2arg == NULL) {
PyErr_NoMemory();
return NULL;
}
/* Find cells which are also arguments. */
for (i = 0; i < n_cellvars; i++) {
Py_ssize_t j;
PyObject *cell = PyTuple_GET_ITEM(cellvars, i);
cell2arg[i] = CO_CELL_NOT_AN_ARG;
for (j = 0; j < total_args; j++) {
PyObject *arg = PyTuple_GET_ITEM(varnames, j);
int cmp = PyUnicode_Compare(cell, arg);
if (cmp == -1 && PyErr_Occurred()) {
PyMem_FREE(cell2arg);
return NULL;
}
if (cmp == 0) {
cell2arg[i] = j;
used_cell2arg = true;
break;
}
}
}
if (!used_cell2arg) {
PyMem_FREE(cell2arg);
cell2arg = NULL;
}
}
co = PyObject_NEW(PyCodeObject, &PyCode_Type);
if (co == NULL) {
if (cell2arg)
PyMem_FREE(cell2arg);
return NULL;
}
co->co_argcount = argcount;
co->co_kwonlyargcount = kwonlyargcount;
co->co_nlocals = nlocals;
co->co_stacksize = stacksize;
co->co_flags = flags;
Py_INCREF(code);
co->co_code = code;
Py_INCREF(consts);
co->co_consts = consts;
Py_INCREF(names);
co->co_names = names;
Py_INCREF(varnames);
co->co_varnames = varnames;
Py_INCREF(freevars);
co->co_freevars = freevars;
Py_INCREF(cellvars);
co->co_cellvars = cellvars;
co->co_cell2arg = cell2arg;
Py_INCREF(filename);
co->co_filename = filename;
Py_INCREF(name);
co->co_name = name;
co->co_firstlineno = firstlineno;
Py_INCREF(lnotab);
co->co_lnotab = lnotab;
co->co_zombieframe = NULL;
co->co_weakreflist = NULL;
co->co_extra = NULL;
return co;
}
static CYTHON_INLINE char* __Pyx_PyObject_AsStringAndSize(PyObject* o, Py_ssize_t *length) {
#if __PYX_DEFAULT_STRING_ENCODING_IS_ASCII || __PYX_DEFAULT_STRING_ENCODING_IS_DEFAULT
if (
#if PY_MAJOR_VERSION < 3 && __PYX_DEFAULT_STRING_ENCODING_IS_ASCII
__Pyx_sys_getdefaultencoding_not_ascii &&
#endif
PyUnicode_Check(o)) {
#if PY_VERSION_HEX < 0x03030000
char* defenc_c;
// borrowed, cached reference
PyObject* defenc = _PyUnicode_AsDefaultEncodedString(o, NULL);
if (!defenc) return NULL;
defenc_c = PyBytes_AS_STRING(defenc);
#if __PYX_DEFAULT_STRING_ENCODING_IS_ASCII
{
char* end = defenc_c + PyBytes_GET_SIZE(defenc);
char* c;
for (c = defenc_c; c < end; c++) {
if ((unsigned char) (*c) >= 128) {
// raise the error
PyUnicode_AsASCIIString(o);
return NULL;
}
}
}
#endif /*__PYX_DEFAULT_STRING_ENCODING_IS_ASCII*/
*length = PyBytes_GET_SIZE(defenc);
return defenc_c;
#else /* PY_VERSION_HEX < 0x03030000 */
if (PyUnicode_READY(o) == -1) return NULL;
#if __PYX_DEFAULT_STRING_ENCODING_IS_ASCII
if (PyUnicode_IS_ASCII(o)) {
// cached for the lifetime of the object
*length = PyUnicode_GET_DATA_SIZE(o);
return PyUnicode_AsUTF8(o);
} else {
// raise the error
PyUnicode_AsASCIIString(o);
return NULL;
}
#else /* __PYX_DEFAULT_STRING_ENCODING_IS_ASCII */
return PyUnicode_AsUTF8AndSize(o, length);
#endif /* __PYX_DEFAULT_STRING_ENCODING_IS_ASCII */
#endif /* PY_VERSION_HEX < 0x03030000 */
} else
#endif /* __PYX_DEFAULT_STRING_ENCODING_IS_ASCII || __PYX_DEFAULT_STRING_ENCODING_IS_DEFAULT */
#if PY_VERSION_HEX >= 0x02060000
if (PyByteArray_Check(o)) {
*length = PyByteArray_GET_SIZE(o);
return PyByteArray_AS_STRING(o);
} else
#endif
{
char* result;
int r = PyBytes_AsStringAndSize(o, &result, length);
if (unlikely(r < 0)) {
return NULL;
} else {
return result;
}
}
}
static void
show_warning(PyObject *filename, int lineno, PyObject *text,
PyObject *category, PyObject *sourceline)
{
PyObject *f_stderr;
PyObject *name;
char lineno_str[128];
_Py_IDENTIFIER(__name__);
PyOS_snprintf(lineno_str, sizeof(lineno_str), ":%d: ", lineno);
name = _PyObject_GetAttrId(category, &PyId___name__);
if (name == NULL) /* XXX Can an object lack a '__name__' attribute? */
goto error;
f_stderr = _PySys_GetObjectId(&PyId_stderr);
if (f_stderr == NULL) {
fprintf(stderr, "lost sys.stderr\n");
goto error;
}
/* Print "filename:lineno: category: text\n" */
if (PyFile_WriteObject(filename, f_stderr, Py_PRINT_RAW) < 0)
goto error;
if (PyFile_WriteString(lineno_str, f_stderr) < 0)
goto error;
if (PyFile_WriteObject(name, f_stderr, Py_PRINT_RAW) < 0)
goto error;
if (PyFile_WriteString(": ", f_stderr) < 0)
goto error;
if (PyFile_WriteObject(text, f_stderr, Py_PRINT_RAW) < 0)
goto error;
if (PyFile_WriteString("\n", f_stderr) < 0)
goto error;
Py_CLEAR(name);
/* Print " source_line\n" */
if (sourceline) {
int kind;
void *data;
Py_ssize_t i, len;
Py_UCS4 ch;
PyObject *truncated;
if (PyUnicode_READY(sourceline) < 1)
goto error;
kind = PyUnicode_KIND(sourceline);
data = PyUnicode_DATA(sourceline);
len = PyUnicode_GET_LENGTH(sourceline);
for (i=0; i<len; i++) {
ch = PyUnicode_READ(kind, data, i);
if (ch != ' ' && ch != '\t' && ch != '\014')
break;
}
truncated = PyUnicode_Substring(sourceline, i, len);
if (truncated == NULL)
goto error;
PyFile_WriteObject(sourceline, f_stderr, Py_PRINT_RAW);
Py_DECREF(truncated);
PyFile_WriteString("\n", f_stderr);
}
else {
_Py_DisplaySourceLine(f_stderr, filename, lineno, 2);
}
error:
Py_XDECREF(name);
PyErr_Clear();
}
/*
read_directory(archive) -> files dict (new reference)
Given a path to a Zip archive, build a dict, mapping file names
(local to the archive, using SEP as a separator) to toc entries.
A toc_entry is a tuple:
(__file__, # value to use for __file__, available for all files,
# encoded to the filesystem encoding
compress, # compression kind; 0 for uncompressed
data_size, # size of compressed data on disk
file_size, # size of decompressed data
file_offset, # offset of file header from start of archive
time, # mod time of file (in dos format)
date, # mod data of file (in dos format)
crc, # crc checksum of the data
)
Directories can be recognized by the trailing SEP in the name,
data_size and file_offset are 0.
*/
static PyObject *
read_directory(PyObject *archive)
{
PyObject *files = NULL;
FILE *fp;
unsigned short flags, compress, time, date, name_size;
unsigned int crc, data_size, file_size, header_size, header_offset;
unsigned long file_offset, header_position;
unsigned long arc_offset; /* Absolute offset to start of the zip-archive. */
unsigned int count, i;
unsigned char buffer[46];
char name[MAXPATHLEN + 5];
PyObject *nameobj = NULL;
PyObject *path;
const char *charset;
int bootstrap;
const char *errmsg = NULL;
fp = _Py_fopen_obj(archive, "rb");
if (fp == NULL) {
if (PyErr_ExceptionMatches(PyExc_OSError)) {
_PyErr_FormatFromCause(ZipImportError,
"can't open Zip file: %R", archive);
}
return NULL;
}
if (fseek(fp, -22, SEEK_END) == -1) {
goto file_error;
}
header_position = (unsigned long)ftell(fp);
if (header_position == (unsigned long)-1) {
goto file_error;
}
assert(header_position <= (unsigned long)LONG_MAX);
if (fread(buffer, 1, 22, fp) != 22) {
goto file_error;
}
if (get_uint32(buffer) != 0x06054B50u) {
/* Bad: End of Central Dir signature */
errmsg = "not a Zip file";
goto invalid_header;
}
header_size = get_uint32(buffer + 12);
header_offset = get_uint32(buffer + 16);
if (header_position < header_size) {
errmsg = "bad central directory size";
goto invalid_header;
}
if (header_position < header_offset) {
errmsg = "bad central directory offset";
goto invalid_header;
}
if (header_position - header_size < header_offset) {
errmsg = "bad central directory size or offset";
goto invalid_header;
}
header_position -= header_size;
arc_offset = header_position - header_offset;
files = PyDict_New();
if (files == NULL) {
goto error;
}
/* Start of Central Directory */
count = 0;
if (fseek(fp, (long)header_position, 0) == -1) {
goto file_error;
}
for (;;) {
PyObject *t;
size_t n;
int err;
n = fread(buffer, 1, 46, fp);
if (n < 4) {
goto eof_error;
}
/* Start of file header */
if (get_uint32(buffer) != 0x02014B50u) {
break; /* Bad: Central Dir File Header */
}
if (n != 46) {
goto eof_error;
}
flags = get_uint16(buffer + 8);
compress = get_uint16(buffer + 10);
time = get_uint16(buffer + 12);
date = get_uint16(buffer + 14);
crc = get_uint32(buffer + 16);
data_size = get_uint32(buffer + 20);
file_size = get_uint32(buffer + 24);
name_size = get_uint16(buffer + 28);
header_size = (unsigned int)name_size +
get_uint16(buffer + 30) /* extra field */ +
get_uint16(buffer + 32) /* comment */;
file_offset = get_uint32(buffer + 42);
if (file_offset > header_offset) {
errmsg = "bad local header offset";
goto invalid_header;
}
file_offset += arc_offset;
if (name_size > MAXPATHLEN) {
name_size = MAXPATHLEN;
}
if (fread(name, 1, name_size, fp) != name_size) {
goto file_error;
}
name[name_size] = '\0'; /* Add terminating null byte */
#if SEP != '/'
for (i = 0; i < name_size; i++) {
if (name[i] == '/') {
name[i] = SEP;
}
}
#endif
/* Skip the rest of the header.
* On Windows, calling fseek to skip over the fields we don't use is
* slower than reading the data because fseek flushes stdio's
* internal buffers. See issue #8745. */
assert(header_size <= 3*0xFFFFu);
for (i = name_size; i < header_size; i++) {
if (getc(fp) == EOF) {
goto file_error;
}
}
bootstrap = 0;
if (flags & 0x0800) {
charset = "utf-8";
}
else if (!PyThreadState_GET()->interp->codecs_initialized) {
/* During bootstrap, we may need to load the encodings
package from a ZIP file. But the cp437 encoding is implemented
in Python in the encodings package.
Break out of this dependency by assuming that the path to
the encodings module is ASCII-only. */
charset = "ascii";
bootstrap = 1;
}
else {
charset = "cp437";
}
nameobj = PyUnicode_Decode(name, name_size, charset, NULL);
if (nameobj == NULL) {
if (bootstrap) {
PyErr_Format(PyExc_NotImplementedError,
"bootstrap issue: python%i%i.zip contains non-ASCII "
"filenames without the unicode flag",
PY_MAJOR_VERSION, PY_MINOR_VERSION);
}
goto error;
}
if (PyUnicode_READY(nameobj) == -1) {
goto error;
}
path = PyUnicode_FromFormat("%U%c%U", archive, SEP, nameobj);
if (path == NULL) {
goto error;
}
t = Py_BuildValue("NHIIkHHI", path, compress, data_size,
file_size, file_offset, time, date, crc);
if (t == NULL) {
goto error;
}
err = PyDict_SetItem(files, nameobj, t);
Py_CLEAR(nameobj);
Py_DECREF(t);
if (err != 0) {
goto error;
}
count++;
}
fclose(fp);
if (Py_VerboseFlag) {
PySys_FormatStderr("# zipimport: found %u names in %R\n",
count, archive);
}
return files;
eof_error:
set_file_error(archive, !ferror(fp));
goto error;
file_error:
PyErr_Format(ZipImportError, "can't read Zip file: %R", archive);
goto error;
invalid_header:
assert(errmsg != NULL);
PyErr_Format(ZipImportError, "%s: %R", errmsg, archive);
goto error;
error:
fclose(fp);
Py_XDECREF(files);
Py_XDECREF(nameobj);
return NULL;
}
/* Returns 0 on error (no new refs), 1 on success */
static int
setup_context(Py_ssize_t stack_level, PyObject **filename, int *lineno,
PyObject **module, PyObject **registry)
{
PyObject *globals;
/* Setup globals and lineno. */
PyFrameObject *f = PyThreadState_GET()->frame;
// Stack level comparisons to Python code is off by one as there is no
// warnings-related stack level to avoid.
if (stack_level <= 0 || is_internal_frame(f)) {
while (--stack_level > 0 && f != NULL) {
f = f->f_back;
}
}
else {
while (--stack_level > 0 && f != NULL) {
f = next_external_frame(f);
}
}
if (f == NULL) {
globals = PyThreadState_Get()->interp->sysdict;
*lineno = 1;
}
else {
globals = f->f_globals;
*lineno = PyFrame_GetLineNumber(f);
}
*module = NULL;
/* Setup registry. */
assert(globals != NULL);
assert(PyDict_Check(globals));
*registry = PyDict_GetItemString(globals, "__warningregistry__");
if (*registry == NULL) {
int rc;
*registry = PyDict_New();
if (*registry == NULL)
return 0;
rc = PyDict_SetItemString(globals, "__warningregistry__", *registry);
if (rc < 0)
goto handle_error;
}
else
Py_INCREF(*registry);
/* Setup module. */
*module = PyDict_GetItemString(globals, "__name__");
if (*module == NULL) {
*module = PyUnicode_FromString("<string>");
if (*module == NULL)
goto handle_error;
}
else
Py_INCREF(*module);
/* Setup filename. */
*filename = PyDict_GetItemString(globals, "__file__");
if (*filename != NULL && PyUnicode_Check(*filename)) {
Py_ssize_t len;
int kind;
void *data;
if (PyUnicode_READY(*filename))
goto handle_error;
len = PyUnicode_GetLength(*filename);
kind = PyUnicode_KIND(*filename);
data = PyUnicode_DATA(*filename);
#define ascii_lower(c) ((c <= 127) ? Py_TOLOWER(c) : 0)
/* if filename.lower().endswith(".pyc"): */
if (len >= 4 &&
PyUnicode_READ(kind, data, len-4) == '.' &&
ascii_lower(PyUnicode_READ(kind, data, len-3)) == 'p' &&
ascii_lower(PyUnicode_READ(kind, data, len-2)) == 'y' &&
ascii_lower(PyUnicode_READ(kind, data, len-1)) == 'c')
{
*filename = PyUnicode_Substring(*filename, 0,
PyUnicode_GET_LENGTH(*filename)-1);
if (*filename == NULL)
goto handle_error;
}
else
Py_INCREF(*filename);
}
else {
*filename = NULL;
if (*module != Py_None && PyUnicode_CompareWithASCIIString(*module, "__main__") == 0) {
PyObject *argv = _PySys_GetObjectId(&PyId_argv);
/* PyList_Check() is needed because sys.argv is set to None during
Python finalization */
if (argv != NULL && PyList_Check(argv) && PyList_Size(argv) > 0) {
int is_true;
*filename = PyList_GetItem(argv, 0);
Py_INCREF(*filename);
/* If sys.argv[0] is false, then use '__main__'. */
is_true = PyObject_IsTrue(*filename);
if (is_true < 0) {
Py_DECREF(*filename);
goto handle_error;
}
else if (!is_true) {
Py_XSETREF(*filename, PyUnicode_FromString("__main__"));
if (*filename == NULL)
goto handle_error;
}
}
else {
/* embedded interpreters don't have sys.argv, see bug #839151 */
*filename = PyUnicode_FromString("__main__");
if (*filename == NULL)
goto handle_error;
}
}
if (*filename == NULL) {
*filename = *module;
Py_INCREF(*filename);
}
}
return 1;
handle_error:
/* filename not XDECREF'ed here as there is no way to jump here with a
dangling reference. */
Py_XDECREF(*registry);
Py_XDECREF(*module);
return 0;
}
/*
read_directory(archive) -> files dict (new reference)
Given a path to a Zip archive, build a dict, mapping file names
(local to the archive, using SEP as a separator) to toc entries.
A toc_entry is a tuple:
(__file__, # value to use for __file__, available for all files,
# encoded to the filesystem encoding
compress, # compression kind; 0 for uncompressed
data_size, # size of compressed data on disk
file_size, # size of decompressed data
file_offset, # offset of file header from start of archive
time, # mod time of file (in dos format)
date, # mod data of file (in dos format)
crc, # crc checksum of the data
)
Directories can be recognized by the trailing SEP in the name,
data_size and file_offset are 0.
*/
static PyObject *
read_directory(PyObject *archive)
{
PyObject *files = NULL;
FILE *fp;
unsigned short flags;
short compress, time, date, name_size;
long crc, data_size, file_size, header_size;
Py_ssize_t file_offset, header_position, header_offset;
long l, count;
Py_ssize_t i;
char name[MAXPATHLEN + 5];
PyObject *nameobj = NULL;
char *p, endof_central_dir[22];
Py_ssize_t arc_offset; /* Absolute offset to start of the zip-archive. */
PyObject *path;
const char *charset;
int bootstrap;
fp = _Py_fopen(archive, "rb");
if (fp == NULL) {
if (!PyErr_Occurred())
PyErr_Format(ZipImportError, "can't open Zip file: %R", archive);
return NULL;
}
fseek(fp, -22, SEEK_END);
header_position = ftell(fp);
if (fread(endof_central_dir, 1, 22, fp) != 22) {
fclose(fp);
PyErr_Format(ZipImportError, "can't read Zip file: %R", archive);
return NULL;
}
if (get_long((unsigned char *)endof_central_dir) != 0x06054B50) {
/* Bad: End of Central Dir signature */
fclose(fp);
PyErr_Format(ZipImportError, "not a Zip file: %R", archive);
return NULL;
}
header_size = get_long((unsigned char *)endof_central_dir + 12);
header_offset = get_long((unsigned char *)endof_central_dir + 16);
arc_offset = header_position - header_offset - header_size;
header_offset += arc_offset;
files = PyDict_New();
if (files == NULL)
goto error;
/* Start of Central Directory */
count = 0;
for (;;) {
PyObject *t;
int err;
fseek(fp, header_offset, 0); /* Start of file header */
l = PyMarshal_ReadLongFromFile(fp);
if (l != 0x02014B50)
break; /* Bad: Central Dir File Header */
fseek(fp, header_offset + 8, 0);
flags = (unsigned short)PyMarshal_ReadShortFromFile(fp);
compress = PyMarshal_ReadShortFromFile(fp);
time = PyMarshal_ReadShortFromFile(fp);
date = PyMarshal_ReadShortFromFile(fp);
crc = PyMarshal_ReadLongFromFile(fp);
data_size = PyMarshal_ReadLongFromFile(fp);
file_size = PyMarshal_ReadLongFromFile(fp);
name_size = PyMarshal_ReadShortFromFile(fp);
header_size = 46 + name_size +
PyMarshal_ReadShortFromFile(fp) +
PyMarshal_ReadShortFromFile(fp);
fseek(fp, header_offset + 42, 0);
file_offset = PyMarshal_ReadLongFromFile(fp) + arc_offset;
if (name_size > MAXPATHLEN)
name_size = MAXPATHLEN;
p = name;
for (i = 0; i < (Py_ssize_t)name_size; i++) {
*p = (char)getc(fp);
if (*p == '/')
*p = SEP;
p++;
}
*p = 0; /* Add terminating null byte */
header_offset += header_size;
bootstrap = 0;
if (flags & 0x0800)
charset = "utf-8";
else if (!PyThreadState_GET()->interp->codecs_initialized) {
/* During bootstrap, we may need to load the encodings
package from a ZIP file. But the cp437 encoding is implemented
in Python in the encodings package.
Break out of this dependency by assuming that the path to
the encodings module is ASCII-only. */
charset = "ascii";
bootstrap = 1;
}
else
charset = "cp437";
nameobj = PyUnicode_Decode(name, name_size, charset, NULL);
if (PyUnicode_READY(nameobj) == -1)
goto error;
if (nameobj == NULL) {
if (bootstrap)
PyErr_Format(PyExc_NotImplementedError,
"bootstrap issue: python%i%i.zip contains non-ASCII "
"filenames without the unicode flag",
PY_MAJOR_VERSION, PY_MINOR_VERSION);
goto error;
}
path = PyUnicode_FromFormat("%U%c%U", archive, SEP, nameobj);
if (path == NULL)
goto error;
t = Py_BuildValue("Nhllnhhl", path, compress, data_size,
file_size, file_offset, time, date, crc);
if (t == NULL)
goto error;
err = PyDict_SetItem(files, nameobj, t);
Py_CLEAR(nameobj);
Py_DECREF(t);
if (err != 0)
goto error;
count++;
}
fclose(fp);
if (Py_VerboseFlag)
PySys_FormatStderr("# zipimport: found %ld names in %R\n",
count, archive);
return files;
error:
fclose(fp);
Py_XDECREF(files);
Py_XDECREF(nameobj);
return NULL;
}
