Odoo Connects Everything | 欧度连接一切

odoo muilt domain dbfilter many to one config and code

odoo支持dbfilter配置用于实现多域名对应多账套，dbfilter是一个python标准正则字符串，并支持特殊的%h、%d这两个动态参数

根据源代码剖析（odoo8、odoo9）
openerp/http.py

def db_filter(dbs, httprequest=None):
    httprequest = httprequest or request.httprequest
    #请求域名去除端口号，如果有的话
    h = httprequest.environ.get('HTTP_HOST', '').split(':')[0]
    #分割域名取主机名，即多级域名最左边段
    d, _, r = h.partition('.')
    #如主机名为www且有下一段取第二段为主机名
    if d == "www" and r:
        d = r.partition('.')[0]
    #动态正则支持，%h替换为域名，%d替换为主机名
    r = openerp.tools.config['dbfilter'].replace('%h', h).replace('%d', d)
    #正则筛选匹配的数据库列表   
    dbs = [i for i in dbs if re.match(r, i)]
    return dbs

def db_filter(dbs, httprequest=None):

httprequest = httprequest or request.httprequest

#请求域名去除端口号，如果有的话

h = httprequest.environ.get('HTTP_HOST', '').split(':')[0]

#分割域名取主机名，即多级域名最左边段

d, _, r = h.partition('.')

#如主机名为www且有下一段取第二段为主机名

if d == "www" and r:

d = r.partition('.')[0]

#动态正则支持，%h替换为域名，%d替换为主机名

r = openerp.tools.config['dbfilter'].replace('%h', h).replace('%d', d)

#正则筛选匹配的数据库列表

dbs = [i for i in dbs if re.match(r, i)]

return dbs

由此可见，绝大部分有规律的映射需求都可以得到满足。惟有那些情况特殊，完全没有规则可言的：比如SEO需要、历史数据库命名混乱、甚至多个一级域名只对一个数据库的需求就悲剧了

这里举一个理想的openerp-server.conf扩展配置特例

dbfilter = %m
dbfilter_zhunlian.net = www-zhunlian-net
dbfilter_www.zhunlian.net = www-zhunlian-net
dbfilter_wap.zhunlian.net = www-zhunlian-net
dbfilter_higk.com = www-higk-com
dbfilter_www.higk.com = www-higk-com
dbfilter_wap.higk.com = www-higk-com
dbfilter_higk.com.cn = www-higk-com
dbfilter_www.higk.com.cn = www-higk-com
dbfilter_higk.net = www-higk-com
dbfilter_www.higk.net = www-higk-com

dbfilter = %m

dbfilter_zhunlian.net = www-zhunlian-net

dbfilter_www.zhunlian.net = www-zhunlian-net

dbfilter_wap.zhunlian.net = www-zhunlian-net

dbfilter_higk.com = www-higk-com

dbfilter_www.higk.com = www-higk-com

dbfilter_wap.higk.com = www-higk-com

dbfilter_higk.com.cn = www-higk-com

dbfilter_www.higk.com.cn = www-higk-com

dbfilter_higk.net = www-higk-com

dbfilter_www.higk.net = www-higk-com

首先新增支持%m特别参数，用来开启域名数据库映射复杂模式
其次新增支持dbfilter_前缀的域名映射独立配置指定数据库名称（www-zhunlian-net、www-higk-com）
最后迫不得已新增插入两行精炼的源代码

def db_filter(dbs, httprequest=None):
    httprequest = httprequest or request.httprequest
    h = httprequest.environ.get('HTTP_HOST', '').split(':')[0]
    d, _, r = h.partition('.')
    if d == "www" and r:
        d = r.partition('.')[0]
    r = openerp.tools.config['dbfilter'].replace('%h', h).replace('%d', d)
    #开启域名数据库映射复杂模式
    if r == '%m':
        #尝试获取域名独立配置，如无指定则映射同名数据库名，相当于保底%h功能以减少标准配置
        r = openerp.tools.config.get('dbfilter_%s' % h, h)
    dbs = [i for i in dbs if re.match(r, i)]
    return dbs

def db_filter(dbs, httprequest=None):

httprequest = httprequest or request.httprequest

h = httprequest.environ.get('HTTP_HOST', '').split(':')[0]

d, _, r = h.partition('.')

if d == "www" and r:

d = r.partition('.')[0]

r = openerp.tools.config['dbfilter'].replace('%h', h).replace('%d', d)

#开启域名数据库映射复杂模式

if r == '%m':

#尝试获取域名独立配置，如无指定则映射同名数据库名，相当于保底%h功能以减少标准配置

r = openerp.tools.config.get('dbfilter_%s' % h, h)

dbs = [i for i in dbs if re.match(r, i)]

return dbs

注：一个域名必备的.号在正则表达式领域可用于匹配任意字符，所以域名全称作为正则也可以根据需要匹配出一组数据库列表，如域名renjie.me可同时匹配出数据库名为renjie.me、renjie-me、renjie@me等等相似名称的数据库

python ldap userPassword salt sha md5 encrypt

odoo自带的auth_ldap模块虽然实现了ldap统一账户的验证，但如果用户一旦修改密码就变成同名的本地用户。这是因为odoo默认先验证本地res.users信息，如果没有记录，或者有记录但密码字段为空才会去ldap服务器验证

根据其原理，可以在修改密码的时候对接上ldap的密码信息就完美了，但用modify指令写入的userPassword值是原文存储，即明文直接写入在服务器也是明文，这就需要在请求端预先加密成ldap标准的密文格式

网上看到有很多php的实现都是要写点代码什么的，照葫芦折腾几下才发现python的passlib库居然有现成拿来即用的函数

>>> from passlib.hash import ldap_salted_sha1 as ssha
>>> ssha.encrypt("test", salt_size=16)
'{SSHA}tiULq3HOY1XRLW+bjGkHe3awnZCHcE5JSam1tnaOce49B4Dw'

>>> from passlib.hash import ldap_salted_sha1 as ssha

>>> ssha.encrypt("test", salt_size=16)

'{SSHA}tiULq3HOY1XRLW+bjGkHe3awnZCHcE5JSam1tnaOce49B4Dw'

以此类推

>>> import passlib.hash
>>> dir(passlib.hash)
['__class__', '__delattr__', '__dict__', '__dir__', '__doc__', '__format__', '__getattr__', '__getattribute__', '__hash__', '__init__', '__module__', '__name__', '__new__', '__package__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', '__weakref__', 'apr_md5_crypt', 'atlassian_pbkdf2_sha1', 'bcrypt', 'bcrypt_sha256', 'bigcrypt', 'bsd_nthash', 'bsdi_crypt', 'cisco_pix', 'cisco_type7', 'crypt16', 'cta_pbkdf2_sha1', 'des_crypt', 'django_bcrypt', 'django_bcrypt_sha256', 'django_des_crypt', 'django_disabled', 'django_pbkdf2_sha1', 'django_pbkdf2_sha256', 'django_salted_md5', 'django_salted_sha1', 'dlitz_pbkdf2_sha1', 'fshp', 'grub_pbkdf2_sha512', 'hex_md4', 'hex_md5', 'hex_sha1', 'hex_sha256', 'hex_sha512', 'htdigest', 'ldap_bcrypt', 'ldap_bsdi_crypt', 'ldap_des_crypt', 'ldap_hex_md5', 'ldap_hex_sha1', 'ldap_md5', 'ldap_md5_crypt', 'ldap_pbkdf2_sha1', 'ldap_pbkdf2_sha256', 'ldap_pbkdf2_sha512', 'ldap_plaintext', 'ldap_salted_md5', 'ldap_salted_sha1', 'ldap_sha1', 'ldap_sha1_crypt', 'ldap_sha256_crypt', 'ldap_sha512_crypt', 'lmhash', 'md5_crypt', 'msdcc', 'msdcc2', 'mssql2000', 'mssql2005', 'mysql323', 'mysql41', 'nthash', 'oracle10', 'oracle11', 'pbkdf2_sha1', 'pbkdf2_sha256', 'pbkdf2_sha512', 'phpass', 'plaintext', 'postgres_md5', 'roundup_plaintext', 'scram', 'sha1_crypt', 'sha256_crypt', 'sha512_crypt', 'sun_md5_crypt', 'unix_disabled', 'unix_fallback']

>>> import passlib.hash

>>> dir(passlib.hash)

['__class__', '__delattr__', '__dict__', '__dir__', '__doc__', '__format__', '__getattr__', '__getattribute__', '__hash__', '__init__', '__module__', '__name__', '__new__', '__package__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', '__weakref__', 'apr_md5_crypt', 'atlassian_pbkdf2_sha1', 'bcrypt', 'bcrypt_sha256', 'bigcrypt', 'bsd_nthash', 'bsdi_crypt', 'cisco_pix', 'cisco_type7', 'crypt16', 'cta_pbkdf2_sha1', 'des_crypt', 'django_bcrypt', 'django_bcrypt_sha256', 'django_des_crypt', 'django_disabled', 'django_pbkdf2_sha1', 'django_pbkdf2_sha256', 'django_salted_md5', 'django_salted_sha1', 'dlitz_pbkdf2_sha1', 'fshp', 'grub_pbkdf2_sha512', 'hex_md4', 'hex_md5', 'hex_sha1', 'hex_sha256', 'hex_sha512', 'htdigest', 'ldap_bcrypt', 'ldap_bsdi_crypt', 'ldap_des_crypt', 'ldap_hex_md5', 'ldap_hex_sha1', 'ldap_md5', 'ldap_md5_crypt', 'ldap_pbkdf2_sha1', 'ldap_pbkdf2_sha256', 'ldap_pbkdf2_sha512', 'ldap_plaintext', 'ldap_salted_md5', 'ldap_salted_sha1', 'ldap_sha1', 'ldap_sha1_crypt', 'ldap_sha256_crypt', 'ldap_sha512_crypt', 'lmhash', 'md5_crypt', 'msdcc', 'msdcc2', 'mssql2000', 'mssql2005', 'mysql323', 'mysql41', 'nthash', 'oracle10', 'oracle11', 'pbkdf2_sha1', 'pbkdf2_sha256', 'pbkdf2_sha512', 'phpass', 'plaintext', 'postgres_md5', 'roundup_plaintext', 'scram', 'sha1_crypt', 'sha256_crypt', 'sha512_crypt', 'sun_md5_crypt', 'unix_disabled', 'unix_fallback']

其中ldap_bcrypt,ldap_bsdi_crypt,ldap_des_crypt,ldap_hex_md5,ldap_hex_sha1,ldap_md5,ldap_md5_crypt,ldap_pbkdf2_sha1,ldap_pbkdf2_sha256,ldap_pbkdf2_sha512,ldap_plaintext,ldap_salted_md5,ldap_salted_sha1,ldap_sha1,ldap_sha1_crypt,ldap_sha256_crypt,ldap_sha512_crypt等都是ldap加密相关的兄弟函数，可谓是应有尽有非常丰富，看来在odoo端直接集成phpLDAPadmin里各种类型的加密密码更新都是易事了

centos install flash player projector content debugger

众所周知，Adobe Flash Player 11.2将是支持Linux平台的最后一个版本，不过Google Chrome自带的Flash Player Plugin版本不受这个限制，但Adobe将继续为Flash Player 11.2 for Linux提供安全更新

Flash基础环境除了依附浏览器的Plugin版本外，还有一种Projector版本，即独立播放器程序。虽然插件版已经支持64位系统了，但播放器版本官方始终只提供32位程序包下载

下载地址（含debug版本）
http://www.adobe.com/support/flashplayer/debug_downloads.html

Linux Help（2011年11月版本）
https://helpx.adobe.com/flash-player/release-note/readme-flash-player-linux.html

文档目前只有FP9和FP10的依赖库列表说明，最新也是最后版本的FP11详情很遗憾还未更新
为了避免安装不必要的32位旧依赖库，可以通过启动时缺少库提示来按需安装，如：

[root@x1c flash]# ./flashplayer
./flashplayer: error while loading shared libraries: libgtk-x11-2.0.so.0: cannot open shared object file: No such file or directory

[root@x1c flash]# yum whatprovides libgtk-x11-2.0.so.0
gtk2-2.24.28-8.el7.i686 : The GIMP ToolKit (GTK+), a library for creating GUIs for X
源    ：base
匹配来源：
提供    ：libgtk-x11-2.0.so.0

[root@x1c flash]# yum install gtk2.i686 -y

[root@x1c flash]# ./flashplayer

./flashplayer: error while loading shared libraries: libgtk-x11-2.0.so.0: cannot open shared object file: No such file or directory

[root@x1c flash]# yum whatprovides libgtk-x11-2.0.so.0

gtk2-2.24.28-8.el7.i686 : The GIMP ToolKit (GTK+), a library for creating GUIs for X

源：base

匹配来源：

提供：libgtk-x11-2.0.so.0

[root@x1c flash]# yum install gtk2.i686 -y

由于每个人的基础环境异同需要补上的32位库也不尽相同，记得上回安装Android SDK时大规模安装过一次，可能有些依赖相同的都省了，本次最终更新列表如下：

glib2.i686
libX11.i686
libXext.i686
libXt.i686
libXcursor.i686
nss.i686(libssl3.so)
gtk2.i686(libgtk-x11-2.0.so.0)

glib2.i686

libX11.i686

libXext.i686

libXt.i686

libXcursor.i686

nss.i686(libssl3.so)

gtk2.i686(libgtk-x11-2.0.so.0)

yum完上述i686基础依赖库，剩下的就是运行时报错的解决了，相关过程如下

GTK默认加载adwaita主题警告：

(flashplayer:13285): Gtk-WARNING **: 无法在模块路径中找到主题引擎：“adwaita”

yum install adwaita-gtk2-theme.i686 -y

(flashplayer:13285): Gtk-WARNING **: 无法在模块路径中找到主题引擎：“adwaita”

yum install adwaita-gtk2-theme.i686 -y

GTK默认加载PackageKit模块提示：

Gtk-Message: Failed to load module "pk-gtk-module"

yum install PackageKit-gtk3-module.i686 -y

Gtk-Message: Failed to load module "pk-gtk-module"

yum install PackageKit-gtk3-module.i686 -y

GTK默认加载libcanberra模块提示：

Gtk-Message: Failed to load module "canberra-gtk-module"

yum install libcanberra-gtk2.i686 -y

Gtk-Message: Failed to load module "canberra-gtk-module"

yum install libcanberra-gtk2.i686 -y

运行时闪退：

(flashplayer:15762): GLib-GObject-WARNING **: instance with invalid (NULL) class pointer
(flashplayer:15762): GLib-GObject-CRITICAL **: g_signal_handlers_disconnect_matched: assertion 'G_TYPE_CHECK_INSTANCE (instance)' failed

yum install libcurl.i686 -y

(flashplayer:15762): GLib-GObject-WARNING **: instance with invalid (NULL) class pointer

(flashplayer:15762): GLib-GObject-CRITICAL **: g_signal_handlers_disconnect_matched: assertion 'G_TYPE_CHECK_INSTANCE (instance)' failed

yum install libcurl.i686 -y

运行时无声：

ALSA lib dlmisc.c:252:(snd1_dlobj_cache_get) Cannot open shared library /usr/lib/alsa-lib/libasound_module_pcm_pulse.so

yum install alsa-plugins-pulseaudio.i686 -y

ALSA lib dlmisc.c:252:(snd1_dlobj_cache_get) Cannot open shared library /usr/lib/alsa-lib/libasound_module_pcm_pulse.so

yum install alsa-plugins-pulseaudio.i686 -y

折腾完毕之后运行flashplayer或flashplayerdebugger一切顺利

freeswitch http https ws wss nginx domain default port config

现代H5浏览器产业链越来越丰富，http+websocket+webrtc+sip组合已经是一种非常成熟的web原生音视频通讯解决方案

FreeSWITCH是一个开源的电话软交换平台，早在SIP年代就非常有名，自从支持WebSocket传输sip信令后与Web结合玩法多了许多

WebRTC提供了音视频通讯的核心技术，包括音视频的采集、编解码、网络传输、显示等功能，并且还支持跨平台，但就是没有信令协议部分

Chrome、Firefox等浏览器为了从保护用户隐私绝对安全角度考虑，纷纷提高了Web原生应用访问摄像头、麦克风的门槛：如不是用https、wss安全协议则会有各种警告及确认甚至直接拒绝

WS、WSS其实就是HTTP、HTTPS的WebSocket协议名称，其使用方式及默认端口都完全一致

综上所述，如果都用同一个域名地址承载web sip相关服务方便用户使用，就需要在nginx上做一些优雅的配置

map $http_upgrade $connection_upgrade {
    default upgrade;
    ''      close;
}

server {
    listen       80;
    listen       443 ssl;
    root         /var/www/sip;

    proxy_read_timeout 3600;
    proxy_http_version 1.1;
    proxy_set_header Upgrade $http_upgrade;
    proxy_set_header Connection $connection_upgrade;

    location / {
        if ($scheme = 'http') {
            set $ws_port 5066;
        }
        if ($scheme = 'https') {
            set $ws_port 7443;
        }
        if ($http_upgrade = 'websocket') {
            proxy_pass    $scheme://$server_addr:$ws_port;
        }
    }
}

map $http_upgrade $connection_upgrade {

default upgrade;

'' close;

}

server {

listen 80;

listen 443 ssl;

root /var/www/sip;

proxy_read_timeout 3600;

proxy_http_version 1.1;

proxy_set_header Upgrade $http_upgrade;

proxy_set_header Connection $connection_upgrade;

location / {

if ($scheme = 'http') {

set $ws_port 5066;

}

if ($scheme = 'https') {

set $ws_port 7443;

}

if ($http_upgrade = 'websocket') {

proxy_pass $scheme://$server_addr:$ws_port;

}

其对应freeswitch相关端口配置片断如下

<!-- Internal SIP Profile -->
<X-PRE-PROCESS cmd="set" data="internal_sip_port=5060"/>
<!-- RTP port range -->
<param name="rtp-start-port" value="16384"/>
<param name="rtp-end-port" value="32768"/>
<!-- for sip over websocket support -->
<param name="ws-binding"  value=":5066"/>
<!-- for sip over secure websocket support -->
<param name="wss-binding" value=":7443"/>

<X-PRE-PROCESS cmd="set" data="internal_sip_port=5060"/>

注：
sip和rtp端口不经nginx直连，但需要防火墙做相应的白名单
ws5066端口和wss7443端口间接通过nginx的http端口和https端口反向代理提供服务
nginx默认的http80端口和https443端口又用来提供webrtc前端相关资源访问

阿里云ecs部署freeswitch默认event socket listen port 8021 ipv6问题

阿里云ecs服务器目前还不支持ipv6
freeswitch默认event_socket模块配置
/etc/freeswitch/autoload_configs/event_socket.conf.xml

<configuration name="event_socket.conf" description="Socket Client">
  <settings>
    <param name="nat-map" value="false"/>
    <param name="listen-ip" value="::"/>
    <param name="listen-port" value="8021"/>
    <param name="password" value="ClueCon"/>
    <!--<param name="apply-inbound-acl" value="loopback.auto"/>-->
    <!--<param name="stop-on-bind-error" value="true"/>-->
  </settings>
</configuration>

</settings>

</configuration>

里的<param name=”listen-ip” value=”::”/>是ipv6地址的表现形式，其等同于ipv4的0.0.0.0
这就造成服务启动后，由于无法绑定ipv6地址，也就不能通过fs_cli命令行进行控制台管理了
这就解释了同样的配置在本机一切正常，部署到阿里云上就无法管理

知道原理解决起来就简单了，将listen-ip的value根据实际需要改成127.0.0.1、0.0.0.0或内外网卡的ipv4地址即可

odoo addons xml develop debug auto update

odoo以开发者模式运行的话，能自动检测所有的addons目录的代码变动情况自动进行reload
这个机制对于python代码来说是够用的，但是对csv安全规则、xml数据及视图模版来说还需要更新模块才能使其生效

今天由于逐个测试安全规则条目，不断的手工更新模块不胜其烦，只好深入研究一下

odoo9开始已经将开发者模式依赖的pyinotify替换为watchdog，可能是后者的应用范围更广把，不仅提供python库，还附带shell工具

那就学习下这个叫watchmedo的命令行工具吧

(python)[srj@x1c odoo9]$ watchmedo --help
usage: watchmedo [-h] [--version]
                 
                 {tricks-from,tricks-generate-yaml,log,shell-command,auto-restart}
                 ...

positional arguments:
  {tricks-from,tricks-generate-yaml,log,shell-command,auto-restart}
    tricks-from         Subcommand to execute tricks from a tricks
                        configuration file. :param args: Command line argument
                        options.
    tricks-generate-yaml
                        Subcommand to generate Yaml configuration for tricks
                        named on the command line. :param args: Command line
                        argument options.
    log                 Subcommand to log file system events to the console.
                        :param args: Command line argument options.
    shell-command       Subcommand to execute shell commands in response to
                        file system events. :param args: Command line argument
                        options.
    auto-restart        Subcommand to start a long-running subprocess and
                        restart it on matched events. :param args: Command
                        line argument options.

optional arguments:
  -h, --help            show this help message and exit
  --version             show program's version number and exit

(python)[srj@x1c odoo9]$ watchmedo --help

usage: watchmedo [-h] [--version]

{tricks-from,tricks-generate-yaml,log,shell-command,auto-restart}

...

positional arguments:

{tricks-from,tricks-generate-yaml,log,shell-command,auto-restart}

tricks-from Subcommand to execute tricks from a tricks

configuration file. :param args: Command line argument

options.

tricks-generate-yaml

Subcommand to generate Yaml configuration for tricks

named on the command line. :param args: Command line

argument options.

log Subcommand to log file system events to the console.

:param args: Command line argument options.

shell-command Subcommand to execute shell commands in response to

file system events. :param args: Command line argument

options.

auto-restart Subcommand to start a long-running subprocess and

restart it on matched events. :param args: Command

line argument options.

optional arguments:

-h, --help show this help message and exit

--version show program's version number and exit

看到auto-restart参数后真是让人眼前一亮，根据提示继续学习二级参数

(python)[srj@x1c odoo9]$ watchmedo auto-restart -h
usage: watchmedo auto-restart [-h] [-d directory] [-p PATTERNS]
                              [-i IGNORE_PATTERNS] [-D] [-R]
                              [--interval TIMEOUT] [--signal SIGNAL]
                              [--kill-after KILL_AFTER]
                              command [arg [arg ...]]

    Subcommand to start a long-running subprocess and restart it
    on matched events.

    :param args:
        Command line argument options.
    

positional arguments:
  command               Long-running command to run in a subprocess.
  arg                   Command arguments. Note: Use -- before the command
                        arguments, otherwise watchmedo will try to interpret
                        them. (default: -)

optional arguments:
  -h, --help            show this help message and exit
  -d directory, --directory directory
                        Directory to watch. Use another -d or --directory
                        option for each directory. (default: -)
  -p PATTERNS, --pattern PATTERNS, --patterns PATTERNS
                        matches event paths with these patterns (separated by
                        ;). (default: '*')
  -i IGNORE_PATTERNS, --ignore-pattern IGNORE_PATTERNS, --ignore-patterns IGNORE_PATTERNS
                        ignores event paths with these patterns (separated by
                        ;). (default: '')
  -D, --ignore-directories
                        ignores events for directories (default: False)
  -R, --recursive       monitors the directories recursively (default: False)
  --interval TIMEOUT, --timeout TIMEOUT
                        use this as the polling interval/blocking timeout
                        (default: 1.0)
  --signal SIGNAL       stop the subprocess with this signal (default SIGINT)
                        (default: 'SIGINT')
  --kill-after KILL_AFTER
                        when stopping, kill the subprocess after the specified
                        timeout (default 10) (default: 10.0)

(python)[srj@x1c odoo9]$ watchmedo auto-restart -h

usage: watchmedo auto-restart [-h] [-d directory] [-p PATTERNS]

[-i IGNORE_PATTERNS] [-D] [-R]

[--interval TIMEOUT] [--signal SIGNAL]

[--kill-after KILL_AFTER]

command [arg [arg ...]]

Subcommand to start a long-running subprocess and restart it

on matched events.

:param args:

Command line argument options.

positional arguments:

command Long-running command to run in a subprocess.

arg Command arguments. Note: Use -- before the command

arguments, otherwise watchmedo will try to interpret

them. (default: -)

optional arguments:

-h, --help show this help message and exit

-d directory, --directory directory

Directory to watch. Use another -d or --directory

option for each directory. (default: -)

-p PATTERNS, --pattern PATTERNS, --patterns PATTERNS

matches event paths with these patterns (separated by

;). (default: '*')

-i IGNORE_PATTERNS, --ignore-pattern IGNORE_PATTERNS, --ignore-patterns IGNORE_PATTERNS

ignores event paths with these patterns (separated by

;). (default: '')

-D, --ignore-directories

ignores events for directories (default: False)

-R, --recursive monitors the directories recursively (default: False)

--interval TIMEOUT, --timeout TIMEOUT

use this as the polling interval/blocking timeout

(default: 1.0)

--signal SIGNAL stop the subprocess with this signal (default SIGINT)

(default: 'SIGINT')

--kill-after KILL_AFTER

when stopping, kill the subprocess after the specified

timeout (default 10) (default: 10.0)

真是太强大了，本以为还要写点小代码什么的，没想到几个选项就可以灵活组合出一个小型的监控工具

watchmedo auto-restart -R -p="*.py;*.xml;*.csv" -d /home/srj/Git/Odoo/custom_addons/account_simple/ -- ./odoo.py -c c.conf -d test9 -u account_simple

1	watchmedo auto-restart -R -p=".py;.xml;*.csv" -d /home/srj/Git/Odoo/custom_addons/account_simple/ -- ./odoo.py -c c.conf -d test9 -u account_simple

-R是目录递归监控
-p是文件模式匹配，主要是排除前端资源和pyc
-d是指定监控目录
–是watchmedo与odoo各自的命令行参数分割线

至此odoo的模块开发效率又提高了一个层次：）

centos odoo decoder jpeg not available

云服务器上传jpg图片提示IOError错误

  File "/usr/lib64/python2.7/site-packages/PIL/ImageFile.py", line 200, in load
    d = Image._getdecoder(self.mode, d, a, self.decoderconfig)
  File "/usr/lib64/python2.7/site-packages/PIL/Image.py", line 417, in _getdecoder
    raise IOError("decoder %s not available" % decoder_name)
IOError: decoder jpeg not available

File "/usr/lib64/python2.7/site-packages/PIL/ImageFile.py", line 200, in load

d = Image._getdecoder(self.mode, d, a, self.decoderconfig)

File "/usr/lib64/python2.7/site-packages/PIL/Image.py", line 417, in _getdecoder

raise IOError("decoder %s not available" % decoder_name)

IOError: decoder jpeg not available

初步判断是python pil库依赖的底层jpeg lib问题
但是本机正常无法重现，于是检查环境

[srj@x1c ~]$ yum list installed|grep jpeg
libjpeg-turbo.x86_64               1.2.90-5.el7                        @anaconda
libjpeg-turbo-devel.x86_64         1.2.90-5.el7                        @anaconda
openjpeg-libs.x86_64               1.5.1-10.el7                        @anaconda

[srj@x1c ~]$ yum list installed|grep jpeg

libjpeg-turbo.x86_64 1.2.90-5.el7 @anaconda

libjpeg-turbo-devel.x86_64 1.2.90-5.el7 @anaconda

openjpeg-libs.x86_64 1.5.1-10.el7 @anaconda

对比了下发现服务器环境没有libjpeg-turbo-devel.x86_64立马补装一个，重启odoo服务错误依旧，于是继续更新Pillow库

[root@demo ~]# pip install --upgrade Pillow
Collecting Pillow
  Downloading Pillow-3.1.0.tar.gz (9.9MB)
    100% |████████████████████████████████| 9.9MB 42kB/s 
Installing collected packages: Pillow
  Found existing installation: Pillow 2.7.0
    Uninstalling Pillow-2.7.0:
      Successfully uninstalled Pillow-2.7.0
  Running setup.py install for Pillow
Successfully installed Pillow-3.1.0

[root@demo ~]# pip install --upgrade Pillow

Collecting Pillow

Downloading Pillow-3.1.0.tar.gz (9.9MB)

100% |████████████████████████████████| 9.9MB 42kB/s

Installing collected packages: Pillow

Found existing installation: Pillow 2.7.0

Uninstalling Pillow-2.7.0:

Successfully uninstalled Pillow-2.7.0

Running setup.py install for Pillow

Successfully installed Pillow-3.1.0

重启odoo服务问题解决
最后回顾了下odoo requirements.txt：
odoo8 Pillow==2.5.1
odoo9 Pillow==2.7.0
其实所谓更新无非是重新安装编译一下Pillow库，没想到直接升级到3.1.0版本，希望没有什么副作用：）

odoo Many2one field create use custom view open

odoo联系人相关的信息全部存储在res.partner对象里，包括供应商、客户、员工等个人及公司相关的信息都可以归纳入内

partner_id = fields.Many2one('res.partner', required=True)

1	partner_id = fields.Many2one('res.partner', required=True)

二次开发经常会直接关联该对象可以少造很多轮子，但由于其新建联系人的默认视图view_partner_form是一个非常重且包含好多类型设置又被各种业务模块继承扩展过的大视图，这就需要为不同的应用场景开发对应业务信息的轻量级自定义视图

最开始我是从继承res.partner并重写其@api.model的fields_view_get方法来完成的

@api.model
def fields_view_get(self, view_id=None, view_type=False, toolbar=False, submenu=False):
    def get_view_id(name):
        view = self.env['ir.ui.view'].search([('name', '=', name)], limit=1)
        if not view:
            return None
        return view.id
    context_view_name = self._context.get('view_name')
    if context_view_name:
        ##重置context指定视图
        view_id = get_view_id(context_view_name)
    return super(res_partner, self).fields_view_get(
        view_id=view_id, view_type=view_type, toolbar=toolbar, submenu=submenu)

@api.model

def fields_view_get(self, view_id=None, view_type=False, toolbar=False, submenu=False):

def get_view_id(name):

view = self.env['ir.ui.view'].search([('name', '=', name)], limit=1)

if not view:

return None

return view.id

context_view_name = self._context.get('view_name')

if context_view_name:

##重置context指定视图

view_id = get_view_id(context_view_name)

return super(res_partner, self).fields_view_get(

view_id=view_id, view_type=view_type, toolbar=toolbar, submenu=submenu)

使用很简单view context新增一个view_name属性指定自定义视图名称即可

<field name="partner_id" context="{'view_name': 'res.partner.daybook.form'}"/>

1	<field name="partner_id" context="{'view_name': 'res.partner.daybook.form'}"/>

当后来对其它对象也有同样需求的时候就基本确定我悲剧的造了一个小轮子
因为这个需求是非常合理，如果要一个个对象去继承重写是非常不实现的，至少也要在共同的父类上做对接
根据这个思路从openerp.osv.Model一直寻找到openerp.models.Model的fields_view_get父级方法，其中有一段代码让人眼前一亮

# try to find a view_id if none provided
if not view_id:
    # <view_type>_view_ref in context can be used to overrride the default view
    view_ref_key = view_type + '_view_ref'
    view_ref = context.get(view_ref_key)
    if view_ref:
        if '.' in view_ref:
            module, view_ref = view_ref.split('.', 1)
            cr.execute("SELECT res_id FROM ir_model_data WHERE model='ir.ui.view' AND module=%s AND name=%s", (module, view_ref))
            view_ref_res = cr.fetchone()
            if view_ref_res:
                view_id = view_ref_res[0]

# try to find a view_id if none provided

if not view_id:

# <view_type>_view_ref in context can be used to overrride the default view

view_ref_key = view_type + '_view_ref'

view_ref = context.get(view_ref_key)

if view_ref:

if '.' in view_ref:

module, view_ref = view_ref.split('.', 1)

cr.execute("SELECT res_id FROM ir_model_data WHERE model='ir.ui.view' AND module=%s AND name=%s", (module, view_ref))

view_ref_res = cr.fetchone()

if view_ref_res:

view_id = view_ref_res[0]

这不就是传说中解决分类自选视图的相关代码：类型、模块名、视图ID确定一个视图
马上删除之前所有相关代码，view context按如下规则修改直接应用

<field name="partner_id" context="{'form_view_ref': '模块名.视图ID'}"/>

1	<field name="partner_id" context="{'form_view_ref': '模块名.视图ID'}"/>

至此，对odoo一切都是models的理念又有了进一步的理解

ECShop动静分离transport.js get jsonp跨域对接

ECShop transport.js定义了Ajax Transport

/* 定义两个别名 */
var Ajax = Transport;
Ajax.call = Transport.run;

/* 定义两个别名 */

var Ajax = Transport;

Ajax.call = Transport.run;

其中run方法实现了类似jQuery $.ajax的功能但缺少了对jsonp方式的支持

  /* *
  * 调用此方法发送HTTP请求。
  *
  * @public
  * @param   {string}    url             请求的URL地址
  * @param   {mix}       params          发送参数
  * @param   {Function}  callback        回调函数
  * @param   {string}    ransferMode     请求的方式，有"GET"和"POST"两种
  * @param   {string}    responseType    响应类型，有"JSON"、"XML"和"TEXT"三种
  * @param   {boolean}   asyn            是否异步请求的方式
  * @param   {boolean}   quiet           是否安静模式请求
  */
  run : function (url, params, callback, transferMode, responseType, asyn, quiet)

/* *

* 调用此方法发送HTTP请求。

* @public

* @param {string} url 请求的URL地址

* @param {mix} params 发送参数

* @param {Function} callback 回调函数

* @param {string} ransferMode 请求的方式，有"GET"和"POST"两种

* @param {string} responseType 响应类型，有"JSON"、"XML"和"TEXT"三种

* @param {boolean} asyn 是否异步请求的方式

* @param {boolean} quiet 是否安静模式请求

run : function (url, params, callback, transferMode, responseType, asyn, quiet)

可以在第一个get方法判断入口前新增下列代码：

if (transferMode === "GET")
{
  if(params.indexOf("?") == -1){
    if(params){
     params+= "&"; 
    }
    params += "callback=?";
  }
  return $.getJSON(url, params, callback);
}

if (transferMode === "GET")

{

if(params.indexOf("?") == -1){

if(params){

params+= "&";

}

params += "callback=?";

}

return $.getJSON(url, params, callback);

}

巧妙的将原来ajax get方法对接上jquery script jsonp实现
排除一些自制的特殊调用：比如callback的第二个参数以及this的深度使用
其它上层Ajax.call相关调用就不用一一更新了

Linux下反编译flash swf文件查看actionscript代码

今天有个需求参照物的关键代码封装在swf里
想了解下实现才想起来本地没有相应的反编译软件

虽说现在flash反编译软件多如牛毛，但支持多平台又免费开源的并不多
还好优秀软件不在多，让我发现了JPEXS Free Flash Decompiler

源码库：
https://github.com/jindrapetrik/jpexs-decompiler

官网：
https://www.free-decompiler.com/flash/
This product is completely FREE with no hidden fees. No premium version exists. No trial limitations. You get one version with all features included. No ads anywhere. You can look in app source code. That is fair, seriously. Don’t you think?

GitHub上可以看到，该软件更新维护非常活跃，今天下载的最新7.1.2版本是月初发布的，非常新鲜，就是国内下载好慢，最后去掉https后快了一些
http://www.free-decompiler.com/flash/download/ffdec_7.1.2.zip

Linux下依赖几乎没，就是对最新java1.8以上的版本要求稍稍高了些
下载完解压到/opt/ffdec，然后运行目录下的ffdec.sh即可，真正的开箱即用

JPEXS Free Flash Decompiler

原本以为就一命令行，没想到全程GUI操作还自带中文语言包切换下即可
那ActionScript都省的导出另外看，直接当作Linux下简易IDE，其基本的Run、Debug功能都集成好了