大师兄的Python源码学习笔记(三十三）: 运行环境初始化(五)
大师兄的Python源码学习笔记(三十五）: 模块的动态加载机制（二）

• 前面的章节聚焦在一个模块内，但现实程序中通常有多个模块。
• 因此模块间的引用和交互也是程序中的重要组成部分。

一、从import案例开始

• 创建一个import案例，并查看对应的指令字节码：

demo.py 

import os

  1           0 LOAD_CONST               0 (0)
              2 LOAD_CONST               1 (None)
              4 IMPORT_NAME              0 (os)
              6 STORE_NAME               0 (os)
              8 LOAD_CONST               1 (None)
             10 RETURN_VALUE

• LOAD_CONST和STORE_NAME指令已经很熟悉了，可以看出与import对应的是字节码指令IMPORT_NAME。
• IMPORT_NAME在虚拟机中对应的代码如下：

Python\ceval.c

    TARGET(IMPORT_NAME) {
            PyObject *name = GETITEM(names, oparg);
            PyObject *fromlist = POP();
            PyObject *level = TOP();
            PyObject *res;
            res = import_name(f, name, fromlist, level);
            Py_DECREF(level);
            Py_DECREF(fromlist);
            SET_TOP(res);
            if (res == NULL)
                goto error;
            DISPATCH();
        }

• 这里的核心是调用import_name函数，并传入了f、name、from_list和level四个参数：

• f：帧栈(FrameObject)。
• name：一个PyUnicodeObject对象，表示调用模块的名字。
• from_list: 为None，对应2 LOAD_CONST 1 (None)。
• level: 为0，对应0 LOAD_CONST 0 (0)。

Python\ceval.c

static PyObject *
import_name(PyFrameObject *f, PyObject *name, PyObject *fromlist, PyObject *level)
{
    _Py_IDENTIFIER(__import__);
    PyObject *import_func, *res;
    PyObject* stack[5];

    import_func = _PyDict_GetItemId(f->f_builtins, &PyId___import__);
    if (import_func == NULL) {
        PyErr_SetString(PyExc_ImportError, "__import__ not found");
        return NULL;
    }

    /* Fast path for not overloaded __import__. */
    if (import_func == PyThreadState_GET()->interp->import_func) {
        int ilevel = _PyLong_AsInt(level);
        if (ilevel == -1 && PyErr_Occurred()) {
            return NULL;
        }
        res = PyImport_ImportModuleLevelObject(
                        name,
                        f->f_globals,
                        f->f_locals == NULL ? Py_None : f->f_locals,
                        fromlist,
                        ilevel);
        return res;
    }

    Py_INCREF(import_func);

    stack[0] = name;
    stack[1] = f->f_globals;
    stack[2] = f->f_locals == NULL ? Py_None : f->f_locals;
    stack[3] = fromlist;
    stack[4] = level;
    res = _PyObject_FastCall(import_func, stack, 5);
    Py_DECREF(import_func);
    return res;
}

• import_name函数首先通过_Py_IDENTIFIER获取内建函数__import__并将其包装为可执行的PyCFunctionObject对象:

/* This structure helps managing static strings. The basic usage goes like this:
   Instead of doing

       r = PyObject_CallMethod(o, "foo", "args", ...);

   do

       _Py_IDENTIFIER(foo);
       ...
       r = _PyObject_CallMethodId(o, &PyId_foo, "args", ...);

   PyId_foo is a static variable, either on block level or file level. On first
   usage, the string "foo" is interned, and the structures are linked. On interpreter
   shutdown, all strings are released (through _PyUnicode_ClearStaticStrings).

   Alternatively, _Py_static_string allows choosing the variable name.
   _PyUnicode_FromId returns a borrowed reference to the interned string.
   _PyObject_{Get,Set,Has}AttrId are __getattr__ versions using _Py_Identifier*.
*/
typedef struct _Py_Identifier {
    struct _Py_Identifier *next;
    const char* string;
    PyObject *object;
} _Py_Identifier;

#define _Py_static_string_init(value) { .next = NULL, .string = value, .object = NULL }
#define _Py_static_string(varname, value)  static _Py_Identifier varname = _Py_static_string_init(value)
#define _Py_IDENTIFIER(varname) _Py_static_string(PyId_##varname, #varname)

• 接着判断__import__是否被重载，如果没有，则调用PyImport_ImportModuleLevelObject:

Python\import.c

PyObject *
PyImport_ImportModuleLevelObject(PyObject *name, PyObject *globals,
                                 PyObject *locals, PyObject *fromlist,
                                 int level)
{
    _Py_IDENTIFIER(_handle_fromlist);
    PyObject *abs_name = NULL;
    PyObject *final_mod = NULL;
    PyObject *mod = NULL;
    PyObject *package = NULL;
    PyInterpreterState *interp = PyThreadState_GET()->interp;
    int has_from;

    if (name == NULL) {
        PyErr_SetString(PyExc_ValueError, "Empty module name");
        goto error;
    }

    /* The below code is importlib.__import__() & _gcd_import(), ported to C
       for added performance. */

    if (!PyUnicode_Check(name)) {
        PyErr_SetString(PyExc_TypeError, "module name must be a string");
        goto error;
    }
    if (PyUnicode_READY(name) < 0) {
        goto error;
    }
    if (level < 0) {
        PyErr_SetString(PyExc_ValueError, "level must be >= 0");
        goto error;
    }

    if (level > 0) {
        abs_name = resolve_name(name, globals, level);
        if (abs_name == NULL)
            goto error;
    }
    else {  /* level == 0 */
        if (PyUnicode_GET_LENGTH(name) == 0) {
            PyErr_SetString(PyExc_ValueError, "Empty module name");
            goto error;
        }
        abs_name = name;
        Py_INCREF(abs_name);
    }

    mod = PyImport_GetModule(abs_name);
    if (mod != NULL && mod != Py_None) {
        _Py_IDENTIFIER(__spec__);
        _Py_IDENTIFIER(_initializing);
        _Py_IDENTIFIER(_lock_unlock_module);
        PyObject *value = NULL;
        PyObject *spec;
        int initializing = 0;

        /* Optimization: only call _bootstrap._lock_unlock_module() if
           __spec__._initializing is true.
           NOTE: because of this, initializing must be set *before*
           stuffing the new module in sys.modules.
         */
        spec = _PyObject_GetAttrId(mod, &PyId___spec__);
        if (spec != NULL) {
            value = _PyObject_GetAttrId(spec, &PyId__initializing);
            Py_DECREF(spec);
        }
        if (value == NULL)
            PyErr_Clear();
        else {
            initializing = PyObject_IsTrue(value);
            Py_DECREF(value);
            if (initializing == -1)
                PyErr_Clear();
            if (initializing > 0) {
                value = _PyObject_CallMethodIdObjArgs(interp->importlib,
                                                &PyId__lock_unlock_module, abs_name,
                                                NULL);
                if (value == NULL)
                    goto error;
                Py_DECREF(value);
            }
        }
    }
    else {
        Py_XDECREF(mod);
        mod = import_find_and_load(abs_name);
        if (mod == NULL) {
            goto error;
        }
    }

    has_from = 0;
    if (fromlist != NULL && fromlist != Py_None) {
        has_from = PyObject_IsTrue(fromlist);
        if (has_from < 0)
            goto error;
    }
    if (!has_from) {
        Py_ssize_t len = PyUnicode_GET_LENGTH(name);
        if (level == 0 || len > 0) {
            Py_ssize_t dot;

            dot = PyUnicode_FindChar(name, '.', 0, len, 1);
            if (dot == -2) {
                goto error;
            }

            if (dot == -1) {
                /* No dot in module name, simple exit */
                final_mod = mod;
                Py_INCREF(mod);
                goto error;
            }

            if (level == 0) {
                PyObject *front = PyUnicode_Substring(name, 0, dot);
                if (front == NULL) {
                    goto error;
                }

                final_mod = PyImport_ImportModuleLevelObject(front, NULL, NULL, NULL, 0);
                Py_DECREF(front);
            }
            else {
                Py_ssize_t cut_off = len - dot;
                Py_ssize_t abs_name_len = PyUnicode_GET_LENGTH(abs_name);
                PyObject *to_return = PyUnicode_Substring(abs_name, 0,
                                                        abs_name_len - cut_off);
                if (to_return == NULL) {
                    goto error;
                }

                final_mod = PyImport_GetModule(to_return);
                Py_DECREF(to_return);
                if (final_mod == NULL) {
                    PyErr_Format(PyExc_KeyError,
                                 "%R not in sys.modules as expected",
                                 to_return);
                    goto error;
                }
            }
        }
        else {
            final_mod = mod;
            Py_INCREF(mod);
        }
    }
    else {
        final_mod = _PyObject_CallMethodIdObjArgs(interp->importlib,
                                                  &PyId__handle_fromlist, mod,
                                                  fromlist, interp->import_func,
                                                  NULL);
    }

  error:
    Py_XDECREF(abs_name);
    Py_XDECREF(mod);
    Py_XDECREF(package);
    if (final_mod == NULL)
        remove_importlib_frames();
    return final_mod;
}

• 在这里会用到level参数，如果level大于0，则在响应的父目录寻找abs_name。
• 如果level等于0，则将abs_name赋值为name。
• 在Python中，一个模块被导入后，会加入到sys.modules中，并首先从sys.modules中查找模块，这就是为什么Python的模块只会被加载一次。
• 接着，经过一系列判断，虚拟机调用PyImport_GetModule获取模块。

Python\import.c

PyObject *
PyImport_GetModule(PyObject *name)
{
    PyObject *m;
    PyObject *modules = PyImport_GetModuleDict();
    if (modules == NULL) {
        PyErr_SetString(PyExc_RuntimeError, "unable to get sys.modules");
        return NULL;
    }
    Py_INCREF(modules);
    if (PyDict_CheckExact(modules)) {
        m = PyDict_GetItemWithError(modules, name);  /* borrowed */
        Py_XINCREF(m);
    }
    else {
        m = PyObject_GetItem(modules, name);
        if (m == NULL && PyErr_ExceptionMatches(PyExc_KeyError)) {
            PyErr_Clear();
        }
    }
    Py_DECREF(modules);
    return m;
}

• 最后调用_PyObject_CallMethodIdObjArgs函数，导入模块，并通过STORE_NAME将其压入栈中：

Objects\call.c

PyObject *
_PyObject_CallMethodIdObjArgs(PyObject *obj,
                              struct _Py_Identifier *name, ...)
{
    va_list vargs;
    PyObject *callable, *result;

    if (obj == NULL || name == NULL) {
        return null_error();
    }

    callable = _PyObject_GetAttrId(obj, name);
    if (callable == NULL) {
        return NULL;
    }

    va_start(vargs, name);
    result = object_vacall(callable, vargs);
    va_end(vargs);

    Py_DECREF(callable);
    return result;
}