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()
