Avoid AddTraceback() if stringtab isn't set up (GH-4378)

This can happen (rarely) with exceptions that occur very early in the module init process.

Fixes https://github.com/cython/cython/issues/4377

Uses a fake Numpy module for testing to make a version of "import_array" that always fails.

3 files changed, 77 insertions, 25 deletions

diff --git a/Cython/Compiler/Code.py b/Cython/Compiler/Code.py
index a2affbe89..8f0af0ce2 100644
--- a/Cython/Compiler/Code.py
+++ b/Cython/Compiler/Code.py
@@ -1138,7 +1138,8 @@ class GlobalState(object):
         'pystring_table',
         'cached_builtins',
         'cached_constants',
-        'init_globals',
+        'init_constants',
+        'init_globals',  # (utility code called at init-time)
         'init_module',
         'cleanup_globals',
         'cleanup_module',
@@ -1209,6 +1210,11 @@ class GlobalState(object):
         w.putln("")
         w.putln("static CYTHON_SMALL_CODE int __Pyx_InitGlobals(void) {")
 
+        w = self.parts['init_constants']
+        w.enter_cfunc_scope()
+        w.putln("")
+        w.putln("static CYTHON_SMALL_CODE int __Pyx_InitConstants(void) {")
+
         if not Options.generate_cleanup_code:
             del self.parts['cleanup_globals']
         else:
@@ -1284,13 +1290,14 @@ class GlobalState(object):
         w.putln("}")
         w.exit_cfunc_scope()
 
-        w = self.parts['init_globals']
-        w.putln("return 0;")
-        if w.label_used(w.error_label):
-            w.put_label(w.error_label)
-            w.putln("return -1;")
-        w.putln("}")
-        w.exit_cfunc_scope()
+        for part in ['init_globals', 'init_constants']:
+            w = self.parts[part]
+            w.putln("return 0;")
+            if w.label_used(w.error_label):
+                w.put_label(w.error_label)
+                w.putln("return -1;")
+            w.putln("}")
+            w.exit_cfunc_scope()
 
         if Options.generate_cleanup_code:
             w = self.parts['cleanup_globals']
@@ -1510,7 +1517,7 @@ class GlobalState(object):
             return
 
         decl = self.parts['decls']
-        init = self.parts['init_globals']
+        init = self.parts['init_constants']
         cnames = []
         for (type_cname, method_name), cname in sorted(self.cached_cmethods.items()):
             cnames.append(cname)
@@ -1560,7 +1567,7 @@ class GlobalState(object):
                 decls_writer.putln("static Py_UNICODE %s[] = { %s };" % (cname, utf16_array))
                 decls_writer.putln("#endif")
 
-        init_globals = self.parts['init_globals']
+        init_constants = self.parts['init_constants']
         if py_strings:
             self.use_utility_code(UtilityCode.load_cached("InitStrings", "StringTools.c"))
             py_strings.sort()
@@ -1575,9 +1582,9 @@ class GlobalState(object):
             decls_writer.putln("#if !CYTHON_USE_MODULE_STATE")
             not_limited_api_decls_writer = decls_writer.insertion_point()
             decls_writer.putln("#endif")
-            init_globals.putln("#if CYTHON_USE_MODULE_STATE")
-            init_globals_in_module_state = init_globals.insertion_point()
-            init_globals.putln("#endif")
+            init_constants.putln("#if CYTHON_USE_MODULE_STATE")
+            init_constants_in_module_state = init_constants.insertion_point()
+            init_constants.putln("#endif")
             for idx, py_string_args in enumerate(py_strings):
                 c_cname, _, py_string = py_string_args
                 if not py_string.is_str or not py_string.encoding or \
@@ -1627,20 +1634,20 @@ class GlobalState(object):
                     py_string.is_str,
                     py_string.intern
                     ))
-                init_globals_in_module_state.putln("if (__Pyx_InitString(%s[%d], &%s) < 0) %s;" % (
+                init_constants_in_module_state.putln("if (__Pyx_InitString(%s[%d], &%s) < 0) %s;" % (
                     Naming.stringtab_cname,
                     idx,
                     py_string.cname,
-                    init_globals.error_goto(self.module_pos)))
+                    init_constants.error_goto(self.module_pos)))
             w.putln("{0, 0, 0, 0, 0, 0, 0}")
             w.putln("};")
 
-            init_globals.putln("#if !CYTHON_USE_MODULE_STATE")
-            init_globals.putln(
+            init_constants.putln("#if !CYTHON_USE_MODULE_STATE")
+            init_constants.putln(
                 "if (__Pyx_InitStrings(%s) < 0) %s;" % (
                     Naming.stringtab_cname,
-                    init_globals.error_goto(self.module_pos)))
-            init_globals.putln("#endif")
+                    init_constants.error_goto(self.module_pos)))
+            init_constants.putln("#endif")
 
     def generate_num_constants(self):
         consts = [(c.py_type, c.value[0] == '-', len(c.value), c.value, c.value_code, c)
@@ -1648,7 +1655,7 @@ class GlobalState(object):
         consts.sort()
         decls_writer = self.parts['decls']
         decls_writer.putln("#if !CYTHON_USE_MODULE_STATE")
-        init_globals = self.parts['init_globals']
+        init_constants = self.parts['init_constants']
         for py_type, _, _, value, value_code, c in consts:
             cname = c.cname
             self.parts['module_state'].putln("PyObject *%s;" % cname)
@@ -1669,9 +1676,9 @@ class GlobalState(object):
                 function = "PyInt_FromLong(%sL)"
             else:
                 function = "PyInt_FromLong(%s)"
-            init_globals.putln('%s = %s; %s' % (
+            init_constants.putln('%s = %s; %s' % (
                 cname, function % value_code,
-                init_globals.error_goto_if_null(cname, self.module_pos)))
+                init_constants.error_goto_if_null(cname, self.module_pos)))
         decls_writer.putln("#endif")
 
     # The functions below are there in a transition phase only
diff --git a/Cython/Compiler/ModuleNode.py b/Cython/Compiler/ModuleNode.py
index d0687624f..99b2f0028 100644
--- a/Cython/Compiler/ModuleNode.py
+++ b/Cython/Compiler/ModuleNode.py
@@ -2939,7 +2939,7 @@ class ModuleNode(Nodes.Node, Nodes.BlockNode):
 
         # start of module init/exec function (pre/post PEP 489)
         code.putln("{")
-
+        code.putln('int stringtab_initialized = 0;')
         tempdecl_code = code.insertion_point()
 
         profile = code.globalstate.directives['profile']
@@ -3012,7 +3012,10 @@ class ModuleNode(Nodes.Node, Nodes.BlockNode):
         code.putln("#endif")
 
         code.putln("/*--- Initialize various global constants etc. ---*/")
-        code.put_error_if_neg(self.pos, "__Pyx_InitGlobals()")
+        code.put_error_if_neg(self.pos, "__Pyx_InitConstants()")
+        code.putln("stringtab_initialized = 1;")
+        code.put_error_if_neg(self.pos, "__Pyx_InitGlobals()")  # calls any utility code
+
 
         code.putln("#if PY_MAJOR_VERSION < 3 && (__PYX_DEFAULT_STRING_ENCODING_IS_ASCII || "
                    "__PYX_DEFAULT_STRING_ENCODING_IS_DEFAULT)")
@@ -3095,7 +3098,9 @@ class ModuleNode(Nodes.Node, Nodes.BlockNode):
         for cname, type in code.funcstate.all_managed_temps():
             code.put_xdecref(cname, type)
         code.putln('if (%s) {' % env.module_cname)
-        code.putln('if (%s) {' % env.module_dict_cname)
+        code.putln('if (%s && stringtab_initialized) {' % env.module_dict_cname)
+        # We can run into errors before the module or stringtab are initialized.
+        # In this case it is not safe to add a traceback (because it uses the stringtab)
         code.put_add_traceback(EncodedString("init %s" % env.qualified_name))
         code.globalstate.use_utility_code(Nodes.traceback_utility_code)
         # Module reference and module dict are in global variables which might still be needed
diff --git a/tests/run/numpy_import_array_error.srctree b/tests/run/numpy_import_array_error.srctree
new file mode 100644
index 000000000..9f2c286f1
--- /dev/null
+++ b/tests/run/numpy_import_array_error.srctree
@@ -0,0 +1,40 @@
+PYTHON setup.py build_ext -i
+PYTHON main.py
+
+############# setup.py ############
+
+from distutils.core import setup
+from Cython.Build import cythonize
+
+setup(ext_modules = cythonize('cimport_numpy.pyx'))
+
+############# numpy.pxd ############
+
+# A fake Numpy module. This defines a version of _import_array
+# that always fails. The Cython-generated call to _import_array
+# happens quite early (before the stringtab is initialized)
+# and thus the error itself handling could cause a segmentation fault
+# https://github.com/cython/cython/issues/4377
+
+cdef extern from *:
+    """
+    #define NPY_NDARRAYOBJECT_H
+    static int _import_array(void) {
+        PyErr_SetString(PyExc_ValueError, "Oh no!");
+        return -1;
+    }
+    """
+    int _import_array() except -1
+    
+############# cimport_numpy.pyx ###########
+
+cimport numpy
+
+############# main.py ####################
+
+try:
+    import cimport_numpy
+except ImportError as e:
+    print(e)
+else:
+    assert(False)