We have around 50 projects configured in CruiseControl.net, and each is individually configured to send an email out on completion / change / whatever.
Is there any way of configuring the user / email address list at a higher (e.g. server) level? I've searched the documentation, and googled for any clues, but I can't find anything.
I need to update who gets sent emails, and updating 50 files isn't a sustainable or ideal solution!
Thanks.
I'm not sure this the exact case of yours, but might work for you too. I have separated all the email settings into separate file, and then use the include in project configuration.
So it looks sth like this:
File: EmailPublisher.xml
<cb:config-template xmlns:cb="urn:ccnet.config.builder">
<email from="notification#example.com" mailhost="mailhost.example.com" mailport="587" includeDetails="true" mailhostUsername="username"
mailhostPassword="mailPwd" useSSL="TRUE" replyto="admin#example.com">
<description>Sending email notifications</description>
<groups>
<group name="developers">
<notifications>
<notificationType>Failed</notificationType>
<notificationType>Fixed</notificationType>
</notifications>
</group>
<group name="buildmaster">
<notifications>
<notificationType>Always</notificationType>
</notifications>
</group>
</groups>
<converters>
<regexConverter find="$" replace="#example.com" />
</converters>
<modifierNotificationTypes>
<NotificationType>Failed</NotificationType>
<NotificationType>Fixed</NotificationType>
</modifierNotificationTypes>
</email>
</cb:config-template>
And then in the project configuration in publishers section:
<!-- email publisher -->
<cb:include href="EmailPublisher.xml"/>
The downside is that you need to have the same configuration for each project, or alternatively use several include files (which starts to cause the same issue again).
Anyway, it seems to work for our team rather well.
[EDIT]
You can also go one step further, and for example only include the users list from separate file, and have the notifications settings applied separately for project.
You could use variables/substitutions like this: $(admins), $(programmers), $(ProductAScrumMaster), $(ProductBEntireTeam), $(HolyCrapTheMachineFriedEmailEveryone) and the like. You can append or nest them as well if you need combinations.
This post might be more useful as an example - it has lots more detail: Using dynamic parameters in email publisher subjectSettings block with CruiseControl.Net
Related
The website that I'm working on uses cooluri to format the urls. In the case of news, the "parameters" column of the "link_cache" table stores the parameters like this:
a:3:{s:5:"cHash";s:32:"eea2db734d63b661abaab43d86fd3bb5";s:2:"id";s:5:"15503";s:18:"tx_ttnews[tt_news]";s:6:"142085";}
, or in a more readable way:
{
"cHash":"eea2db734d63b661abaab43d86fd3bb5",
"id":"15503",
"tx_ttnews[tt_news]":"142085"
}
My problem is, that there are lots of pages where these urls are inserted in content elements like in the example below, and ofcourse the cHash parameters don't match.
website.com/news/some-title/431731b3f9d391a54c9ee48467ca4bb4.html
Now because of this I get the following error message on the news single view page:
no news_id given
Is there a clean way to solve this issue? I was told that changing the links in the backend is not an option. Also, the links are very old, and the option "oldlinksvalidfor" is set to 365.
Your configuration goes in the block. Here is a configuration for Ext:CoolUri. This working fine for me (with tt_news) this will remove cHash from the URL. You have an additional configuration for managing caching and all.
<predefinedparts>
<part>
<parameter>no_cache</parameter>
</part>
<part>
<parameter>cHash</parameter>
</part>
<!-- common patterns -->
<!-- parts defined this way will be removed from URL -->
<part>
<parameter>paramC</parameter>
</part>
<!-- this will prefix a value with "prefix-". Cannot be localized.
-->
<part key="prefix-(.*)" regexp="1">
<parameter>paramD</parameter>
</part>
<part key="page-(.*)" regexp="1">
<parameter>array[k5]</parameter>
</part>
<!-- if parameter matches value, key will be added to URL
with mutliple values, use valuemaps
-->
<part key="thisWillAppearInUrl">
<parameter>paramE</parameter>
<value>ifParamEMatcesThisValue</value>
</part>
</predefinedparts>
Hope this will helpful to you.
Greetings!
I am currently writing shell code that pushes data from LDAP into ejabberd, concerning MUC rooms. The last step I need to do is add some MUC rooms to users' bookmarks so they are auto-joined in their clients. There does not seem to be a module for that like there is in Prosody.
So I assume I need to mnipulate the users' private XML storage, in particular the storage:bookmarks part. I can get all existing bookmarks with:
ejabberdctl private_get user host storage storage:bookmarks
Then, there is ejabberdctl private_set, but I do not really understand it. From what I get, it seems that I need to replace the entire storage element at once, with old and new entries merged together.
Is there another way to add conference sub-elements to the node, or add bookmarks in some other way?
I tried to use private_set, but it seems to break on spaces in the element string. I tried escaping them in all possible ways, but to no avail.
OK, I found out for sure that adding new conferences to bookmarks requires re-uploading the entire bookmark storage set. That means the correct way is using private_get as shown in the question, then modify the XML to add a new entry and then use private_set to re-upload all of it.
As to the issue with spaces: Erlang shell (that's what ejabberdctl is) needs another level of quoting with single quotes, so some XML would become "'<storage xmlns="storage:bookmarks"><conference jid=…'" and so on in a shell argument.
You can find a shell script that does this and much more with ejabberdctl here: https://www.teckids.org/gitweb/?p=verein.git;a=blob;f=sysadmin/scripts/teckids-ejmaint
There are two standards for MUC bookmarks in XMPP. The old standard, XEP-0049, uses Private XML storage that can be modified with the private_set command. But the more recent standard is to store the bookmarks in PEP: XEP-0223
Dominik George's answer works for the old standard; for the new PEP method you can use:
sudo ./ejabberdctl send_stanza user#domain.tld user#domain.tld '
<iq type="set" id="asdf">
<pubsub xmlns="http://jabber.org/protocol/pubsub">
<publish node="storage:bookmarks">
<item id="current">
<storage xmlns="storage:bookmarks">
<conference jid="room#conference.domain.tld" autojoin="true" name="name">
<nick>nick</nick>
</conference>
</storage>
</item>
</publish>
<publish-options>
<x xmlns="jabber:x:data" type="submit">
<field var="FORM_TYPE" type="hidden">
<value>http://jabber.org/protocol/pubsub#publish-options</value>
</field>
<field var="pubsub#access_model">
<value>whitelist</value>
</field>
</x>
</publish-options>
</pubsub>
</iq>
'
This command is also accessible using the REST API: https://docs.ejabberd.im/developer/ejabberd-api/admin-api/#send-stanza
i would like to know how to create a new repository in ATG. like what all steps are needed to be included? Do i need to create a properties file?
In order to create a new repository, You need to follow these steps if you want to create a repository that uses sql database in as the datastore.
Create a properties file
Create tables you want to map to the repository
Create a definition XML files to map repository item-descriptors and their properties to the respective tables.
MyRepository.properties
$class=atg.adapter.gsa.GSARepository
repositoryName=MyRepository
definitionFiles=atg/test/repositories/MyRepository.xml
XMLToolsFactory=/atg/dynamo/service/xml/XMLToolsFactory
transactionManager=/atg/dynamo/transaction/TransactionManager
idGenerator=/atg/dynamo/service/IdGenerator
dataSource=/atg/dynamo/service/jdbc/JTDataSource
lockManager=/atg/dynamo/service/ClientLockManager
The above properties files ensures that a new repository component is created and MyRepository.xml will be marked as its definition file.
The content of the MyRepository.xml file should be something like below...
MyRepository.xml
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE gsa-template SYSTEM "http://www.atg.com/dtds/gsa/gsa_1.0.dtd">
<gsa-template>
<header>
<name>New Repository creation</name>
<author>Jyothi Prasad Buddha</author>
</header>
<item-descriptor name="myRepo" cache-mode="simple">
<table name="my_repo" type="primary" id-column-names="id">
<!-- properties that may (or may not) be used as primary keys -->
<property name="name" data-types="String" />
<property name="age" data-types="int" />
</table>
</item-descriptor>
</gsa-template>
However you will have to create the the necessary tables before you start atg instance. The above xml files refers to a table named my_repo which has comlumns name and age.
Yes you need to create properties file's for more details you can read RepositoryGuide.pdf from Oracle.
for more detail you can read below blog
http://immuraliraj.blogspot.in/2011/12/atg-repository-basic-concepts.html
Just type
"ATG Creating a repository" on Google
and you would get lot of relevant results on the first page itself. Also be specific which version of ATG are you using, and what you have already done/found in your research. (just a suggestion, so you get good responses).
Yes, you need a properties file when creating a custom repository component. There are a lot of blogs that answer your question with detailed steps.
Check this for example
I had some errors about invalid definitions coming from Visual Studio 2010 when trying to call a WSDL defined function. The problem was that you cannot use the same message definition in two seperate functions. So I have to create multiple message definitions while they do the same.
For instance:
<message name="Hi">
<part name="input" type="xsd:string">
</message>
<message name="Say_hi_back">
<part name="return" type="xsd:string">
</message>
<message name="I_hate_you">
<part name="return" type="xsd:string">
</message>
<portType name="DataPort">
<operation name="sayHello">
<input message="tns:Hi"/>
<output message="tns:Say_hi_back"/>
</operation>
<operation name="sayIHateYou">
<input message="tns:Hi"/>
<output message="tns:I_hate_you"/>
</operation>
</portType>
Now calling either one of the functions will give you an error. Unless I add a Hi2 with exactly the same parts and change one of the input messages in the operation definitions to tns:Hi2.
Why is this? It makes no sense. I'm building up a service where I'm going to have to add the customerID to all the functions I'm going to build. One function for getting the appointments, one for the payments, one for all. This means I'm going to have to copy the message definition like 10 times and name them getCustomerID*N*.
Also a lot of times I'm going to have to have multiple input parameters. Say for instance someone wants to have all appointments between date x and date y. (And this goes for all the information that is stored like payments etc.) While I only need one message with an int, a date and a date. I'm going to have to write a huge document.
So my question is if there is any other way to do this. I've only been working with WSDL for two days and those were two days full of problems and deceiving 'victories'. Where you solve one problem only to find out that opened the gate to the next.
Thanks. :)
You are creating a WSDL reflecting an RPC style as evidenced by the 'type' attributes in the message part definitions. I am not entirely sure why this would cause a problem with VS, but the RPC style has gone out of vogue in favor of the document style (the modern versions of some tools have dropped support for RPC altogether).
You may have better results using the document style (document/literal/wrapped is our standard). You can read a little more about style differences here (http://www.ibm.com/developerworks/webservices/library/ws-whichwsdl/).
The changes required are not too complex and this site (http://wso2.org/library/knowledge-base/convert-rpc-encoded-wsdl-document-literal-wrapped-wsdl) gives some help, although I think the author flipped his rpc vs literal output definitions in the #Output message section.
I have a feature which is provisioning 1 document library and 2 custom lists. A folder is included for each list containing the schema.xml for that list. Each folder also contains the associated forms (AllItems, DispForm, EditForm, NewForm, etc.). Everything deploys/works correctly but it seems a little redundant having the same forms copied into each list's folder. There is nothing special about these lists - the are basically a default doc library/generic list with additional fields provided through new content types (derived from Item/Document).
As far as I can tell these forms are pretty generic. Are there pre-installed forms that I can reference from my list so I don't have to deploy all of these extra files? Is there any reason I would not want to do this?
Update - moving xml in comment to original question for readability:
<Forms>
<Form Type="DisplayForm" Url="Forms/DispForm.aspx" WebPartZoneID="Main"/>
<Form Type="EditForm" Url="Forms/EditForm.aspx" WebPartZoneID="Main"/>
<Form Type="NewForm" Url="Forms/Upload.aspx" WebPartZoneID="Main"/>
<Form Type="NewFormDialog" Path="EditDlg.htm">
....
There are virtual defaults that are used if you don't specify a concrete page.
All lists use these template defaults unless you use a tool like SharePoint designer to customize the page. Then the template is used to create the concrete page and you can customize the look for a particular list without affecting others.
For my custom definitions, I use
<List>
...
<MetaData>
...
<Forms>
<Form Type="DisplayForm" Url="DispForm.aspx" SetupPath="pages\form.aspx" WebPartZoneID="Main" />
<Form Type="EditForm" Url="EditForm.aspx" SetupPath="pages\form.aspx" WebPartZoneID="Main" />
<Form Type="NewForm" Url="NewForm.aspx" SetupPath="pages\form.aspx" WebPartZoneID="Main" />
</Forms>
</MetaData>
</List>
If you have no reason to customize the out of the box version of these forms, you can use the virtual form and not deploy copies.