BlogJava-Sealyu-随笔分类-Python

MySQLdb User's Guide

seal — Mon, 17 May 2010 15:56:00 GMT

Introduction

MySQLdb is an thread-compatible interface to the popular MySQL database server that provides the Python database API.

Installation

The README file has complete installation instructions.

_mysql

If you want to write applications which are portable across databases, use MySQLdb, and avoid using this module directly. _mysql provides an interface which mostly implements the MySQL C API. For more information, see the MySQL documentation. The documentation for this module is intentionally weak because you probably should use the higher-level MySQLdb module. If you really need it, use the standard MySQL docs and transliterate as necessary.

MySQL C API translation

The MySQL C API has been wrapped in an object-oriented way. The only MySQL data structures which are implemented are the MYSQL (database connection handle) and MYSQL_RES (result handle) types. In general, any function which takes MYSQL *mysql as an argument is now a method of the connection object, and any function which takes MYSQL_RES *result as an argument is a method of the result object. Functions requiring none of the MySQL data structures are implemented as functions in the module. Functions requiring one of the other MySQL data structures are generally not implemented. Deprecated functions are not implemented. In all cases, the mysql_ prefix is dropped from the name. Most of the conn methods listed are also available as MySQLdb Connection object methods. Their use is non-portable.

MySQL C API function mapping

C API	`_mysql`
`mysql_affected_rows()`	`conn.affected_rows()`
`mysql_autocommit()`	`conn.autocommit()`
`mysql_character_set_name()`	`conn.character_set_name()`
`mysql_close()`	`conn.close()`
`mysql_commit()`	`conn.commit()`
`mysql_connect()`	`_mysql.connect()`
`mysql_data_seek()`	`result.data_seek()`
`mysql_debug()`	`_mysql.debug()`
`mysql_dump_debug_info`	`conn.dump_debug_info()`
`mysql_escape_string()`	`_mysql.escape_string()`
`mysql_fetch_row()`	`result.fetch_row()`
`mysql_get_character_set_info()`	`conn.get_character_set_info()`
`mysql_get_client_info()`	`_mysql.get_client_info()`
`mysql_get_host_info()`	`conn.get_host_info()`
`mysql_get_proto_info()`	`conn.get_proto_info()`
`mysql_get_server_info()`	`conn.get_server_info()`
`mysql_info()`	`conn.info()`
`mysql_insert_id()`	`conn.insert_id()`
`mysql_num_fields()`	`result.num_fields()`
`mysql_num_rows()`	`result.num_rows()`
`mysql_options()`	various options to `_mysql.connect()`
`mysql_ping()`	`conn.ping()`
`mysql_query()`	`conn.query()`
`mysql_real_connect()`	`_mysql.connect()`
`mysql_real_query()`	`conn.query()`
`mysql_real_escape_string()`	`conn.escape_string()`
`mysql_rollback()`	`conn.rollback()`
`mysql_row_seek()`	`result.row_seek()`
`mysql_row_tell()`	`result.row_tell()`
`mysql_select_db()`	`conn.select_db()`
`mysql_set_character_set()`	`conn.set_character_set()`
`mysql_ssl_set()`	`ssl` option to `_mysql.connect()`
`mysql_stat()`	`conn.stat()`
`mysql_store_result()`	`conn.store_result()`
`mysql_thread_id()`	`conn.thread_id()`
`mysql_thread_safe_client()`	`conn.thread_safe_client()`
`mysql_use_result()`	`conn.use_result()`
`mysql_warning_count()`	`conn.warning_count()`
`CLIENT_*`	`MySQLdb.constants.CLIENT.*`
`CR_*`	`MySQLdb.constants.CR.*`
`ER_*`	`MySQLdb.constants.ER.*`
`FIELD_TYPE_*`	`MySQLdb.constants.FIELD_TYPE.*`
`FLAG_*`	`MySQLdb.constants.FLAG.*`

Some _mysql examples

Okay, so you want to use _mysql anyway. Here are some examples.

The simplest possible database connection is:

import _mysql

db=_mysql.connect()

This creates a connection to the MySQL server running on the local machine using the standard UNIX socket (or named pipe on Windows), your login name (from the USER environment variable), no password, and does not USE a database. Chances are you need to supply more information.:

db=_mysql.connect("localhost","joebob","moonpie","thangs")

This creates a connection to the MySQL server running on the local machine via a UNIX socket (or named pipe), the user name "joebob", the password "moonpie", and selects the initial database "thangs".

We haven't even begun to touch upon all the parameters connect() can take. For this reason, I prefer to use keyword parameters:

db=_mysql.connect(host="localhost",user="joebob",

passwd="moonpie",db="thangs")

This does exactly what the last example did, but is arguably easier to read. But since the default host is "localhost", and if your login name really was "joebob", you could shorten it to this:

db=_mysql.connect(passwd="moonpie",db="thangs")

UNIX sockets and named pipes don't work over a network, so if you specify a host other than localhost, TCP will be used, and you can specify an odd port if you need to (the default port is 3306):

db=_mysql.connect(host="outhouse",port=3307,passwd="moonpie",db="thangs")

If you really had to, you could connect to the local host with TCP by specifying the full host name, or 127.0.0.1.

Generally speaking, putting passwords in your code is not such a good idea:

db=_mysql.connect(host="outhouse",db="thangs",read_default_file="~/.my.cnf")

This does what the previous example does, but gets the username and password and other parameters from ~/.my.cnf (UNIX-like systems). Read about option files for more details.

So now you have an open connection as db and want to do a query. Well, there are no cursors in MySQL, and no parameter substitution, so you have to pass a complete query string to db.query():

db.query("""SELECT spam, eggs, sausage FROM breakfast

WHERE price < 5""")

There's no return value from this, but exceptions can be raised. The exceptions are defined in a separate module, _mysql_exceptions, but _mysql exports them. Read DB API specification PEP-249 to find out what they are, or you can use the catch-all MySQLError.

At this point your query has been executed and you need to get the results. You have two options:

r=db.store_result()

# ...or...

r=db.use_result()

Both methods return a result object. What's the difference? store_result() returns the entire result set to the client immediately. If your result set is really large, this could be a problem. One way around this is to add a LIMIT clause to your query, to limit the number of rows returned. The other is to use use_result(), which keeps the result set in the server and sends it row-by-row when you fetch. This does, however, tie up server resources, and it ties up the connection: You cannot do any more queries until you have fetched all the rows. Generally I recommend using store_result() unless your result set is really huge and you can't use LIMIT for some reason.

Now, for actually getting real results:

>>> r.fetch_row()

(('3','2','0'),)

This might look a little odd. The first thing you should know is, fetch_row() takes some additional parameters. The first one is, how many rows (maxrows) should be returned. By default, it returns one row. It may return fewer rows than you asked for, but never more. If you set maxrows=0, it returns all rows of the result set. If you ever get an empty tuple back, you ran out of rows.

The second parameter (how) tells it how the row should be represented. By default, it is zero which means, return as a tuple. how=1 means, return it as a dictionary, where the keys are the column names, or table.column if there are two columns with the same name (say, from a join). how=2 means the same as how=1 except that the keys are always table.column; this is for compatibility with the old Mysqldb module.

OK, so why did we get a 1-tuple with a tuple inside? Because we implicitly asked for one row, since we didn't specify maxrows.

The other oddity is: Assuming these are numeric columns, why are they returned as strings? Because MySQL returns all data as strings and expects you to convert it yourself. This would be a real pain in the ass, but in fact, _mysql can do this for you. (And MySQLdb does do this for you.) To have automatic type conversion done, you need to create a type converter dictionary, and pass this to connect() as the conv keyword parameter.

The keys of conv should be MySQL column types, which in the C API are FIELD_TYPE_*. You can get these values like this:

from MySQLdb.constants import FIELD_TYPE

By default, any column type that can't be found in conv is returned as a string, which works for a lot of stuff. For our purposes, we probably want this:

my_conv = { FIELD_TYPE.LONG: int }

This means, if it's a FIELD_TYPE_LONG, call the builtin int() function on it. Note that FIELD_TYPE_LONG is an INTEGER column, which corresponds to a C long, which is also the type used for a normal Python integer. But beware: If it's really an UNSIGNED INTEGER column, this could cause overflows. For this reason, MySQLdb actually uses long() to do the conversion. But we'll ignore this potential problem for now.

Then if you use db=_mysql.connect(conv=my_conv...), the results will come back ((3, 2, 0),), which is what you would expect.

MySQLdb

MySQLdb is a thin Python wrapper around _mysql which makes it compatible with the Python DB API interface (version 2). In reality, a fair amount of the code which implements the API is in _mysql for the sake of efficiency.

The DB API specification PEP-249 should be your primary guide for using this module. Only deviations from the spec and other database-dependent things will be documented here.

Functions and attributes

Only a few top-level functions and attributes are defined within MySQLdb.

connect(parameters...)

Constructor for creating a connection to the database. Returns a Connection Object. Parameters are the same as for the MySQL C API. In addition, there are a few additional keywords that correspond to what you would pass mysql_options() before connecting. Note that some parameters must be specified as keyword arguments! The default value for each parameter is NULL or zero, as appropriate. Consult the MySQL documentation for more details. The important parameters are:

host

name of host to connect to. Default: use the local host via a UNIX socket (where applicable)

user

user to authenticate as. Default: current effective user.

passwd

password to authenticate with. Default: no password.

db

database to use. Default: no default database.

port

TCP port of MySQL server. Default: standard port (3306).

unix_socket

location of UNIX socket. Default: use default location or TCP for remote hosts.

conv

type conversion dictionary. Default: a copy of MySQLdb.converters.conversions

compress

Enable protocol compression. Default: no compression.

connect_timeout

Abort if connect is not completed within given number of seconds. Default: no timeout (?)

named_pipe

Use a named pipe (Windows). Default: don't.

init_command

Initial command to issue to server upon connection. Default: Nothing.

read_default_file

MySQL configuration file to read; see the MySQL documentation for mysql_options().

read_default_group

Default group to read; see the MySQL documentation for mysql_options().

cursorclass

cursor class that cursor() uses, unless overridden. Default: MySQLdb.cursors.Cursor. This must be a keyword parameter.

use_unicode

If True, CHAR and VARCHAR and TEXT columns are returned as Unicode strings, using the configured character set. It is best to set the default encoding in the server configuration, or client configuration (read with read_default_file). If you change the character set after connecting (MySQL-4.1 and later), you'll need to put the correct character set name in connection.charset.

If False, text-like columns are returned as normal strings, but you can always write Unicode strings.

This must be a keyword parameter.

charset

If present, the connection character set will be changed to this character set, if they are not equal. Support for changing the character set requires MySQL-4.1 and later server; if the server is too old, UnsupportedError will be raised. This option implies use_unicode=True, but you can override this with use_unicode=False, though you probably shouldn't.

If not present, the default character set is used.

This must be a keyword parameter.

sql_mode

If present, the session SQL mode will be set to the given string. For more information on sql_mode, see the MySQL documentation. Only available for 4.1 and newer servers.

If not present, the session SQL mode will be unchanged.

This must be a keyword parameter.

ssl

This parameter takes a dictionary or mapping, where the keys are parameter names used by the mysql_ssl_set MySQL C API call. If this is set, it initiates an SSL connection to the server; if there is no SSL support in the client, an exception is raised. This must be a keyword parameter.

apilevel

String constant stating the supported DB API level. '2.0'

threadsafety

Integer constant stating the level of thread safety the interface supports. This is set to 1, which means: Threads may share the module.

The MySQL protocol can not handle multiple threads using the same connection at once. Some earlier versions of MySQLdb utilized locking to achieve a threadsafety of 2. While this is not terribly hard to accomplish using the standard Cursor class (which uses mysql_store_result()), it is complicated by SSCursor (which uses mysql_use_result(); with the latter you must ensure all the rows have been read before another query can be executed. It is further complicated by the addition of transactions, since transactions start when a cursor execute a query, but end when COMMIT or ROLLBACK is executed by the Connection object. Two threads simply cannot share a connection while a transaction is in progress, in addition to not being able to share it during query execution. This excessively complicated the code to the point where it just isn't worth it.

The general upshot of this is: Don't share connections between threads. It's really not worth your effort or mine, and in the end, will probably hurt performance, since the MySQL server runs a separate thread for each connection. You can certainly do things like cache connections in a pool, and give those connections to one thread at a time. If you let two threads use a connection simultaneously, the MySQL client library will probably upchuck and die. You have been warned.

For threaded applications, try using a connection pool. This can be done using the Pool module.

charset

The character set used by the connection. In MySQL-4.1 and newer, it is possible (but not recommended) to change the connection's character set with an SQL statement. If you do this, you'll also need to change this attribute. Otherwise, you'll get encoding errors.

paramstyle

String constant stating the type of parameter marker formatting expected by the interface. Set to 'format' = ANSI C printf format codes, e.g. '...WHERE name=%s'. If a mapping object is used for conn.execute(), then the interface actually uses 'pyformat' = Python extended format codes, e.g. '...WHERE name=%(name)s'. However, the API does not presently allow the specification of more than one style in paramstyle.

Note that any literal percent signs in the query string passed to execute() must be escaped, i.e. %%.

Parameter placeholders can only be used to insert column values. They can not be used for other parts of SQL, such as table names, statements, etc.

conv

A dictionary or mapping which controls how types are converted from MySQL to Python and vice versa.

If the key is a MySQL type (from FIELD_TYPE.*), then the value can be either:

a callable object which takes a string argument (the MySQL value),' returning a Python value
a sequence of 2-tuples, where the first value is a combination of flags from MySQLdb.constants.FLAG, and the second value is a function as above. The sequence is tested until the flags on the field match those of the first value. If both values are None, then the default conversion is done. Presently this is only used to distinquish TEXT and BLOB columns.

If the key is a Python type or class, then the value is a callable Python object (usually a function) taking two arguments (value to convert, and the conversion dictionary) which converts values of this type to a SQL literal string value.

This is initialized with reasonable defaults for most types. When creating a Connection object, you can pass your own type converter dictionary as a keyword parameter. Otherwise, it uses a copy of MySQLdb.converters.conversions. Several non-standard types are returned as strings, which is how MySQL returns all columns. For more details, see the built-in module documentation.

Connection Objects

Connection objects are returned by the connect() function.

commit(): If the database and the tables support transactions, this commits the current transaction; otherwise this method successfully does nothing.
rollback(): If the database and tables support transactions, this rolls back (cancels) the current transaction; otherwise a NotSupportedError is raised.
cursor([cursorclass]): MySQL does not support cursors; however, cursors are easily emulated. You can supply an alternative cursor class as an optional parameter. If this is not present, it defaults to the value given when creating the connection object, or the standard Cursor class. Also see the additional supplied cursor classes in the usage section.

There are many more methods defined on the connection object which are MySQL-specific. For more information on them, consult the internal documentation using pydoc.

Cursor Objects

callproc(procname, args)

Calls stored procedure procname with the sequence of arguments in args. Returns the original arguments. Stored procedure support only works with MySQL-5.0 and newer.

Compatibility note: PEP-249 specifies that if there are OUT or INOUT parameters, the modified values are to be returned. This is not consistently possible with MySQL. Stored procedure arguments must be passed as server variables, and can only be returned with a SELECT statement. Since a stored procedure may return zero or more result sets, it is impossible for MySQLdb to determine if there are result sets to fetch before the modified parmeters are accessible.

The parameters are stored in the server as @_*procname*_*n*, where n is the position of the parameter. I.e., if you cursor.callproc('foo', (a, b, c)), the parameters will be accessible by a SELECT statement as @_foo_0, @_foo_1, and @_foo_2.

Compatibility note: It appears that the mere act of executing the CALL statement produces an empty result set, which appears after any result sets which might be generated by the stored procedure. Thus, you will always need to use nextset() to advance result sets.

close()

Closes the cursor. Future operations raise ProgrammingError. If you are using server-side cursors, it is very important to close the cursor when you are done with it and before creating a new one.

info()

Returns some information about the last query. Normally you don't need to check this. If there are any MySQL warnings, it will cause a Warning to be issued through the Python warning module. By default, Warning causes a message to appear on the console. However, it is possible to filter these out or cause Warning to be raised as exception. See the MySQL docs for mysql_info(), and the Python warning module. (Non-standard)

setinputsizes()

Does nothing, successfully.

setoutputsizes()

Does nothing, successfully.

nextset()

Advances the cursor to the next result set, discarding the remaining rows in the current result set. If there are no additional result sets, it returns None; otherwise it returns a true value.

Note that MySQL doesn't support multiple result sets until 4.1.

Some examples

The connect() method works nearly the same as with _mysql:

import MySQLdb

db=MySQLdb.connect(passwd="moonpie",db="thangs")

To perform a query, you first need a cursor, and then you can execute queries on it:

c=db.cursor()

max_price=5

c.execute("""SELECT spam, eggs, sausage FROM breakfast

WHERE price < %s""", (max_price,))

In this example, max_price=5 Why, then, use %s in the string? Because MySQLdb will convert it to a SQL literal value, which is the string '5'. When it's finished, the query will actually say, "...WHERE price < 5".

Why the tuple? Because the DB API requires you to pass in any parameters as a sequence. Due to the design of the parser, (max_price) is interpreted as using algebraic grouping and simply as max_price and not a tuple. Adding a comma, i.e. (max_price,) forces it to make a tuple.

And now, the results:

>>> c.fetchone()

(3L, 2L, 0L)

Quite unlike the _mysql example, this returns a single tuple, which is the row, and the values are properly converted by default... except... What's with the L's?

As mentioned earlier, while MySQL's INTEGER column translates perfectly into a Python integer, UNSIGNED INTEGER could overflow, so these values are converted to Python long integers instead.

If you wanted more rows, you could use c.fetchmany(n) or c.fetchall(). These do exactly what you think they do. On c.fetchmany(n), the n is optional and defaults to c.arraysize, which is normally 1. Both of these methods return a sequence of rows, or an empty sequence if there are no more rows. If you use a weird cursor class, the rows themselves might not be tuples.

Note that in contrast to the above, c.fetchone() returns None when there are no more rows to fetch.

The only other method you are very likely to use is when you have to do a multi-row insert:

c.executemany(

"""INSERT INTO breakfast (name, spam, eggs, sausage, price)

VALUES (%s, %s, %s, %s, %s)""",

[

("Spam and Sausage Lover's Plate", 5, 1, 8, 7.95 ),

("Not So Much Spam Plate", 3, 2, 0, 3.95 ),

("Don't Wany ANY SPAM! Plate", 0, 4, 3, 5.95 )

] )

Here we are inserting three rows of five values. Notice that there is a mix of types (strings, ints, floats) though we still only use %s. And also note that we only included format strings for one row. MySQLdb picks those out and duplicates them for each row.

Using and extending

In general, it is probably wise to not directly interact with the DB API except for small applicatons. Databases, even SQL databases, vary widely in capabilities and may have non-standard features. The DB API does a good job of providing a reasonably portable interface but some methods are non-portable. Specifically, the parameters accepted by connect() are completely implementation-dependent.

If you believe your application may need to run on several different databases, the author recommends the following approach, based on personal experience: Write a simplified API for your application which implements the specific queries and operations your application needs to perform. Implement this API as a base class which should be have few database dependencies, and then derive a subclass from this which implements the necessary dependencies. In this way, porting your application to a new database should be a relatively simple matter of creating a new subclass, assuming the new database is reasonably standard.

Because MySQLdb's Connection and Cursor objects are written in Python, you can easily derive your own subclasses. There are several Cursor classes in MySQLdb.cursors:

BaseCursor: The base class for Cursor objects. This does not raise Warnings.
CursorStoreResultMixIn: Causes the Cursor to use the mysql_store_result() function to get the query result. The entire result set is stored on the client side.
CursorUseResultMixIn: Causes the cursor to use the mysql_use_result() function to get the query result. The result set is stored on the server side and is transferred row by row using fetch operations.
CursorTupleRowsMixIn: Causes the cursor to return rows as a tuple of the column values.

CursorDictRowsMixIn

Causes the cursor to return rows as a dictionary, where the keys are column names and the values are column values. Note that if the column names are not unique, i.e., you are selecting from two tables that share column names, some of them will be rewritten as table.column. This can be avoided by using the SQL AS keyword. (This is yet-another reason not to use * in SQL queries, particularly where JOIN is involved.)

Cursor: The default cursor class. This class is composed of CursorWarningMixIn, CursorStoreResultMixIn, CursorTupleRowsMixIn, and BaseCursor, i.e. it raises Warning, uses mysql_store_result(), and returns rows as tuples.
DictCursor: Like Cursor except it returns rows as dictionaries.
SSCursor: A "server-side" cursor. Like Cursor but uses CursorUseResultMixIn. Use only if you are dealing with potentially large result sets.
SSDictCursor: Like SSCursor except it returns rows as dictionaries.

Embedded Server

Instead of connecting to a stand-alone server over the network, the embedded server support lets you run a full server right in your Python code or application server.

If you have built MySQLdb with embedded server support, there are two additional functions you will need to make use of:

server_init(args, groups)

Initialize embedded server. If this client is not linked against the embedded server library, this function does nothing.

args
sequence of command-line arguments
groups
sequence of groups to use in defaults files

server_end()
Shut down embedded server. If not using an embedded server, this does nothing.

See the MySQL documentation for more information on the embedded server.

Title:	MySQLdb: a Python interface for MySQL
Author:	Andy Dustman
Version:	$Revision: 421 $

seal 2010-05-17 23:56 发表评论

简明 Python 编程规范（转）

seal — Thu, 13 May 2010 18:59:00 GMT

本文最初发表于赖勇浩（恋花蝶）的博客：http://blog.csdn.net/lanphaday，如蒙转载，必须保留全文完整，未经本人同意，不得用于商业目的。

简明 Python 编程规范

编码

    所有的 Python 脚本文件都应在文件头标上 # -*- coding:utf-8 -*- 。设置编辑器，默认保存为 utf-8 格式。

注释

    业界普遍认同 Python 的注释分为两种的概念，一种是由 # 开头的“真正的”注释，另一种是 docstrings。前者表明为何选择当前实现以及这种实现的原理和难点，后者表明如何使用这个包、模块、类、函数（方法），甚至包括使用示例和单元测试。
    坚持适当注释原则。对不存在技术难点的代码坚持不注释，对存在技术难点的代码必须注释。但与注释不同，推荐对每一个包、模块、类、函数（方法）写 docstrings，除非代码一目了然，非常简单。

缩进

    Python 依赖缩进来确定代码块的层次，行首空白符主要有两种：tab 和空格，但严禁两者混用。如果使用 tab 缩进，设定 tab 为 4 个空格。
    公司内部推荐使用 4 个空格的 tab 进行缩进。
空格

    空格在 Python 代码中是有意义的，因为 Python 的语法依赖于缩进，在行首的空格称为前导空格。在这一节不讨论前导空格相关的内容，只讨论非前导空格。非前导空格在 Python 代码中没有意义，但适当地加入非前导空格可以增进代码的可读性。
1）在二元算术、逻辑运算符前后加空格：如 a = b + c；
2）在一元前缀运算符后不加空格，如 if !flg: pass；
3） “:” 用在行尾时前后皆不加空格，如分枝、循环、函数和类定义语言；用在非行尾时两端加空格，如 dict 对象的定义 d = {‘key’ : ’value’}。
4）括号（含圆括号、方括号和花括号）前后不加空格，如 do_something(arg1, arg2)，而不是 do_something( arg1, arg2 )；
5）逗号后面加一个空格，前面不加空格；s

空行

    适当的空行有利于增加代码的可读性，加空行可以参考如下几个准则：
1）在类、函数的定义间加空行；
2）在 import 不同种类的模块间加工行；
3）在函数中的逻辑段落间加空行，即把相关的代码紧凑写在一起，作为一个逻辑段落，段落间以空行分隔；

断行

尽管现在的宽屏显示器已经可以单屏显示超过 256 列字符，但本规范仍然坚持行的最大长度不得超过 78 个字符的标准。折叠长行的方法有以下几种方法：
1) 为长变量名换一个短名，如：
    this.is.a.very.long.variable_name = this.is.another.long.variable_name
    应改为：
    variable_name1 = this.is.a.very.long.variable_name
    variable_name2 = this.is.another.variable_name
    variable_name1 = variable_name2s
2) 在括号（包括圆括号、方括号和花括号）内换行，如：
    class Edit(CBase):
        def __init__(self, parent, width,
                                font = FONT, color = BLACK, pos = POS, style = 0):
    或：
    very_very_very_long_variable_name = Edit(parent, "
                                                                            width, "
                                                                            font, "
                                                                            color, "
                                                                            pos)
    如果行长到连第一个括号内的参数都放不下，则每个元素都单独占一行：
    very_very_very_long_variable_name = ui.widgets.Edit( "
                                                                                    panrent, "
                                                                                    width, "
                                                                                    font, "
                                                                                    color, "
                                                                                    pos)
3) 在长行加入续行符强行断行，断行的位置应在操作符前，且换行后多一个缩进，以使维护人员看代码的时候看到代码行首即可判定这里存在换行，如：
if color == WHITE or color == BLACK "
or color == BLUE: # 注意 or 操作符在新行的行首而不是旧行的行尾
do_something(color);

命名

    一致的命名可以给开发人员减少许多麻烦，而恰如其分的命名则可以大幅提高代码的可读性，降低维护成本。

常量

常量名所有字母大写，由下划线连接各个单词，如
WHITE = 0XFFFFFF
THIS_IS_A_CONSTANT = 1

变量

    变量名全部小写，由下划线连接各个单词，如
color = WHITE
this_is_a_variable = 1
    不论是类成员变量还是全局变量，均不使用 m 或 g 前缀。私有类成员使用单一下划线前缀标识，多定义公开成员，少定义私有成员。
    变量名不应带有类型信息，因为 Python 是动态类型语言。如 iValue、names_list、dict_obj 等都是不好的命名。

函数

    函数名的命名规则与变量名相同。
类
    类名单词首字母大写，不使用下划线连接单词，也不加入 C、T 等前缀。如：
class ThisIsAClass(object):
    passs
模块
    模块名全部小写，对于包内使用的模块，可以加一个下划线前缀，如
module.py
_internal_module.py

包

    包的命名规范与模块相同。

缩写

    命名应当尽量使用全拼写的单词，缩写的情况有如下两种：
1）常用的缩写，如 XML、ID等，在命名时也应只大写首字母，如
class XmlParser(object):pass
2）命名中含有长单词，对某个单词进行缩写。这时应使用约定成俗的缩写方式，如去除元音、包含辅音的首字符等方式，例如：
function 缩写为 fn
text 缩写为 txt
object 缩写为 obj
count 缩写为 cnt
number 缩写为 num，等。
特定命名方式
    主要是指 __xxx__ 形式的系统保留字命名法。项目中也可以使用这种命名，它的意义在于这种形式的变量是只读的，这种形式的类成员函数尽量不要重载。如
class Base(object):
    def __init__(self, id, parent = None):
        self.__id__ = id
        self.__parent__ = parent
    def __message__(self, msgid):
        # …略
其中 __id__、__parent__ 和 __message__ 都采用了系统保留字命名法。

语句

import
    import 语句有以下几个原则需要遵守：
1） import 的次序，先 import Python 内置模块，再 import 第三方模块，最后 import 自己开发的项目中的其它模块；这几种模块中用空行分隔开来。
2）一条 import 语句 import 一个模块。
3）当从模块中 import 多个对象且超过一行时，使用如下断行法（此语法 py2.5 以上版本才支持）：
from module import (obj1, obj2, obj3, obj4,
                                    obj5, obj6)
4）不要使用 from module import *，除非是 import 常量定义模块或其它你确保不会出现命名空间冲突的模块。
赋值
    对于赋值语言，主要是不要做无谓的对齐，如：
a   = 1     # 这是一个行注释
variable  = 2     # 另一个行注释
fn   = callback_function  # 还是行注释
没有必要做这种对齐，原因有两点：一是这种对齐会打乱编程时的注意力，大脑要同时处理两件事（编程和对齐）；二是以后阅读和维护都很困难，因为人眼的横向视野很窄，把三个字段看成一行很困难，而且维护时要增加一个更长的变量名也会破坏对齐。直接这样写为佳：
a = 1 # 这是一个行注释
variable = 2 # 另一个行注释
fn = callback_function # 还是行注释
分枝和循环
    对于分枝和循环，有如下几点需要注意的：
1) 不要写成一行，如：
if !flg: pass 和 for i in xrange(10): print i都不是好代码，应写成
if !flg:
    pass
for i in xrange(10):
    print i
注：本文档中出现写成一行的例子是因为排版的原因，不得作为编码中不断行的依据。
2) 条件表达式的编写应该足够 pythonic，如以下形式的条件表达式是拙劣的：
if len(alist) != 0: do_something()
if alist != []: do_something()
if s != “”: do_something()
if var != None: do_something()
if var != False: do_something()
上面的语句应该写成：
if seq: do_somethin() # 注意，这里命名也更改了
if var: do_something()
3) 用得着的时候多使用循环语句的 else 分句，以简化代码。

已有代码

    对于项目中已有的代码，可能因为历史遗留原因不符合本规范，应当看作可以容忍的特例，允许存在；但不应在新的代码中延续旧的风格。
    对于第三方模块，可能不符合本规范，也应看作可以容忍的特例，允许存在；但不应在新的代码中使用第三方模块的风格。
    tab 与空格混用的缩进是不可容忍的，在运行项目时应使用 –t 或 –tt 选项排查这种可能性存在。出现混用的情况时，如果是公司开发的基础类库代码，应当通知类库维护人员修改；第三方模块则可以通过提交 patch 等方式敦促开发者修正问题。

已有风格

    开发人员往往在加入项目之前已经形成自有的编码风格，加入项目后应以本规范为准编写代码。特别是匈牙利命名法，因为带有类型信息，并不适合 Python 编程，不应在 Python 项目中应用。

seal 2010-05-14 02:59 发表评论

Django Google Maps form widget(转)

seal — Mon, 10 May 2010 19:42:00 GMT

We’re building a Django application at work, and I had to implement a way for a user to select a location on a map, in order to be able to save the exact coordinates and the zoom level the user chose.
I looked around to see if something similar had been developed, but I did not find anything so simple and useful as what I later developed, so for the sake of open source-ness I’m sharing it here with you.
It is implemented as a standard Django form widget so just do as any other Django widget, for example:

class PersonForm(forms.Form):

name = forms.CharField()

website = forms.URLField()

location = forms.CharField(widget=GMapInput(attrs={'width':'600', 'height':'400'}))

Here’s the code for the GMapInput class (sorry for the docs in Spanish, but I’m in Mexico):

from django.forms.widgets import Input

from django.utils.safestring import mark_safe

 

class GMapInput(Input):

"""

Widget para seleccionar un punto en un mapa de Google

"""

 

def render(self, name, value, attrs=None):

"""

Atributos extras:

- width: ancho del mapa en pixeles

- height: alto del mapa en pixeles

- center: latitud,longitud del punto central del mapa

- zoom: zoom inicial del mapa, 1 - 17

"""

 

final_attrs = self.build_attrs(attrs)

width = final_attrs['width'] if 'width' in final_attrs else '500'

height = final_attrs['height'] if 'height' in final_attrs else '300'

center = final_attrs['center'] if 'center' in final_attrs else '21.983801,-100.964355' # Centro de México

zoom = final_attrs['zoom'] if 'zoom' in final_attrs else '4' # Zoom amplio, se ve todo un país

 

widget = u'''





''' % { 'id': final_attrs['id'], 'center': center, 'zoom': zoom }

else:

values = value.partition('|')

 

widget = widget + u'''

%(id)s_map.setCenter(new GLatLng(%(coords)s), %(zoom)s);

%(id)s_marker = new GMarker(new GLatLng(%(coords)s), {draggable: true});

%(id)s_map.addOverlay(%(id)s_marker);

 

GEvent.addListener(%(id)s_marker, "dragend", %(id)s_updateField);

GEvent.addListener(%(id)s_map, "zoomend", %(id)s_updateField);

GEvent.addListener(%(id)s_map, "dblclick", function (overlay, latlng) { %(id)s_marker.setLatLng(latlng); %(id)s_updateField(); });

''' % { 'id': final_attrs['id'], 'coords': values[0], 'zoom': values[2] }

 

return mark_safe(widget)

Don’t forget to add the