php脚本cli 模式运行

参考文章 http://rapheal.sinaapp.com/2013/11/20/php_zend_hello_world/

　　　　http://www.douban.com/note/337885681/

 5.3 与以前版本的不同 http://my.oschina.net/mickelfeng/blog/201382

php添加新语法  http://www.it165.net/pro/html/201208/3477.html

　　　　　　　　http://nikic.github.io/2012/07/27/How-to-add-new-syntactic-features-to-PHP.html

　　　　　　　　http://zsong.me/posts/1

　　　　　　　　http://www.laruence.com/2009/02/21/662.html

　　　　　　　　http://www.verydemo.com/cm.jsp?c=14&u=shen-ru-li-jie-php-yuan-li-zhi

　　　　　　　　http://bbs.phpchina.com/thread-100825-1-1.html

 

static http://www.nowamagic.net/librarys/veda/detail/1402

cli方式运行php脚本

 

$PHPSRC/sapi/cli/php_cli.c中的do_cli函数里边接收了命令行的参数输入

zend_execute_scripts()  $PHPSRC/main/main.c里边有定义

 EG(active_op_array) = zend_compile_file(file_handle, type TSRMLS_CC);
zend_execute(EG(active_op_array) TSRMLS_CC);

 

通过zend_compile_file（实际上为compile_file()<定义在Zend/zend_language_scanner.c的555行>）把脚本文件编译出opcode，

ZEND_API int zend_execute_scripts(int type TSRMLS_DC, zval **retval, int file_count, ...) /* {{{ */
{
    va_list files;
    int i;
    zend_file_handle *file_handle;
    zend_op_array *orig_op_array = EG(active_op_array);
    zval **orig_retval_ptr_ptr = EG(return_value_ptr_ptr);
    long orig_interactive = CG(interactive);

    va_start(files, file_count);
    for (i = 0; i < file_count; i++) {
        file_handle = va_arg(files, zend_file_handle *);
        if (!file_handle) {
            continue;
        }

        if (orig_interactive) {
            if (file_handle->filename[0] != '-' || file_handle->filename[1]) {
                CG(interactive) = 0;
            } else {
                CG(interactive) = 1;
            }
        }
       
        EG(active_op_array) = zend_compile_file(file_handle, type TSRMLS_CC);
        if (file_handle->opened_path) {
            int dummy = 1;
            zend_hash_add(&EG(included_files), file_handle->opened_path, strlen(file_handle->opened_path) + 1, (void *)&dummy, sizeof(int), NULL);
        }
        zend_destroy_file_handle(file_handle TSRMLS_CC);
        if (EG(active_op_array)) {
            EG(return_value_ptr_ptr) = retval ? retval : NULL;
            zend_execute(EG(active_op_array) TSRMLS_CC);
            zend_exception_restore(TSRMLS_C);
            if (EG(exception)) {
                if (EG(user_exception_handler)) {
                    zval *orig_user_exception_handler;
                    zval **params[1], *retval2, *old_exception;
                    old_exception = EG(exception);
                    EG(exception) = NULL;
                    params[0] = &old_exception;
                    orig_user_exception_handler = EG(user_exception_handler);
                    if (call_user_function_ex(CG(function_table), NULL, orig_user_exception_handler, &retval2, 1, params, 1, NULL TSRMLS_CC) == SUCCESS) {
                        if (retval2 != NULL) {
                            zval_ptr_dtor(&retval2);
                        }
                        if (EG(exception)) {
                            zval_ptr_dtor(&EG(exception));
                            EG(exception) = NULL;
                        }
                        zval_ptr_dtor(&old_exception);
                    } else {
                        EG(exception) = old_exception;
                        zend_exception_error(EG(exception), E_ERROR TSRMLS_CC);
                    }
                } else {
                    zend_exception_error(EG(exception), E_ERROR TSRMLS_CC);
                }
            }
            destroy_op_array(EG(active_op_array) TSRMLS_CC);
            efree(EG(active_op_array));
        } else if (type==ZEND_REQUIRE) {
            va_end(files);
            EG(active_op_array) = orig_op_array;
            EG(return_value_ptr_ptr) = orig_retval_ptr_ptr;
            CG(interactive) = orig_interactive;
            return FAILURE;
        }
    }
    va_end(files);
    EG(active_op_array) = orig_op_array;
    EG(return_value_ptr_ptr) = orig_retval_ptr_ptr;
    CG(interactive) = orig_interactive;

    return SUCCESS;
}


zend_compile_file = compile_file;

 

ZEND_API zend_op_array *compile_file(zend_file_handle *file_handle, int type TSRMLS_DC)
{
    zend_lex_state original_lex_state;
    zend_op_array *op_array = (zend_op_array *) emalloc(sizeof(zend_op_array));
    zend_op_array *original_active_op_array = CG(active_op_array);
    zend_op_array *retval=NULL;
    int compiler_result;
    zend_bool compilation_successful=0;
    znode retval_znode;
    zend_bool original_in_compilation = CG(in_compilation);

    retval_znode.op_type = IS_CONST;
    retval_znode.u.constant.type = IS_LONG;
    retval_znode.u.constant.value.lval = 1;
    Z_UNSET_ISREF(retval_znode.u.constant);
    Z_SET_REFCOUNT(retval_znode.u.constant, 1);

    zend_save_lexical_state(&original_lex_state TSRMLS_CC);

    retval = op_array; /* success oriented */

    if (open_file_for_scanning(file_handle TSRMLS_CC)==FAILURE) {
        if (type==ZEND_REQUIRE) {
            zend_message_dispatcher(ZMSG_FAILED_REQUIRE_FOPEN, file_handle->filename TSRMLS_CC);
            zend_bailout();
        } else {
            zend_message_dispatcher(ZMSG_FAILED_INCLUDE_FOPEN, file_handle->filename TSRMLS_CC);
        }
        compilation_successful=0;
    } else {
        init_op_array(op_array, ZEND_USER_FUNCTION, INITIAL_OP_ARRAY_SIZE TSRMLS_CC); //初始化op_array
        CG(in_compilation) = 1;
        CG(active_op_array) = op_array;
        zend_stack_push(&CG(context_stack), (void *) &CG(context), sizeof(CG(context)));
        zend_init_compiler_context(TSRMLS_C);
        compiler_result = zendparse(TSRMLS_C);//对于非函数，都是设置opline的各个属性，对于函数，要额外分配一个op_array
        zend_do_return(&retval_znode, 0 TSRMLS_CC);
        CG(in_compilation) = original_in_compilation;
        if (compiler_result==1) { /* parser error */
            zend_bailout();
        }
        compilation_successful=1;
    }

    if (retval) {
        CG(active_op_array) = original_active_op_array;
        if (compilation_successful) {
            pass_two(op_array TSRMLS_CC);//循环遍历op_array中的opcodes,设置每一个opline中的属性hanlder
            zend_release_labels(TSRMLS_C);
        } else {
            efree(op_array);
            retval = NULL;
        }
    }
    zend_restore_lexical_state(&original_lex_state TSRMLS_CC);
    return retval;
}

 

ZEND_API void execute(zend_op_array *op_array TSRMLS_DC)
{
    zend_execute_data *execute_data;
    zend_bool nested = 0;
    zend_bool original_in_execution = EG(in_execution);


    if (EG(exception)) {
        return;
    }

    EG(in_execution) = 1;

zend_vm_enter:
    /* Initialize execute_data */
    execute_data = (zend_execute_data *)zend_vm_stack_alloc(
        ZEND_MM_ALIGNED_SIZE(sizeof(zend_execute_data)) +
        ZEND_MM_ALIGNED_SIZE(sizeof(zval**) * op_array->last_var * (EG(active_symbol_table) ? 1 : 2)) +
        ZEND_MM_ALIGNED_SIZE(sizeof(temp_variable)) * op_array->T TSRMLS_CC);

    EX(CVs) = (zval***)((char*)execute_data + ZEND_MM_ALIGNED_SIZE(sizeof(zend_execute_data)));
    memset(EX(CVs), 0, sizeof(zval**) * op_array->last_var);
    EX(Ts) = (temp_variable *)(((char*)EX(CVs)) + ZEND_MM_ALIGNED_SIZE(sizeof(zval**) * op_array->last_var * (EG(active_symbol_table) ? 1 : 2)));
    EX(fbc) = NULL;
    EX(called_scope) = NULL;
    EX(object) = NULL;
    EX(old_error_reporting) = NULL;
    EX(op_array) = op_array;
    EX(symbol_table) = EG(active_symbol_table);
    EX(prev_execute_data) = EG(current_execute_data);
    EG(current_execute_data) = execute_data;
    EX(nested) = nested;
    nested = 1;

    if (op_array->start_op) {
        ZEND_VM_SET_OPCODE(op_array->start_op);
    } else {
        ZEND_VM_SET_OPCODE(op_array->opcodes);
    }

    if (op_array->this_var != -1 && EG(This)) {
        Z_ADDREF_P(EG(This)); /* For $this pointer */
        if (!EG(active_symbol_table)) {
            EX(CVs)[op_array->this_var] = (zval**)EX(CVs) + (op_array->last_var + op_array->this_var);
            *EX(CVs)[op_array->this_var] = EG(This);
        } else {
            if (zend_hash_add(EG(active_symbol_table), "this", sizeof("this"), &EG(This), sizeof(zval *), (void**)&EX(CVs)[op_array->this_var])==FAILURE) {
                Z_DELREF_P(EG(This));
            }
        }
    }

    EG(opline_ptr) = &EX(opline);

    EX(function_state).function = (zend_function *) op_array;
    EX(function_state).arguments = NULL;
    
    while (1) {
        int ret;
#ifdef ZEND_WIN32
        if (EG(timed_out)) {
            zend_timeout(0);
        }
#endif

        if ((ret = EX(opline)->handler(execute_data TSRMLS_CC)) > 0) {
            switch (ret) {
                case 1:
                    EG(in_execution) = original_in_execution;
                    return;
                case 2:
                    op_array = EG(active_op_array);
                    goto zend_vm_enter;
                case 3:
                    execute_data = EG(current_execute_data);
                default:
                    break;
            }
        }

    }
    zend_error_noreturn(E_ERROR, "Arrived at end of main loop which shouldn't happen");
}

 zendparse (LALR1 将代码分隔成token，再解析后存放在opcode的数组中)

pass_two解析opcode的每一行，将其handler绑定到op上

complie之后会得到一个op_array，在execute之前创建一个zend_execute_data的数据结构

struct _zend_execute_data {
        struct _zend_op *opline;
        zend_function_state function_state;
        zend_op_array *op_array;
        zval *object;
        HashTable *symbol_table;
        struct _zend_execute_data *prev_execute_data;
        zval *old_error_reporting;
        zend_bool nested;
        zval **original_return_value;
        zend_class_entry *current_scope;
        zend_class_entry *current_called_scope;
        zval *current_this;
        struct _zend_op *fast_ret; /* used by FAST_CALL/FAST_RET (finally keyword) */
        call_slot *call_slots;
        call_slot *call;
};

在zend_execute中最重要的一句
ret = OPLINE->handler(execute_data TSRMLS_CC)

调用在op上绑定的handler，完成对opcode的执行

//zend_opcode.c

ZEND_API int pass_two(zend_op_array *op_array TSRMLS_DC)
{
zend_op *opline, *end;

if (op_array->type!=ZEND_USER_FUNCTION && op_array->type!=ZEND_EVAL_CODE) {
return 0;
}
if (CG(compiler_options) & ZEND_COMPILE_EXTENDED_INFO) {
zend_update_extended_info(op_array TSRMLS_CC);
}
if (CG(compiler_options) & ZEND_COMPILE_HANDLE_OP_ARRAY) {
zend_llist_apply_with_argument(&zend_extensions, (llist_apply_with_arg_func_t) zend_extension_op_array_handler, op_array TSRMLS_CC);
}

if (!(op_array->fn_flags & ZEND_ACC_INTERACTIVE) && CG(context).vars_size != op_array->last_var) {
op_array->vars = (zend_compiled_variable *) erealloc(op_array->vars, sizeof(zend_compiled_variable)*op_array->last_var);
CG(context).vars_size = op_array->last_var;
}
if (!(op_array->fn_flags & ZEND_ACC_INTERACTIVE) && CG(context).opcodes_size != op_array->last) {
op_array->opcodes = (zend_op *) erealloc(op_array->opcodes, sizeof(zend_op)*op_array->last);
CG(context).opcodes_size = op_array->last;
}
if (!(op_array->fn_flags & ZEND_ACC_INTERACTIVE) && CG(context).literals_size != op_array->last_literal) {
op_array->literals = (zend_literal*)erealloc(op_array->literals, sizeof(zend_literal) * op_array->last_literal);
CG(context).literals_size = op_array->last_literal;
}

opline = op_array->opcodes;
end = opline + op_array->last;
while (opline < end) {
if (opline->op1_type == IS_CONST) {
opline->op1.zv = &op_array->literals[opline->op1.constant].constant;
}
if (opline->op2_type == IS_CONST) {
opline->op2.zv = &op_array->literals[opline->op2.constant].constant;
}
switch (opline->opcode) {
case ZEND_GOTO:
if (Z_TYPE_P(opline->op2.zv) != IS_LONG) {
zend_resolve_goto_label(op_array, opline, 1 TSRMLS_CC);
}
/* break omitted intentionally */
case ZEND_JMP:
opline->op1.jmp_addr = &op_array->opcodes[opline->op1.opline_num];
break;
case ZEND_JMPZ:
case ZEND_JMPNZ:
case ZEND_JMPZ_EX:
case ZEND_JMPNZ_EX:
case ZEND_JMP_SET:
case ZEND_JMP_SET_VAR:
opline->op2.jmp_addr = &op_array->opcodes[opline->op2.opline_num];
break;
}
ZEND_VM_SET_OPCODE_HANDLER(opline);
opline++;
}

op_array->fn_flags |= ZEND_ACC_DONE_PASS_TWO;
return 0;
}

 

//zend_vm_execute.h

ZEND_API void zend_vm_set_opcode_handler(zend_op* op)
{
op->handler = zend_vm_get_opcode_handler(zend_user_opcodes[op->opcode], op);
}