Custom build process - get build parameters

I'm using TFS 2013 with build process template: TfvcTemplate.12.xaml. I created my own activity and I need to send it some input parameters that include the following data:
1. source path (for example: c:\a\src)
2. binaries path (for example c:\a\bin)
3. project name
I tried to find parameters in the 'WellKnownEnvironmentVariables' variable which can give me that values but with no luck.
the question is which parameters can I use in order to get these values??
T-O-M

Hi T-O-M,
Thia issue is related with TFS. You can refer to :
https://social.msdn.microsoft.com/Forums/en-US/5fd9c7db-7848-49a7-8802-464fc9fe0197/custom-process-parameters-in-tfs-builds?forum=tfsbuild
If you still have any concerns, please feel free to ask in TFS forum.
Best regards,
We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
Click
HERE to participate the survey.

Similar Messages

Custom Data Processing Extension and Parameters

Hello,
I have successfully implemented a custom data processing extension. Now I'm trying to add parameters, but the documentation is very weak on this topic. Also all examples of custam data processing extensions I have found so far are working without parameter support. Using a search engine I have found a custom data processing extension's sequence diagram during report execution. The related article also gives some hints about the IDbCommandAnalysis interface.
The GetParameters implementation creates a parameter collection based on the current command text and returns this collection. So far no problem.
My question is how to implement the IDbCommand's CreateParameter method and Parameters property? In CreateParameter I just return a new instance of my IDataParameter implementation, and the Parameters property returns a new parameter collection based on the current command text.
Is this ok so? At least this is the first implementation which seems to work in the data designer. If I share an instance variable for GetParameters and Parameters I will either get all parameters twice or BIDS will crash when pressing the execute button in the dataset designer - depending on my actual implementation.
What I'm still missing is how to get the parameters added automaticall to the dataset's properties parameter register?
thanx in advance and kind regards, Gerald

I think that finally (after some days of trying) I could make it work. The dataset and report parameters are reflecting the command text's parameters now (BTW: to answer my last question - the dataset's properties are updated after pressing the refresh button only). The road to success was not to use the same member variable for the collections returned by IDbCommandAnalysis:GetParameters() and IDbCommand:Parameter and not to fill the latter one with the command text's parameters (see partial source code below).
public sealed class MyCommand : IDbCommand, IDbCommandAnalysis
    private string m_commandText;
    private MyParameterCollection m_parameters = new MyParameterCollection();
    public string CommandText
        get { return this.m_commandText; }
        set { SetCommandText(value); }
    public IDataParameterCollection Parameters
        get { return m_parameters; }
    public IDataParameter CreateParameter()
        return new MyParameter();
    public IDataParameterCollection GetParameters()
        return GetParameterCollection();
    private void SetCommandText(string commandText)
        m_commandText = commandText;
    private MyParameterCollection GetParameterCollection()
        // ... create parameter collection based on m_commandText
Still there are some open questions:
How can I update the report parameter's properties? Now the parameter's data type is always string, but I'd like to assign the right data type. I'd also like to tag the parameters as hidden and supply a non-queried default value. Is this possible? How?
Although I my implementation of IDbCommandAnalysis:GetParameters() returns a collection containing the parameter names and values the "prompt query parameters" dialog (after pressing the query designer's execute button in the toolbar) shows a "<blank>" value for the parameters. What could be the problem?
Thanx in advance and regards,
Gerald

[SOLVED]Awesome build process gets stuck

Hi there Archers,
When I try to build the latest version of Awesome it manages to install all the dependencies, but when it comes to actually building awesome it consistently gets stuck on the following line:
Scanning dependencies of target man
[ 57%] Generating manpages/man1/awesome.1.xml
[ 57%] Generating manpages/man1/awesome.1
It stays there indefinitely.
On the AUR-page for this package I got the advice that running makepkg with root privileges might solve the problem, but I don't think this is a proper solution. Posts from Bionic_Apple seem to have some connection to my problem.
With my limited understanding of the make files inside the build directory, I've concluded that it all looks OK, but that doesn't stop it from freezing half-way through the process.
Any ideas on how I could make the install finish that doesn't involve root privileges?
Last edited by linduxed (2012-01-19 00:48:12)

Solved by building as root in a chroot environment.

XOQ-01703: error during parse of build process script: (clear, compile)", "

Hi All ,
I was just clearing the dimensions using the below in SQL commandline throwing below error . However this works fine with my another 5 dimensions .Any advice ?
SQL> exec dbms_cube.build('POSITION using (clear, compile)');
BEGIN dbms_cube.build('POSITION using (clear, compile)'); END;
ERROR at line 1:
ORA-37162: OLAP error
XOQ-01703: error during parse of build process script: "BUILD POSITION using
(clear, compile)", ""
ORA-06512: at "SYS.DBMS_CUBE", line 236
ORA-06512: at "SYS.DBMS_CUBE", line 280
ORA-06512: at line 1
Thanks,
DxP

I believe in this case the parser that is parsing the build process script is treating the dimension name "POSITION" as a keyword and not as an OLAP Cube Dimension name. You may wish to rename the dimension to a name other than "POSITION" or double quotes as specified previously.
Edited by: user713408 on Aug 1, 2011 12:12 PM

[svn] 3347: -remove the javascript directory since it is getting cleaned and recreated during the build process .

Revision: 3347
Author: [email protected]
Date: 2008-09-25 06:27:28 -0700 (Thu, 25 Sep 2008)
Log Message:
-remove the javascript directory since it is getting cleaned and recreated during the build process.
-this should make SVN happy and get the builds working again
Removed Paths:
flex/sdk/trunk/frameworks/javascript/FABridge/samples/srcview/source/fabridge/javascript/

- Where? That depends upon what you what. But, going by the letter of what you asked, you shouldn't deploy any applications or packages if you want them on your newly images systems, you should make them a part of the TS using Install Software and Install
Application tasks.
- How many times? You should only need one although depending upon what you are deploying and if you are using offline updates, some folks add more. I typically use two one right after the Setup Windows and ConfigMgr task (just like the task sequence wizard
builds for you) to update the OS and "things" in the image and one at the end to catch additional updates for apps and components added during the TS. Note that you also have to initiate an update scan cycle for additional install update tasks to work. Do
this by adding a run-command line task before the second Install Updates task with the proper WMIC incantation.
- Offline Updates only injects CBS updates into the image which are typically just core OS updates. You should also update your image though to include non-CBS updates. Using a build and catpure task sequence makes this easy.
Jason | http://blog.configmgrftw.com

[svn:fx-trunk] 11546: fix so the textLayout. swc does not get modified during the build process.

Revision: 11546
Author:   [email protected]
Date:     2009-11-06 15:42:05 -0800 (Fri, 06 Nov 2009)
Log Message:
fix so the textLayout.swc does not get modified during the build process.
QE notes: no
Doc notes: no
Bugs: no
Reviewer:
Tests run: checkintests
Is noteworthy for integration: no
Modified Paths:
    flex/sdk/trunk/build.xml

Obtaining custom HTTP GET / POST parameters of variable length in Application Process

Hi, I'm trying to connect a javascript UI control within my page to an APEX Application Process.
The control calls the application process via AJAX and appends a variable number of GET / POST parameters to its URL.
What is the best way to obtain these parameters from within the PLSQL procedure of the process? Or is there a better way to connect my javascript AJAX control to the Database behind my APEX app?
Thanks,
Steffi

Steffi,
I'm using APEX 4.1 and I've not used Treegrid before. It does look pretty interesting.
Firstly, I'm not sure if you are aware but GET requests aren't as secure as POST requests. GET simply appends a key/value pair to the server as part of the url which can be easily sniffed. POST request on the other hand sends the data as key/value pairs to the server in the header which makes it harder to intercept. This is why APEX outputs forms with POST methods.
Secondly, back to your original question.I'm not sure this is possible with an Application Process. In the passed I've used POST methods to pass values to my Application Process. I have some code using native jQuery if you would like to see how its done. Alternatively, if you convert your application process to a standard pl/sql procedure (i.e SQL Workshop>Object Browser>Procedures) you can call this procedure and pass in values as you would with any PHP page.
Simple example (untested)
CREATE OR REPLACE PROCEDURE "foo" ( my_name varchar(20)
     , count NUMBER)
IS
     p_name varchar(20) := my_name;
     p_count NUMBER     := count;
BEGIN
--Do what you want
--output something
   htp.p(p_count || '.: '|| p_name);
END;
The url on the page (or ajax call) would be something like "/pls/apex/{MY SCHEMA}.foo?my_name=Alistair&count=1"
Of course for text you will need to escape single colons and special characters etc.
I would really like to help you further so if you could create a working demo using a free workspace on http://www.apex.oracle.com and provide me with demo username/password. I can try and have a better look for you.
In all honesty, I'm sure you would be looking to create this as a plugin so that you can easily reuse it in various parts of your application and can easily update it or maintain it. I can help you with that as well if you want.
Alistair
P.S You might want to update your profile on this forum to provide your name so people dont just see "31742965-2d09-4f42-849f-e39eb2cfbc9e" your profile name

How do setup FlexUnit in your CI build process on 64-Bit Linux

Hello everyone. I am trying to setup FlexUnit in my build process and am using Rackspace cloud servers for my build server. I only have the option of Linux 64 bit distros or Windows 32b/64b. I started setting up and configuring everything on Linux but ran into problems installing the required stand alone player for test execution using FlexUnit. There is no 64-bit linux compatitble version of the stand alone player. In configuring the system to use the browser plugin on the machine instead of the stand alone player for executing the tests, I then get a manual interaction component which doesn't fit well with an automated CI build.
Am I the only person out there trying to integrate FlexUnit into my CI process on a 64b Linux machine? Any pointers, tips, workarounds or general hypothetical ideas would be great!

@mbbender - Answers below:
> I have been able to get the Flash Plugin working in a browser on my 64b Ubuntu 10.4 system. However, from my understanding I can not use the flash plugin from a browser to properly execute a headless execution of my tests during the CI process. To execute the tests as part of the CI process I need the Flash Stand Alone Player.
-- As of FlexUnit 4.1-beta2, you have the option of executing your tests using the stand-alone debug Flash Player using the default mechanism to launch that player or using a custom command, which could potentially point to the browser. Check out the docs on the Ant task @ http://docs.flexunit.org/index.php?title=Ant_Task#Usage and the sample CI build using a browser @ http://github.com/flexunit/flexunit/blob/master/FlexUnit4SampleCIProject/build.browser.xml for more details on how to do this.
> Are you saying that I can somehow use the Flash Plugin dependencies to install the Flash Stand Alone Player? If so, would I then need a 32b emulator wrapper for running the Stand Alone Player? Or, are you suggesting that I can somehow use the Flash Plugin in the browser to do what I need to execute my tests during the Hudson build?
-- Although I'm not a savvy Linux admin, it's my understanding that recent 64-bit Linux distros can execute 32-bit applications as well as 64-bit applications. The only req I'm aware of is that the 32-bit version of the dependencies must be available. In this case, executing the 32-bit version of the SA player should work from what I understand.
> Do you have a CI build process setup on a Linux 64b CentOS system with FlexUnit? To be honest, I'm more concerned with the end result right now than learning how to do it on my own and would like to know if you can potentially help me get this done a little quicker if you are interested.
-- I currently have a setup using 64-bit CentOS and running the 32-bit SA player, although it's not on Adobe's proper list of supported OSes. We've worked with other folks who have the 32-bit SA player running with FlexUnit on Ubuntu 64-bit. There was some feedback on some issues using older versions of openSUSE, but I don't know too many details re: the problem.
Hope this helps.
-Brian

What I learned in the mortgage and build process

What I learned in my 8 month loan and new build process. -don't sweat the small stuff -no news is good news. If you send in the requested docs, and you don't hear back quickly, than all is good. Understand the farther out from closing you are, the deeper in the stack your file is. You are not the only client. At any one time the money guy has a closing on the horizon that will have priority, if your new build is due to finish in 4 more months, well guess what? All I asked was when MY closing happens in 4 days to have my file at the top, and I was reassured that would be the case. -In my situation the lender, sales, and builder all came under one umbrella. There are advantages and disadvantages of this scenario. I read the horror stories no one was looking out for me, I was going to get roasted etc. I had none of that. They were very professional. In the end I have a great home, and they made some money. Its called business. Because all worked for the same company they were committed to getting me in this house. All 3 were responsive, and each knew what the other 2 were doing. I was quickly preapproved, they only wanted 500 down in earnest, and they broke ground within 30 days after the contract was signed. They paid ALL closing costs, over 12,0000 dollars worth. The house is beautiful and well built. I've had a few minor issues (ALL new houses do) that were promptly addressed and satisfied. -My way to a new house is the middle method. 1st you pick out a spec house in the community and just move in. Works for most people. 2nd you do like I did. Pick a plan from the builder for that community. You get to pick out colors, options etc. The price given is gonna be the basic house, check that list carefully. In my case the house had many "Upgraded" options as standard. As a incentive I was given another 8,000 to use at the design center. Well we spent that and another 7 LOL. Realize it does not take long to spend 10k at the design center. 3rd option is the true custom build. You buy the lot, get the architect to draw up precisely what you want, and go find the builder, and a construction loan. You have to know what you're doing if you go this way. Its a lot of work, and I entertained no notion of this option. I mean you're gonna spend hours (days?) in Lowe's looking at door knobs, backspaslh, lights. In my route the designers had all that narrowed down for ya, so instead of looking at hundreds of tiles you chose from about 30. I liked that. Understand the builder has vendors for all that stuff, and buy it at a good price, and why they were able to build me a pretty good house for the price. -Stuff is gonna happen. Docs will be misplaced, wrongly shredded, or deleted and have to be reexcuted. Weather will be bad=more delays. Suppliers get behind and can't get stuff to the job site. Don't let it ruin your day. Just resend the docs. -the people in the pipeline work best under pressure, they're let ya know when its time to sweat. -I kept a accordion file of the stuff I turned into the money people, I kept it on my desk for quick reference for months. It became a part of me. A month after closing it now sits on my bookcase. I look to it when I want to be reminded what it took to get me here. It truly represents almost 40 years of work and life. I think I'm gonna have it buried with me. I suggest you keep a file. -don't try to match wits with these guys. A couple of times I was asked for stuff that I thought had no revelance, but I did my duty and just got it up and sent the requested info. -They want you to buy THIS house. -Know the scrore going in=understand and know YOUR credit file and how it got there and how to get it better. It will make the process much easier. I really didn't do that because I didn't know anything. I was lucky, I had a clean file, my only problem was lack of revolving credit. Because I had 6 months to work it, I was advised to go out and pick 2 more cards. I did and my scores improved 25 points right away. -if you are going to engage collections on your file, do it before applying, or get some solid advice if you're gonna do it while in the process. -do not become a slave to the process. While all this going on, live your life. Do the stuff you like doing. When you sit around and wonder what they're gonna do about your driveway all the time you whip up all kinds of ugly scenarios. -if you have issues of confrontaion, controversey, and uncooperation among any of the 3, and its not gettin better, step back and take a look. -try to get good snacks at the closing. I wanted nachos but got Keebler oatmeal cookies and mtn dew. And thats all I can think of right now.

Which means tomorrow is when we get the new update for our iPhones and I just found the answer to my question, I don't care for the new iPods, just wanted to know when the update will be available
Message was edited by: Schmitty83

DVD Build Process Fails

I am using DVD Studio Pro 4.0.3 and I have attempted the Build process on my project several times but the progress bar never gets past 0%. But DVD Studio Pro never crashes or even hangs. What I have noticed, upon looking under the name column in the activity moniter is "DSPX_AEncoderServer (Not Responding)". Although I don't know much about this app, I assume it is essential to the build process.
Several months ago, I was able to burn a DVD from a simpler version of this project. Things that have NOT changed are that the Main Feature is about ten minutes long and also certain menus and screens are based on unrendered Motion files. What has changed is that two more unrendered Motion menus have been added and one of these has become very complex, containing several video windows such as you might see from a DVD Chapter Selection Screen.
I had assumed these Motion files would be rendered as part of the DVD Build as they were in my original version of this project. I assumed that was the best approach to avoid artifacts. The Motion files run quite sluggishly however and I am concerned that perhaps they are overloading the processor and causing the encode process to crash. Do you think that the best way to resolve this issue would be to first render these Motion files as Quicktime using the Animation Codec and then replace the Motion files with the Quicktime files, Before going through this, I just wanted another opinion as to whether there was anything I had overlooked.
Thank you so very much for any suggestions you may be able to offer.
Julianna

Thanks for your advice. I'm just not sure I understand what application to use to convert a Quicktime file into an m2v. Are you also saying that motion files aren't supposed to be directly imported into DVD SP? Would I choose the CODEC for DVD Best Quality? After encoding, do I move the old files and relink to the encoded files? I'm somewhat new at this, a little more explaination would be very helpful. Thanks so much.
Usuallly the best way in general is to encode
everything you can outside of DVD SP (Compressor,
bitVice, etc.) to m2vs and AC3s and then bring those
assets in - often stalls on builds can be cured by
that.
G5 2.5Ghz Dual Processor Mac OS X (10.4.8) 2GB RAM, 9800 Pro Video Card

Custom icon in application builder

I am trying to create an executable with custom icon using Application Builder in LabVIEW.
I converted a .png image to .ico for the icon, but when I select that .ico file as a custom icon, I get the following message:
"The selected file doesn't contain all of the needed icons. Without them, your application will show an incorrect icon in some views.
Missing icons:
32X32, 16 color
16X16, 16 color."
Any comments\suggestions on how to get custom icon would be appreciated.
Thanks.

It's possible that you don't have both 16x16 and 32x32 icons in 16 bit color. When you convert the .png to an .ico, see if there is an option to convert into necessary ico images that you need.
As a way to double check this without using application builder, load the VI : <vi.lib>\Platform\icon.llb\Read Icons for Application.vi
It will show you the different icons that are read from the .ico file.

Build process aborting due to following error

I should also probably add that machines usually build fine, it's only been with this new batch of PCs that the build has been failing.

So I'm trying to build a machine over the network, (the operating system I'm installing is Windows 7 Professional) however I'm getting this error message when building a machine from our new batch of PCs;
'Build process aborting due to following error:
Disk partitioning/formatting exception: VDS API
IVdsCreatePartitionEz::CreatePartitionEx returned 0x80042571 size = 45788 MB,
offset = 1 MB, format = 39'
Any suggestions greatly appreciated.
This topic first appeared in the Spiceworks Community

Properties Files during the build process.

Hi everyone, I have a question about files inside the src directory and how they relate to files in the /WEB-INF/classess directory.
From what I've read, eclipse copies the files inside "src" into "WEB-INF/classes" when building a project.
Inside my "src" folder I have a few properties files for my application, let's call them qe.properties, production.properties and final.properties
When I build my app (using Ant) I currently copy qe.properties into final.properties when building my application in QE. And production.properties into final.properties when building my application for production.
Here is the command:
<copy file = "src/qe.properties" tofile="src/final.properties" overwrite="true" />
Every once in a while a production build will have a qe properties file and vice versa (a qe build will have a production properties files).
I tried changing this using a symbolic link but I ran into the same issue, and worst yet, the file that get's copied inot "WEB-INF/classess" is not a symbolic link it simply contains the contents of the symbolic link.
How can I get around this issue?

I replaced the files with new files they are in the same location but they have different names. It still does not work.
It's very frusterating.
Everytime I make ANY changes to my original files I need to rebuild in encore. For example, I have a dynamic link and I adjust a transition and then resave. I have to build in encore from scratch (it will not build).
When encore works if I just save my encore file, close and then reopen it does not work anymore.
If I render the files and change anything in encore it does not work.
BTW the preview works perfectly. It just freezes during the Build Process.
I also tried to build an image and this did not work (it froze too).
I'm assuming CS5 works with intel i7 and windows 7 and 8GB of Ram per the Adobe requirements.
Any other ideas (other than leave every instance of Encore up and never close the program down)?
Jonathan

Log File for Build Process?

Is there a log file that gets written to while a project is
getting built to a particular target? I'm testing my X5 project
inside RH 6 Trial Version, and it's crashing about midway through
the Updating Files step of building a HTML Help (CHM) file.
The sample projects that come with RH 6 compile fine, so I'm
guessing there's a problem with my project it doesn't like; I'm
trying to track down what exactly.
A log file would be much faster than my current approach
which is to mark folders as Print only until it builds without
error, then slowly removing the Print only conditional and
rebuilding to find the problem...
I know a log is created during the build process in the
Output tab inside RH, but when the app crashes I can't get to it to
view it. Is this information stored externally somewhere?

I looked in the temporary output folder inside the
appropriate subdirectory in !SSL! but it looks like all the files
are there. In any case, I'm fairly certain I'm experiencing the
same problem Peter mentioned in his post about the conditional
build tags and merged tables. I did a build without any conditional
build tags and things worked fine. With conditional build tags I
still get the crash.
Conditional
Build Tags and Merged Table Cells Bug

Determining previous TFS 2013/2013 Build Process Template

I am looking at previous builds in TFS 2012 using a custom Build Process Template, and while it is easy to determine what template will be used for my Next build, how do I determine what was used on a previous build? Example: TeamA builds ProjectA
using TemplateA. Then they say something was not correct, and they switch to a new TemplateB. I can see TemplateB in their configuration, but need to make changes to TemplateA (it has custom hooks already in place for other things that are needed).
How do I determine what TemplateA was? It is not in any of the ActivityLogs or in their Build Logs. Just for the sake of argument, assume that there are 100s of build process templates, not easy to determine which was originally used.

Hi Rob,
Thanks for your post.
Yes, the build template name information not be included/generated in TFS build log or
ActivityLogs. If there’s something was not correct in TemplateA, we usually edit on this TemplateA and save it TFS Source Control, then reusing it in build definition.
As far as I know there’s no default way to determine which previous build templates be used in one build definition before. For this scenario, please
submit it to User Voice site at: http://visualstudio.uservoice.com/forums/121579-visual-studio. Microsoft engineers will evaluate them seriously.
We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
Click
HERE to participate the survey.

Custom build process - get build parameters

Similar Messages

Maybe you are looking for