Can you coordinate with Kelvin on this? He's got CLOUDSTACK-1276, and had previously mentioned this exact same thing (using the xml file vs. annotations).
On Thu, Feb 21, 2013 at 07:26:42PM +0530, Rohit Yadav wrote: > Just to add, we will have separate xmls for every kind of profile, I > can make every plugin have its own spring xml which is copied and > injected to the web.xml comments, suggestions, flames? > > Possible future experiments: > - Spring xmls artefacts based > - Have a way to fix AOP, don't add pointcuts on all classes, methods > - Make DI framework replaceable in CloudStack, I would go for Guice ;) > > Regards. > > On Thu, Feb 21, 2013 at 7:22 PM, Rohit Yadav <[email protected]> wrote: > > I'm experimenting with removing all @Components and moving beans into > > an xml that could be read by spring to do DI. > > This decreases the load time and > > > > Errors: > > - Abstract classes annotated as spring components > > - Annotations for project which are not included in either build > > profiles: usage, simulator, oss (default) and nonoss > > - Some project don't even build or are not part of any build > > artefacts, looks like the @Component annotation was done in a hurry :) > > - Interfaces being annotated as spring components > > - Prasanna, Abhi and I spoke with Spring gurus Vijay V and Vaibhav on > > how to fix this bloat, so in applicationContext.xml we do some aop > > stuff to have transaction stuff we get from @DB annotation and this > > needs fixing as it's applying the following for all classes and > > methods (dude bloat, run! :) > > > > <aop:config proxy-target-class="true"> > > <aop:aspect id="dbContextBuilder" ref="transactionContextBuilder"> > > <aop:pointcut id="captureAnyMethod" > > expression="execution(* *(..))" > > /> > > > > Methodology and results: > > I wrote a python program that would go through the whole source code > > and fix all java files (except test, awsapi) who have @Component and > > create bean rules in an xml file. Next I had to manually fix few > > things (see error above) and I got the mgmt server running, the load > > time had improved a lot (little faster now) after disabling > > autoscanning and just using the xml generated. I see the benefit of > > just using the xml and not having spring as a build time dependency > > (we're using @Component in our classes so :). > > > > This was an experiment mostly because a lot of folks are having spring > > memory issues, especially Prasanna's devcloud-ci cannot run and check > > our commits as memory kills dom0. So, if no one objects may I go ahead > > and commit it? For me the jvm took about 650MB real memory on osx. > > > > Right now I made it for all build profiles, nonoss being the superset, > > my experiment can be checked out from here: > > https://github.com/bhaisaab/cloudstack/tree/spring-annotation-to-xml > > > > Regards. > > > > The python program, in case anyone wants to replicate or is curious: > > (defaulter is just a file containing result of: file . | grep java$ > > > defaulter) > > f = open('defaulter', 'r') > > files = f.read().split('\n')[:-1] > > files = filter(lambda x: not x.startswith('./awsapi') and > > x.find('/test/') == -1, files) > > xmldata = "" > > bean = """ <bean id="%s" class="%s"/>\n""" > > xml = """<!-- > > Licensed to the Apache Software Foundation (ASF) under one > > or more contributor license agreements. See the NOTICE file > > distributed with this work for additional information > > regarding copyright ownership. The ASF licenses this file > > to you under the Apache License, Version 2.0 (the > > "License"); you may not use this file except in compliance > > with the License. You may obtain a copy of the License at > > > > http://www.apache.org/licenses/LICENSE-2.0 > > > > Unless required by applicable law or agreed to in writing, > > software distributed under the License is distributed on an > > "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY > > KIND, either express or implied. See the License for the > > specific language governing permissions and limitations > > under the License. > > --> > > <beans xmlns="http://www.springframework.org/schema/beans"; > > xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"; > > xmlns:context="http://www.springframework.org/schema/context"; > > xmlns:tx="http://www.springframework.org/schema/tx"; > > xmlns:aop="http://www.springframework.org/schema/aop"; > > xsi:schemaLocation="http://www.springframework.org/schema/beans > > > > http://www.springframework.org/schema/beans/spring-beans-3.0.xsd > > http://www.springframework.org/schema/tx > > > > http://www.springframework.org/schema/tx/spring-tx-3.0.xsd > > http://www.springframework.org/schema/aop > > > > http://www.springframework.org/schema/aop/spring-aop-3.0.xsd > > http://www.springframework.org/schema/context > > > > http://www.springframework.org/schema/context/spring-context-3.0.xsd";> > > > > %s > > </beans> > > """ > > misses = [] > > for file in files: > > name = file.split('/')[-1].split('.')[0] > > name = name[0].lower() + name[1:] > > pkgname = "" > > idx = 0 > > paths = file.split('/')[1:] > > paths[-1] = paths[-1].split('.')[0] > > for str in paths: > > if str == 'org' or str == 'com': > > pkgname = '.'.join(paths[idx:]) > > idx += 1 > > f = open(file, 'r') > > data = f.read() > > f.close() > > idx = data.find('@Component') > > if idx == -1: > > misses.append(file) > > continue > > string = '@Component' > > if data[len(string)+idx] == '(': > > endidx = data.find(')', len(string)+idx) > > name = data[len(string)+idx+1:endidx].split('=')[-1].replace('"', '') > > name = name.replace("'", '') > > data = data[:idx] + data[endidx+2:] > > else: > > data = data.replace(string, '') > > data = data.replace('import > > org.springframework.stereotype.Component;\n', '') > > f = open(file, 'w') > > f.write(data) > > f.close() > > print name, pkgname > > xmldata += bean % (name, pkgname) > > xmldata += "\n" > > > > xml = xml % xmldata > > f = open('./client/tomcatconf/defaultComponentContext.xml.in', 'w') > > f.write(xml) > > f.close() > > > > f = open('misses', 'w') > > f.write('\n'.join(misses)) > > f.close() >
