AW: Determining next steps for the NPanday
Hi Roman, even though it hurts I think retiring the project is the right thing to do. Not because of the quality of NPanday, but just because the (developer!) community is effectively gone. While I'm still using NPanday every day (1.5 snapshot), it is all on the same project and there is no need for further development currently. And just the fact that we haven't been able to release 1.5 for ages shows the lack of a functioning community. I might invest in releasing 1.5 - but I might also keep saying this the years to come. Although for releasing 1.5 we wouldn't need more than somebody just doing the release with how 1.5 is now. http://incubator.apache.org/npanday/docs/1.5.0-incubating-SNAPSHOT/developers/releasing.html So if someone volunteers - just go ahead. _ Lars -Ursprüngliche Nachricht- Von: shaposh...@gmail.com [mailto:shaposh...@gmail.com] Im Auftrag von Roman Shaposhnik Gesendet: Sonntag, 14. Dezember 2014 19:29 An: npanday-dev@incubator.apache.org Cc: npanday-us...@incubator.apache.org Betreff: Re: Determining next steps for the NPanday On Sun, Dec 14, 2014 at 2:08 AM, David Akehurst d...@akehurst.net wrote: I would like it to stay I have done some development for it Please note that retirement doesn't make source code go away -- it will still be available for you to use and hack on. Retirement is ASF's way of telling outside world that even though source is still there and available, the community is gone. Now, as I said, the measure of the community being there could be interpreted as *at the very least* having 3 active participants since all decisions would require 3 votes minimum. Not sure if NPanday has that, but would love to be contradicted. Thanks, Roman. P.S. Finally, should 3 active participants show up after the retirement, reviving the project takes very little time. Hence its not like a permanent state of things.
AW: Release 1.5.0 ??
I guess the deadline passed by :( :-) I hope to be able to use some time on npanday in march/april next year -Ursprüngliche Nachricht- Von: Konstantin Boudnik [mailto:c...@apache.org] Gesendet: Sonntag, 9. November 2014 06:44 An: npanday-dev@incubator.apache.org Betreff: Re: Release 1.5.0 ?? Guys, the report was due on 5th, so the shepherds can put on their notes by the end of tomorrow. Is anyone works on it? Thanks, Cos On Mon, Nov 03, 2014 at 03:41PM, Brett Porter wrote: I won't have any ability to do that, or a report, for at least two weeks (and lots of stuff going on the month after that). It'd be great if someone else can pick it up - it doesn't require being a committer. - Brett On 3 Nov 2014, at 5:56 am, Konstantin Boudnik c...@apache.org wrote: Any one in the community has cycles to do the pre-release docs review? I would be happy to, but I am not really familiar with the code to make sane documentation comments. Regards, Cos On Mon, Oct 20, 2014 at 03:02PM, Brett Porter wrote: Hi Konstantin, I don't think there's been any progress. I've been occupied on other projects for the last couple of months. It is ready to go other than a review of the docs which are a little stale. Has anyone else had a chance to look at the docs? Anyone want to try the release process? - Brett On 17 Oct 2014, at 3:23 pm, Konstantin Boudnik c...@apache.org wrote: Guys, are we getting anywhere closer to the 1.5.0 release? It seems to be a chunk of activity wrt preparation to it, which then stopped at some point. Can I help anyhow? Regards, Cos On Mon, Aug 25, 2014 at 02:53PM, Brett Porter wrote: I closed out or moved everything on the 1.5.0-incubating list last Friday - only thing I wanted to do next was take a pass over the getting started documentation and check that it is correct and easy to use. Does anyone have some cycles to do the same? It might be particularly helpful to get fresh eyes from a non-committer, and contributions to the documentation would be most welcome. - Brett On 24 Aug 2014, at 9:17 am, Konstantin Boudnik c...@apache.org wrote: I don't see why not. If the current state of the software is good enough to be released from the community standpoint - I'd go for it. But as always - it should be the PPMC, developers, and users call, of course. Cos On Thu, Aug 21, 2014 at 10:50AM, Lars Corneliussen // Zen wrote: I think we should just release it as it is and take it from there. The snapshot-version has been in use for quite a while now. Last issue to resolve would be https://issues.apache.org/jira/browse/NPANDAY-599 Brett, you talked about somehow caching the resolver per project. I think if that's done we're fine and can close 599 and thus 402. -UrsprЭngliche Nachricht- Von: Brett Porter [mailto:br...@porterclan.net] Im Auftrag von Brett Porter Gesendet: Donnerstag, 21. August 2014 02:43 An: npanday-dev@incubator.apache.org Betreff: Re: Release 1.5.0 ?? Thanks for the review and the nudge here Konstantin. I've just dropped a couple out, and will take a quick pass over the non-blockers and ones assigned to me to see what else can be closed or bumped. Do any others have a moment to look through them? Thanks, Brett On 19 Aug 2014, at 7:37 am, Konstantin Boudnik c...@apache.org wrote: Ladies and gentlemen. I've pocked around a little bit and it seems that there are only 4 tickets (of which only 2 are real blockers) are effectively standing between today and the next release of the project. One of the critical is also an improvement. Here's what I am referring to http://is.gd/WzgibS I am thinking if this would be a sane idea to deal with the blockers and push Major bugs into 1.6-incubating? This tactic will help to get the release out so the users have something new to work with. And it also will provide a new development baseline for the contributors? What do you guys think? Cos
[jira] [Commented] (NPANDAY-598) Problem: resolving PDBs in a reactor module makes them a dependency for subsequent modules
[ https://issues.apache.org/jira/browse/NPANDAY-598?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14108850#comment-14108850 ] Lars Corneliussen commented on NPANDAY-598: --- Thanks for the clarification. Maybe {{ResolutionContext}} or {{ResolvedAttachedArtifactsCollector}} would be better names. Problem: resolving PDBs in a reactor module makes them a dependency for subsequent modules -- Key: NPANDAY-598 URL: https://issues.apache.org/jira/browse/NPANDAY-598 Project: NPanday Issue Type: Sub-task Components: Maven Plugins Reporter: Lars Corneliussen Fix For: 1.5.0-incubating -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-599) Cache resolve-result reactor-wide to avoid pdb/gac/++ beeing resolved multiple times
[ https://issues.apache.org/jira/browse/NPANDAY-599?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14106741#comment-14106741 ] Lars Corneliussen commented on NPANDAY-599: --- So the wrong key made the cache mess up things between projects in the reactor? Cache resolve-result reactor-wide to avoid pdb/gac/++ beeing resolved multiple times Key: NPANDAY-599 URL: https://issues.apache.org/jira/browse/NPANDAY-599 Project: NPanday Issue Type: Sub-task Components: Maven Plugins Reporter: Lars Corneliussen Assignee: Brett Porter Fix For: 1.5.0-incubating Currently it seems like our custom contributors/resolvers execute for every possible path (artifact.getDependencyTrail() changes...). We can cache what we resolved reactor-wide in a separate component. Cachekey would be artifact.getId(), right? Should we let each contributor implement the cache (ArtifactResolvingContributor) - or should it be generically handled in the artifact resolver? -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-598) Problem: resolving PDBs in a reactor module makes them a dependency for subsequent modules
[ https://issues.apache.org/jira/browse/NPANDAY-598?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14106838#comment-14106838 ] Lars Corneliussen commented on NPANDAY-598: --- I'm confused. Problem: resolving PDBs in a reactor module makes them a dependency for subsequent modules -- Key: NPANDAY-598 URL: https://issues.apache.org/jira/browse/NPANDAY-598 Project: NPanday Issue Type: Sub-task Components: Maven Plugins Reporter: Lars Corneliussen Fix For: 1.5.0-incubating -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Updated] (NPANDAY-402) Add support to automatically attach and resolve PDB-symbols
[ https://issues.apache.org/jira/browse/NPANDAY-402?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Corneliussen updated NPANDAY-402: -- Summary: Add support to automatically attach and resolve PDB-symbols (was: Add support to automatically attach and resolve PDB-symbols and Code-Documentation (comment.xml) ) Add support to automatically attach and resolve PDB-symbols Key: NPANDAY-402 URL: https://issues.apache.org/jira/browse/NPANDAY-402 Project: NPanday Issue Type: Improvement Components: Maven Plugins Affects Versions: 1.4-incubating, 1.5.0-incubating Environment: $ mvn -v Apache Maven 2.2.1 (rdebian-4) Java version: 1.6.0_24 Java home: /usr/lib/jvm/java-6-sun-1.6.0.24/jre Default locale: en_US, platform encoding: UTF-8 OS name: linux version: 2.6.35-28-generic arch: amd64 Family: unix Reporter: John R. Fallows Assignee: Lars Corneliussen Priority: Critical Fix For: 1.5.0-incubating h2. ISSUES/TODO * Integration tests missing * impl for attaching comments.xml missing * created subtasks (/) h2. Original Description The comments.xml file generated during CSharp compilation needs to be shipped to developers for use in Visual Studio, similar to the javadoc artifact produced by Maven for Java projects. This is not currently attached as an artifact during the NPanday build for dotnet-library. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Created] (NPANDAY-634) Automatically attach (and resolve) VS Code Documentation files (comment.xml)
Lars Corneliussen created NPANDAY-634: - Summary: Automatically attach (and resolve) VS Code Documentation files (comment.xml) Key: NPANDAY-634 URL: https://issues.apache.org/jira/browse/NPANDAY-634 Project: NPanday Issue Type: Improvement Components: Maven Plugins Affects Versions: 1.5.0-incubating Reporter: Lars Corneliussen Priority: Minor Fix For: Backlog With NPANDAY-402 we introduced attaching and resolving PDBs. Also handling Code-Documentation files should be an quite easy addition. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Updated] (NPANDAY-402) Add support to automatically attach and resolve PDB-symbols
[ https://issues.apache.org/jira/browse/NPANDAY-402?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Corneliussen updated NPANDAY-402: -- Summary: Add support to automatically attach and resolve PDB-symbols (was: Add support to automatically attach and resolve PDB-symbols ) Add support to automatically attach and resolve PDB-symbols --- Key: NPANDAY-402 URL: https://issues.apache.org/jira/browse/NPANDAY-402 Project: NPanday Issue Type: Improvement Components: Maven Plugins Affects Versions: 1.4-incubating, 1.5.0-incubating Environment: $ mvn -v Apache Maven 2.2.1 (rdebian-4) Java version: 1.6.0_24 Java home: /usr/lib/jvm/java-6-sun-1.6.0.24/jre Default locale: en_US, platform encoding: UTF-8 OS name: linux version: 2.6.35-28-generic arch: amd64 Family: unix Reporter: John R. Fallows Assignee: Lars Corneliussen Priority: Critical Fix For: 1.5.0-incubating h2. ISSUES/TODO * Integration tests missing * impl for attaching comments.xml missing * created subtasks (/) h2. Original Description The comments.xml file generated during CSharp compilation needs to be shipped to developers for use in Visual Studio, similar to the javadoc artifact produced by Maven for Java projects. This is not currently attached as an artifact during the NPanday build for dotnet-library. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-599) Cache resolve-result reactor-wide to avoid pdb/gac/++ beeing resolved multiple times
[ https://issues.apache.org/jira/browse/NPANDAY-599?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14105206#comment-14105206 ] Lars Corneliussen commented on NPANDAY-599: --- We talked about making i a per-project somehow, didn't we? [code] [07.08.2014 12:04:04] Lars Corneliussen: iirc somehow the deps were cached between reactor projects [07.08.2014 12:04:34] Lars Corneliussen: but I can't remember what happend then... if it failed, or if it just included pdbs you wouldn't want... [07.08.2014 12:06:07] Brett Porter (NPanday team): can see why it could be a problem, maybe better just to have the cache be mapped per project instead [07.08.2014 12:09:19] Lars Corneliussen: if per-project is available, it seems much more approprieate than per-lookup1 [07.08.2014 12:22:07] Brett Porter (NPanday team): no per project lookup, but we can do that in the maps [code] Cache resolve-result reactor-wide to avoid pdb/gac/++ beeing resolved multiple times Key: NPANDAY-599 URL: https://issues.apache.org/jira/browse/NPANDAY-599 Project: NPanday Issue Type: Sub-task Components: Maven Plugins Reporter: Lars Corneliussen Fix For: 1.5.0-incubating Currently it seems like our custom contributors/resolvers execute for every possible path (artifact.getDependencyTrail() changes...). We can cache what we resolved reactor-wide in a separate component. Cachekey would be artifact.getId(), right? Should we let each contributor implement the cache (ArtifactResolvingContributor) - or should it be generically handled in the artifact resolver? -- This message was sent by Atlassian JIRA (v6.2#6252)
AW: Release 1.5.0 ??
I think we should just release it as it is and take it from there. The snapshot-version has been in use for quite a while now. Last issue to resolve would be https://issues.apache.org/jira/browse/NPANDAY-599 Brett, you talked about somehow caching the resolver per project. I think if that's done we're fine and can close 599 and thus 402. -Ursprüngliche Nachricht- Von: Brett Porter [mailto:br...@porterclan.net] Im Auftrag von Brett Porter Gesendet: Donnerstag, 21. August 2014 02:43 An: npanday-dev@incubator.apache.org Betreff: Re: Release 1.5.0 ?? Thanks for the review and the nudge here Konstantin. I've just dropped a couple out, and will take a quick pass over the non-blockers and ones assigned to me to see what else can be closed or bumped. Do any others have a moment to look through them? Thanks, Brett On 19 Aug 2014, at 7:37 am, Konstantin Boudnik c...@apache.org wrote: Ladies and gentlemen. I've pocked around a little bit and it seems that there are only 4 tickets (of which only 2 are real blockers) are effectively standing between today and the next release of the project. One of the critical is also an improvement. Here's what I am referring to http://is.gd/WzgibS I am thinking if this would be a sane idea to deal with the blockers and push Major bugs into 1.6-incubating? This tactic will help to get the release out so the users have something new to work with. And it also will provide a new development baseline for the contributors? What do you guys think? Cos
[jira] [Assigned] (NPANDAY-454) maven-compile-plugin: includeSources section has issues with the path delimiter
[ https://issues.apache.org/jira/browse/NPANDAY-454?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Corneliussen reassigned NPANDAY-454: - Assignee: Brett Porter (was: Lars Corneliussen) maven-compile-plugin: includeSources section has issues with the path delimiter - Key: NPANDAY-454 URL: https://issues.apache.org/jira/browse/NPANDAY-454 Project: NPanday Issue Type: Bug Environment: Apache Maven 2.2.1 (r801777; 2009-08-06 21:16:01+0200) Java version: 1.6.0_24 Java home: /System/Library/Java/JavaVirtualMachines/1.6.0.jdk/Contents/Home Default locale: en_US, platform encoding: MacRoman OS name: mac os x version: 10.6.7 arch: x86_64 Family: mac Mono 2.6.7 NPanday: 1.4.0-incubating Reporter: Matthias Weßendorf Assignee: Brett Porter Fix For: 1.5.0-incubating The pom.xml (sample) files in NPanday are using the following syntax to include the AssemblyInfo.cs file: {code} includeSources includeSourceProperties\AssemblyInfo.cs/includeSource /includeSources {code} However on Mac/Mono this seems to be ignored. Changing to this {code} includeSources includeSourceProperties/AssemblyInfo.cs/includeSource /includeSources {code} make it working... -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Assigned] (NPANDAY-599) Cache resolve-result reactor-wide to avoid pdb/gac/++ beeing resolved multiple times
[ https://issues.apache.org/jira/browse/NPANDAY-599?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Corneliussen reassigned NPANDAY-599: - Assignee: Brett Porter (was: Lars Corneliussen) Cache resolve-result reactor-wide to avoid pdb/gac/++ beeing resolved multiple times Key: NPANDAY-599 URL: https://issues.apache.org/jira/browse/NPANDAY-599 Project: NPanday Issue Type: Sub-task Components: Maven Plugins Reporter: Lars Corneliussen Assignee: Brett Porter Fix For: 1.5.0-incubating Currently it seems like our custom contributors/resolvers execute for every possible path (artifact.getDependencyTrail() changes...). We can cache what we resolved reactor-wide in a separate component. Cachekey would be artifact.getId(), right? Should we let each contributor implement the cache (ArtifactResolvingContributor) - or should it be generically handled in the artifact resolver? -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-499) Make configuration for compiler-plugins and executable-plugins more flexible
[ https://issues.apache.org/jira/browse/NPANDAY-499?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14056374#comment-14056374 ] Lars Corneliussen commented on NPANDAY-499: --- [~brettporter] It's a long time ago :-( - I guess we had plans to do more for better clarity, but it is optional work. Make configuration for compiler-plugins and executable-plugins more flexible Key: NPANDAY-499 URL: https://issues.apache.org/jira/browse/NPANDAY-499 Project: NPanday Issue Type: Improvement Components: Maven Plugins Reporter: Lars Corneliussen Assignee: Lars Corneliussen Fix For: Backlog Currently all configuration for compiler-plugins and executable-plugins are located centrally in {{dotnet-core/main/resources/META-INF/npanday/*.xml}}. h3. Design * (/) Plugins should be able to contribute their own configurations (for example: the test-plugin should specify where to find nunit; currently it is specified in dotnet-core) * Mojos should accept additional configurations as parameters (if probingPaths do not match, or a new version has not yet made it into NPanday) * It should be configurable per 'plugin' and overrideable through a Mojo param / project-level-property, whether or not NPanday should try to run the command from the %PATH% h3. Location of the right executable * Use 'identifier' to instead of 'profile' for locating the correct executable-plugin; still allowing profile for customizing. * Enable version ranges for matching of versions in executables and compilers for vendorVersion, frameworkVersion and executableVersion(to come) {{[1.1,)}] or {{[2.0,3.0)}} * (/) It should be possible to configure executableVersion additionally to vendorVersion and frameworkVersion (for example in order to locate path's for Azure Tools 1.6 specifically, but only when on MICROSOFT 4.0+ and Windows. h3. Path detection * (/) Enable filtering in configuration files, supporting Windows Registry, Environment Variables and (!) project properties h3. Naming and overall structure * (/) (A) Mojo comes with *Requirement and *Config, (B) Repository delivers *Capability which is based on *Plugins; Then we try to match (A) with (B). * Rebrand 'plugin' to 'capability-configuration', hence CompilerExecutableCapabilityConfiguration (?) * Rename 'vendor' and 'vendorVersion' to 'frameworkVendor' and 'frameworkVendorVersion' for better clarity (?) * Join executable-plugins and compiler-plugins into one model (MDO + base classes for Capability * distinguish three types of executables: ** *VendorExecutable* (is provided by the framework vendor ** *CompilerExecutable* (is also provided by the framwork vendor, but has some extra requirements/capabilities ** *ThirdpartyExecutable* (is provided by a third party, hence neither NPanday nor the vendor; adds 'probingPaths' and 'executableVersion') h3. Execution and Mojo Interaction * Refactor ExecutableRequirement.ctor to accept VendorRequirment + executableIdentifier, executableVersion and executableProfile. * Remove get/setCommands() from ExecutableConfig, and require them to be passed in to NetExecutable#execute() as parameter - NetExecutable#execute(ListString commands) / filtering will happen inside * Provide utility methods, that handle exception conversions for Mojos (wrap PlatformUnsupportedException and ExecutionException) in MojoExecutionExceptions -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Assigned] (NPANDAY-499) Make configuration for compiler-plugins and executable-plugins more flexible
[ https://issues.apache.org/jira/browse/NPANDAY-499?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Corneliussen reassigned NPANDAY-499: - Assignee: Brett Porter (was: Lars Corneliussen) Make configuration for compiler-plugins and executable-plugins more flexible Key: NPANDAY-499 URL: https://issues.apache.org/jira/browse/NPANDAY-499 Project: NPanday Issue Type: Improvement Components: Maven Plugins Reporter: Lars Corneliussen Assignee: Brett Porter Fix For: Backlog Currently all configuration for compiler-plugins and executable-plugins are located centrally in {{dotnet-core/main/resources/META-INF/npanday/*.xml}}. h3. Design * (/) Plugins should be able to contribute their own configurations (for example: the test-plugin should specify where to find nunit; currently it is specified in dotnet-core) * Mojos should accept additional configurations as parameters (if probingPaths do not match, or a new version has not yet made it into NPanday) * It should be configurable per 'plugin' and overrideable through a Mojo param / project-level-property, whether or not NPanday should try to run the command from the %PATH% h3. Location of the right executable * Use 'identifier' to instead of 'profile' for locating the correct executable-plugin; still allowing profile for customizing. * Enable version ranges for matching of versions in executables and compilers for vendorVersion, frameworkVersion and executableVersion(to come) {{[1.1,)}] or {{[2.0,3.0)}} * (/) It should be possible to configure executableVersion additionally to vendorVersion and frameworkVersion (for example in order to locate path's for Azure Tools 1.6 specifically, but only when on MICROSOFT 4.0+ and Windows. h3. Path detection * (/) Enable filtering in configuration files, supporting Windows Registry, Environment Variables and (!) project properties h3. Naming and overall structure * (/) (A) Mojo comes with *Requirement and *Config, (B) Repository delivers *Capability which is based on *Plugins; Then we try to match (A) with (B). * Rebrand 'plugin' to 'capability-configuration', hence CompilerExecutableCapabilityConfiguration (?) * Rename 'vendor' and 'vendorVersion' to 'frameworkVendor' and 'frameworkVendorVersion' for better clarity (?) * Join executable-plugins and compiler-plugins into one model (MDO + base classes for Capability * distinguish three types of executables: ** *VendorExecutable* (is provided by the framework vendor ** *CompilerExecutable* (is also provided by the framwork vendor, but has some extra requirements/capabilities ** *ThirdpartyExecutable* (is provided by a third party, hence neither NPanday nor the vendor; adds 'probingPaths' and 'executableVersion') h3. Execution and Mojo Interaction * Refactor ExecutableRequirement.ctor to accept VendorRequirment + executableIdentifier, executableVersion and executableProfile. * Remove get/setCommands() from ExecutableConfig, and require them to be passed in to NetExecutable#execute() as parameter - NetExecutable#execute(ListString commands) / filtering will happen inside * Provide utility methods, that handle exception conversions for Mojos (wrap PlatformUnsupportedException and ExecutionException) in MojoExecutionExceptions -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-402) Add support to automatically attach and resolve PDB-symbols and Code-Documentation (comment.xml)
[ https://issues.apache.org/jira/browse/NPANDAY-402?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14056398#comment-14056398 ] Lars Corneliussen commented on NPANDAY-402: --- [~brettporter] PDB-symbols get attached. comment.xml is not yet getting attatched. Also the subtasks are still real issues - but we could say they are known bugs :-) Add support to automatically attach and resolve PDB-symbols and Code-Documentation (comment.xml) - Key: NPANDAY-402 URL: https://issues.apache.org/jira/browse/NPANDAY-402 Project: NPanday Issue Type: Improvement Components: Maven Plugins Affects Versions: 1.4-incubating, 1.5.0-incubating Environment: $ mvn -v Apache Maven 2.2.1 (rdebian-4) Java version: 1.6.0_24 Java home: /usr/lib/jvm/java-6-sun-1.6.0.24/jre Default locale: en_US, platform encoding: UTF-8 OS name: linux version: 2.6.35-28-generic arch: amd64 Family: unix Reporter: John R. Fallows Assignee: Lars Corneliussen Priority: Critical Fix For: 1.5.0-incubating h2. ISSUES/TODO * Integration tests missing * created subtasks (/) h2. Original Description The comments.xml file generated during CSharp compilation needs to be shipped to developers for use in Visual Studio, similar to the javadoc artifact produced by Maven for Java projects. This is not currently attached as an artifact during the NPanday build for dotnet-library. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Resolved] (NPANDAY-597) Disable transitivity for pdb-artifacts
[ https://issues.apache.org/jira/browse/NPANDAY-597?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Corneliussen resolved NPANDAY-597. --- Resolution: Fixed Disable transitivity for pdb-artifacts -- Key: NPANDAY-597 URL: https://issues.apache.org/jira/browse/NPANDAY-597 Project: NPanday Issue Type: Sub-task Components: Maven Plugins Reporter: Lars Corneliussen Assignee: Lars Corneliussen Fix For: 1.5.0-incubating -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Closed] (NPANDAY-595) Option to disable PDB-resolving (disabled by default?)
[ https://issues.apache.org/jira/browse/NPANDAY-595?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Corneliussen closed NPANDAY-595. - Resolution: Fixed Option to disable PDB-resolving (disabled by default?) -- Key: NPANDAY-595 URL: https://issues.apache.org/jira/browse/NPANDAY-595 Project: NPanday Issue Type: Sub-task Components: Maven Plugins Reporter: Lars Corneliussen Fix For: 1.5.0-incubating I think it is not necessary to resolve pdbs by default. It should at least be enough to resolve them for testing (although, if it slows down passing tests, it might be better not to have nice stacktraces - but that could be optional). -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Resolved] (NPANDAY-566) Custom NPanday-aware copy-dependencies / list-dependencies plugin
[ https://issues.apache.org/jira/browse/NPANDAY-566?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Corneliussen resolved NPANDAY-566. --- Resolution: Fixed Custom NPanday-aware copy-dependencies / list-dependencies plugin - Key: NPANDAY-566 URL: https://issues.apache.org/jira/browse/NPANDAY-566 Project: NPanday Issue Type: New Feature Components: Maven Plugins Affects Versions: 1.5.0-incubating Reporter: Lars Corneliussen Assignee: Lars Corneliussen Fix For: 1.5.0-incubating When copying dependencies we have to run our own resolver, and we'd also like to copy along pdb+xml+... as soon as we have those... -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Resolved] (NPANDAY-575) Copy pdb files into test assembly directory
[ https://issues.apache.org/jira/browse/NPANDAY-575?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Corneliussen resolved NPANDAY-575. --- Resolution: Fixed Copy pdb files into test assembly directory --- Key: NPANDAY-575 URL: https://issues.apache.org/jira/browse/NPANDAY-575 Project: NPanday Issue Type: New Feature Components: Maven Plugins Affects Versions: 1.5.0-incubating Reporter: Victor Stefoglo Assignee: Lars Corneliussen Priority: Minor Fix For: 1.5.0-incubating Attachments: ExampleProject.zip, dotnet-core.patch, maven-test-plugin.patch Some coverage tools as OpenCover, PartCover require pdb files to be present. We use test assembly directory, for coverage tools, and we have a workaround to copy pdb files to test assembly directory (using copy-maven-plugin). I think this should be done by Npanday maven-test-plugin, now It copy all dependency artifacts but generated pdb files are ignored. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Updated] (NPANDAY-402) Add support to automatically attach and resolve PDB-symbols and Code-Documentation (comment.xml)
[ https://issues.apache.org/jira/browse/NPANDAY-402?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Corneliussen updated NPANDAY-402: -- Description: h2. ISSUES/TODO * Integration tests missing * impl for attaching comments.xml missing * created subtasks (/) h2. Original Description The comments.xml file generated during CSharp compilation needs to be shipped to developers for use in Visual Studio, similar to the javadoc artifact produced by Maven for Java projects. This is not currently attached as an artifact during the NPanday build for dotnet-library. was: h2. ISSUES/TODO * Integration tests missing * created subtasks (/) h2. Original Description The comments.xml file generated during CSharp compilation needs to be shipped to developers for use in Visual Studio, similar to the javadoc artifact produced by Maven for Java projects. This is not currently attached as an artifact during the NPanday build for dotnet-library. Add support to automatically attach and resolve PDB-symbols and Code-Documentation (comment.xml) - Key: NPANDAY-402 URL: https://issues.apache.org/jira/browse/NPANDAY-402 Project: NPanday Issue Type: Improvement Components: Maven Plugins Affects Versions: 1.4-incubating, 1.5.0-incubating Environment: $ mvn -v Apache Maven 2.2.1 (rdebian-4) Java version: 1.6.0_24 Java home: /usr/lib/jvm/java-6-sun-1.6.0.24/jre Default locale: en_US, platform encoding: UTF-8 OS name: linux version: 2.6.35-28-generic arch: amd64 Family: unix Reporter: John R. Fallows Assignee: Lars Corneliussen Priority: Critical Fix For: 1.5.0-incubating h2. ISSUES/TODO * Integration tests missing * impl for attaching comments.xml missing * created subtasks (/) h2. Original Description The comments.xml file generated during CSharp compilation needs to be shipped to developers for use in Visual Studio, similar to the javadoc artifact produced by Maven for Java projects. This is not currently attached as an artifact during the NPanday build for dotnet-library. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-563) Generic MSDeploy synchronization mojo
[ https://issues.apache.org/jira/browse/NPANDAY-563?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14046983#comment-14046983 ] Lars Corneliussen commented on NPANDAY-563: --- agreed. Generic MSDeploy synchronization mojo - Key: NPANDAY-563 URL: https://issues.apache.org/jira/browse/NPANDAY-563 Project: NPanday Issue Type: New Feature Components: Maven Plugins Reporter: Lars Corneliussen Assignee: Lars Corneliussen Fix For: 1.5.0-incubating -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-556) Importer for Nuget Packages
[ https://issues.apache.org/jira/browse/NPANDAY-556?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14046985#comment-14046985 ] Lars Corneliussen commented on NPANDAY-556: --- So; close and mark as resolve? Importer for Nuget Packages --- Key: NPANDAY-556 URL: https://issues.apache.org/jira/browse/NPANDAY-556 Project: NPanday Issue Type: New Feature Components: Development Setup, Maven Plugins Affects Versions: 1.5.0-incubating Reporter: Lars Corneliussen Assignee: Lars Corneliussen Labels: nuget Fix For: 1.5.0-incubating Nuget Packages are much broader than Maven Artifacts. Our first thought was to implement a repository layout that would enable for resolving nuget packages directly (see NPANDAY-372). But those packages come in various shapes and are not as easily converted to {{dotnet-library}} artifacts as first thought. h3. Library Importer Instead we will create a plugin, that helps importing nuget packages (and soon also plain DLL-libs) into maven artifacts. As a general rule, each DLL contained in a single {{nupkg}} will result in a separate maven artifact. Some of the challenges we see: h4. Versions Nuget packages (and dlls) have four version numbers while maven only has three. In most cases those are not used in nuget, but if, there has to be some manual mapping. h4. Multi-targeted packages Nuget packages may contain libraries for different target frameworks (net20, 35, wp, sl, ...); something that we'd like to support in NPanday too (NPANDAY-405) As a first approach, you'll have to specify which of the versions to import. We could also support a mapping from framework versions to maven version qualifiers, as for example {{1.2-net20}} h4. Dependencies Since there is no easy mapping from package to artifact, it is not that easy to convert the dependencies either. For each import it will be necessary to specify how the reps are mapped. h4. Prerelease versions Nuget = 1.6 does also support prerelease versions. These should be mapped to SNAPSHOT, if possible. h4. What more? Please comment or edit, if you see more challenges that I haven't seen yet. h2. Technical approach A plugin with it's own lifecycle will 'compile' import declarations and then perform the import on 'compile'; then install and deploy the imported packages on the corresponding maven life cycles 'install' and 'deploy'. {code:title=example import declaration} libs xmlns=http://npanday.apache.org/library-import/1.0.0; nuget packageNUnit/package version source=2.6.0.12054 mapTo=2.6.0/ version source=2.5.10.11092 mapTo=2.5.10/ libDirs defaultnet40/default /libDirs /nuget /libs {code} -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-556) Importer for Nuget Packages
[ https://issues.apache.org/jira/browse/NPANDAY-556?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14046988#comment-14046988 ] Lars Corneliussen commented on NPANDAY-556: --- ups. didn't see the FIXED :-) thanks. Importer for Nuget Packages --- Key: NPANDAY-556 URL: https://issues.apache.org/jira/browse/NPANDAY-556 Project: NPanday Issue Type: New Feature Components: Development Setup, Maven Plugins Affects Versions: 1.5.0-incubating Reporter: Lars Corneliussen Assignee: Lars Corneliussen Labels: nuget Fix For: 1.5.0-incubating Nuget Packages are much broader than Maven Artifacts. Our first thought was to implement a repository layout that would enable for resolving nuget packages directly (see NPANDAY-372). But those packages come in various shapes and are not as easily converted to {{dotnet-library}} artifacts as first thought. h3. Library Importer Instead we will create a plugin, that helps importing nuget packages (and soon also plain DLL-libs) into maven artifacts. As a general rule, each DLL contained in a single {{nupkg}} will result in a separate maven artifact. Some of the challenges we see: h4. Versions Nuget packages (and dlls) have four version numbers while maven only has three. In most cases those are not used in nuget, but if, there has to be some manual mapping. h4. Multi-targeted packages Nuget packages may contain libraries for different target frameworks (net20, 35, wp, sl, ...); something that we'd like to support in NPanday too (NPANDAY-405) As a first approach, you'll have to specify which of the versions to import. We could also support a mapping from framework versions to maven version qualifiers, as for example {{1.2-net20}} h4. Dependencies Since there is no easy mapping from package to artifact, it is not that easy to convert the dependencies either. For each import it will be necessary to specify how the reps are mapped. h4. Prerelease versions Nuget = 1.6 does also support prerelease versions. These should be mapped to SNAPSHOT, if possible. h4. What more? Please comment or edit, if you see more challenges that I haven't seen yet. h2. Technical approach A plugin with it's own lifecycle will 'compile' import declarations and then perform the import on 'compile'; then install and deploy the imported packages on the corresponding maven life cycles 'install' and 'deploy'. {code:title=example import declaration} libs xmlns=http://npanday.apache.org/library-import/1.0.0; nuget packageNUnit/package version source=2.6.0.12054 mapTo=2.6.0/ version source=2.5.10.11092 mapTo=2.5.10/ libDirs defaultnet40/default /libDirs /nuget /libs {code} -- This message was sent by Atlassian JIRA (v6.2#6252)
AW: NPanday incubator report
Hi Roman, Raphael Bircher signed up as a mentor in late 2013. Currently we have noone investing in the project. Brett and I do from time to time implement new features we need, do fixes whenever we (or our customers) get stuck. We also have some people providing patches here and then - but it's the same for them - they do it, when they get stuck. I think nearby everyone is using 1.5.0-SNAPSHOT at the moment anyway. It could just be released - even with bugs and regression issues. I don't really know how to do the release, and Brett doesn't either find the time, I guess :-) Still I see a good future for NPanday. The official Microsoft alternatives are just too bad. But still it's hard to convince Microsoft-people to have a look at NPanday - they get scared when the hear the word java. :-) _ Lars -Ursprüngliche Nachricht- Von: odeach...@gmail.com [mailto:odeach...@gmail.com] Im Auftrag von Deng Ching-Mallete Gesendet: Freitag, 13. Juni 2014 05:46 An: npanday-dev Cc: mat...@apache.org; r...@apache.org Betreff: Re: NPanday incubator report Hi Roman, My apologies, I didn't see your email immediately. Unfortunately, I am no longer an active mentor for NPanday. I've already stepped down as mentor for the project over a year ago so I am not quite up to date with what's happening on the project. I took a quick look now at the Jira and commit notifications, and the last commit was on April 4. There were also a number of tickets reported in the recent weeks, but there doesn't seem to be anyone actively working on them. Hope this somewhat helps! Thanks, Deng On Fri, Jun 13, 2014 at 11:22 AM, Roman Shaposhnik r...@apache.org wrote: On Tue, Jun 10, 2014 at 1:18 AM, David Akehurst d...@akehurst.net wrote: I hope the project has not gone dormant. I still actively use it, and hoping for its continued development and eventual (soon) promotion from incubating. At the very minimum the project would require some active mentors. Hang in there -- we'll try to resolve this at IPMC level. Thanks, Roman.
[jira] [Commented] (NPANDAY-623) Support reactor reference to other modules at compile phase
[ https://issues.apache.org/jira/browse/NPANDAY-623?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14028953#comment-14028953 ] Lars Corneliussen commented on NPANDAY-623: --- Hi, you are right! Doesn*t like, that compile doesn't use final name - i think it should. But in the .NET-area there are some expectations to dll-names when automatically resolving assemblies - so no easy choice. Just changing final-name in the pom-model if it isn't yet custom-changed could be another option. Support reactor reference to other modules at compile phase Key: NPANDAY-623 URL: https://issues.apache.org/jira/browse/NPANDAY-623 Project: NPanday Issue Type: Bug Components: Maven Plugins Affects Versions: 1.4-incubating, 1.5.0-incubating Environment: Maven 2.2.1 Reporter: Greg Domjan Attachments: example.zip When building multiple artifacts with dependencies in the reactor it appears to be currently required that you build to the install phase. This can actually be a bug - see http://developer-blog.cloudbees.com/2012/12/maven-and-hack.html Even if not a bug, this goes against some common practices such as used in the release-plugin which attempts to test the build by running all projects to compile stage only. In comparison the maven reactor allows for dependency resolution in Java JAR dependencies as soon as the compile phase is run. It does this by pointing the artifact file to the local class folder. NPanday could take similar action and point the file to the local target file as soon as compile. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-624) NPE in ilMerge
[ https://issues.apache.org/jira/browse/NPANDAY-624?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14028964#comment-14028964 ] Lars Corneliussen commented on NPANDAY-624: --- Have you yet looked at the code? It's no magic :-) Also, could it be the same problem as the other issue - that the artifact isn't resolved...? NPE in ilMerge -- Key: NPANDAY-624 URL: https://issues.apache.org/jira/browse/NPANDAY-624 Project: NPanday Issue Type: Bug Components: Maven Plugins Affects Versions: 1.5.0-incubating Environment: Windows, VS2005, VS2012, Platform SDK 6,7,7.1,8.0,8.1 Using npanday.settings to select mininmal framework version 3.0 Maven 2.2.1 Reporter: Greg Domjan ilMerge appears to be looking for the compiler details to select the appropriate ilMerge app. The compiler list is returning multiple options, but then ilMerge gets NPE on following call to ```File assemblyPath = compilerExecutable.getAssemblyPath();``` [DEBUG] NPANDAY-102-003: Apply rule:npanday.vendor.impl.VendorInfoTransitionRuleFactory$8@38c9aa93 [DEBUG] NPANDAY-103-017: Entering State = FFF [DEBUG] NPANDAY-103-052: Set defaults: 3.0 [DEBUG] NPANDAY-102-004: Vendor info requirement after rule:[VendorRequirement for vendor MICROSOFT version 3.0, Framework Version = 3.0] [DEBUG] NPANDAY-065-008: Found vendor [Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0] for requirement [VendorRequirement for vendor MICROSOFT version 3.0, Framework Version = 3.0] [DEBUG] NPANDAY-065-001: Found matching capability: CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='C_SHARP'] [DEBUG] NPANDAY-065-009: Failed to match policy: ExecutableMatchPolicy[profile: 'FULL'] [DEBUG] NPANDAY-065-005: Capability doesn't match: CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='C_SHARP'] [DEBUG] NPANDAY-065-001: Found matching capability: CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='C_SHARP'] [DEBUG] NPANDAY-065-001: Found matching capability: CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='C_SHARP'] [DEBUG] NPANDAY-065-001: Found matching capability: CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='C_SHARP'] [DEBUG] NPANDAY-065-009: Failed to match policy: ExecutableMatchPolicy[language: 'C_SHARP'] [DEBUG] NPANDAY-065-005: Capability doesn't match: CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='VB'] [DEBUG] NPANDAY-065-009: Failed to match policy: ExecutableMatchPolicy[language: 'C_SHARP'] [DEBUG] NPANDAY-065-005: Capability doesn't match: CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='ASP'] [WARNING] NPANDAY-065-010: Found multiple matching capabilities; will choose the first one: [CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='C_SHARP'], CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='C_SHARP'], CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='C_SHARP'], CompilerCapability [vendorInfo=[Configured Vendor Info for MICROSOFT 3.0, Framework Version = 3.0], operatingSystem='Windows', language='C_SHARP']] [DEBUG] NPANDAY-061-006: Artifact Type:gac_msil [DEBUG] NPANDAY-061-007: Artifact Type:false [DEBUG] NPANDAY-061-006: Artifact Type:dotnet-library [DEBUG] NPANDAY-061-007: Artifact Type:false [DEBUG] NPANDAY-061-006: Artifact Type:dotnet-library [DEBUG] NPANDAY-061-007: Artifact Type:false [DEBUG] NPANDAY-061-006: Artifact Type:gac_msil [DEBUG] NPANDAY-061-007: Artifact Type:false [DEBUG] NPANDAY-061-006: Artifact Type:gac_msil [DEBUG] NPANDAY-061-007: Artifact Type:false [WARNING] NPANDAY-231: previously netDependencyId was used to resolve some private bin path... [INFO] [ERROR] FATAL ERROR [INFO] [INFO] null [INFO] [DEBUG] Trace java.lang.NullPointerException
[jira] [Commented] (NPANDAY-623) Support reactor reference to other modules at compile phase
[ https://issues.apache.org/jira/browse/NPANDAY-623?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14015512#comment-14015512 ] Lars Corneliussen commented on NPANDAY-623: --- hmpf. could you provide a zip-file? don*t have too much time to look at it this week, though Support reactor reference to other modules at compile phase Key: NPANDAY-623 URL: https://issues.apache.org/jira/browse/NPANDAY-623 Project: NPanday Issue Type: Bug Components: Maven Plugins Affects Versions: 1.4-incubating, 1.5.0-incubating Environment: Maven 2.2.1 Reporter: Greg Domjan When building multiple artifacts with dependencies in the reactor it appears to be currently required that you build to the install phase. This can actually be a bug - see http://developer-blog.cloudbees.com/2012/12/maven-and-hack.html Even if not a bug, this goes against some common practices such as used in the release-plugin which attempts to test the build by running all projects to compile stage only. In comparison the maven reactor allows for dependency resolution in Java JAR dependencies as soon as the compile phase is run. It does this by pointing the artifact file to the local class folder. NPanday could take similar action and point the file to the local target file as soon as compile. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-542) Support for multiple source dirs
[ https://issues.apache.org/jira/browse/NPANDAY-542?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13993578#comment-13993578 ] Lars Corneliussen commented on NPANDAY-542: --- Hi, actually this should be available in 1.5.0-SNAPSHOT, yes :-) _ Lars Support for multiple source dirs Key: NPANDAY-542 URL: https://issues.apache.org/jira/browse/NPANDAY-542 Project: NPanday Issue Type: Improvement Components: Maven Plugins Affects Versions: 1.4-incubating Reporter: Urs Keller Assignee: Lars Corneliussen Fix For: 1.5.0-incubating If another maven plugin generates sources for other plugins to process this is not picked up by NPanday. Typically we have a plugin adding to the sources with the following code: /** * The source directories containing the sources to be compiled. * * @parameter expression=${project.compileSourceRoots} * @required * @readonly */ private ListString compileSourceRoots; ... compileSourceRoots.add(getTargetDir().getAbsolutePath()); -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (NPANDAY-617) compile:generate-assembly-info Allow for filtering/customization of properties
[ https://issues.apache.org/jira/browse/NPANDAY-617?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13994942#comment-13994942 ] Lars Corneliussen commented on NPANDAY-617: --- Agreed. .NETs Nuget also starts aligning to semver (Semantic Versioning) with only 3 numbers. It is still possible to append a fourth with a dash, and if it needs to be a number, prepend it such that it is sortable as string. Since I use Maven to resolve dependencies and package what I deploy, I don't care about .NETs version resolving. Also though about generating assembly binding xml for cases where the version rules not match. Then using the first 3 numbers from the maven version + .0 (for AssemblyVersion) is a good fit. AssemblyInformationalVersion can be an arbitrary string - i just ignore the warning. (could put CS1607) in the default ignores, if so. compile:generate-assembly-info Allow for filtering/customization of properties -- Key: NPANDAY-617 URL: https://issues.apache.org/jira/browse/NPANDAY-617 Project: NPanday Issue Type: Improvement Components: Maven Plugins Affects Versions: 1.4-incubating Reporter: Greg Domjan Labels: configuration, documentation, filters, versioning The generation of assembly info from the maven pom is desirable, especially in reguard to filling in version and copyright year, so having a static AssemblyInfo.cs to customize the value doesn't seem to be a good work around (unless I've missed an alternate way to apply filtering?) It is unclear where these might be sourced from [assembly: AssemblyCopyright()] [assembly: AssemblyTrademark()] [assembly: AssemblyCulture()] The [assembly: AssemblyProduct()] appears to be a combination of Company-Title and it would be good to be able to customize it. In regard to version, it appears maven2 version format isn't supported? If an alternate source for version could be provided so that the version formatted to .net/npanday standard could be used it would be helpful. For example Using the maven 2 standard format version of major.minor.revision-build reports issues due to generation requiring the 'string' version of major.minor.revision.build which can cause issues with version comparison. ie. version8.1.0-0-SNAPSHOT/version generates [assembly: AssemblyVersion(8.1.0)] [assembly: AssemblyInformationalVersion(8.1.0-0-SNAPSHOT)] warning CS1607: Assembly generation -- The version '8.1.0-0-SNAPSHOT' specified for the 'product version' is not in the normal 'major.minor. build.revision' format Best ref I found is http://mojo.codehaus.org/versions-maven-plugin/version-rules.html For example, Maven arranges the version list in the following manner: 1.0.1.0 1.0.10.1 1.0.10.2 1.0.9.3 Version 1.0.9.3 should come before 1.0.10.1 and 1.0.10.2, but the unexpected fourth field (.3) forced Maven to evaluate the version as a string. -- This message was sent by Atlassian JIRA (v6.2#6252)
AW: NPanday report due Apr 2
+1 -Ursprüngliche Nachricht- Von: Brett Porter [mailto:br...@porterclan.net] Im Auftrag von Brett Porter Gesendet: Dienstag, 1. April 2014 07:30 An: npanday-dev@incubator.apache.org Betreff: NPanday report due Apr 2 Hi, The report has been missed for a couple of months, and is due on April 2. Does anyone have any contributions they'd like to make to it? Here is a draft - I'd appreciate a +1 or comments from those following along. --- NPanday allows projects using the .NET framework to be built with Apache Maven. NPanday has been incubating since 2010-08-13. In previous reports, we indicated that there was a regression that blocked releasing the next version. This has now been resolved, and the only thing that stands in the way is sufficient cycles to push it out. There have been a small number of users (not yet committers) who've stepped forward to help test when that happens, which might help build some much needed momentum. However, as before, it's still a concern whether sufficient release votes could be found as many of the existing PPMC are inactive. The last release was made on 16 May 2011. The steps that have been identified are: - apply patches that have been contributed recently - convert the website to svnpubsub so updated documentation can be published - perform the release steps The NPanday automated builds have been impacted by changes to the Windows build infrastructure again, but this is probably something that needs to be worked around on our end. --- Regards, Brett -- Brett Porter @brettporter http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
AW: svn commit: r1573504 - in /incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src: main/csharp/Converter/Algorithms/ test/resource/MvcApplication1/MvcApplication1/ test/resou
Hi Brett, sorry for still not actively beeing watching the list :-( Still an issue? Would need to dive into it to give you feedback. If you really want an answer ping me on skype :-) Or send a copy to my direct email address - - i'll then answer to the dev group :-) Lars -Ursprüngliche Nachricht- Von: Brett Porter [mailto:br...@porterclan.net] Im Auftrag von Brett Porter Gesendet: Montag, 3. März 2014 11:03 An: npanday-dev@incubator.apache.org Cc: Lars Corneliussen Betreff: Re: svn commit: r1573504 - in /incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/sr c: main/csharp/Converter/Algorithms/ test/resource/MvcApplication1/MvcApplication1/ test/resource/NPANDAY_480_AzureSupportOneWebRole/HelloWorld_WebRo Hi Lars, I'm interested in your feedback here. I've made this change to be more inline with the previous behaviour which used contentPath. I picked up on this because it turns out the Resolve Web Roles mojo in the Azure plugin assumes a contentPath destination, and fails if you use iisApp. Is this change correct, or should it in fact be iisApp now, and the resolve mojo adjusted? (You can verify the behaviour through the Azure integration tests). Cheers, Brett On 3 Mar 2014, at 8:55 pm, br...@apache.org wrote: Author: brett Date: Mon Mar 3 09:55:30 2014 New Revision: 1573504 URL: http://svn.apache.org/r1573504 Log: [NPANDAY-488] Use contentPath by default This was the previous behaviour: http://svn.apache.org/viewvc/incubator/npanday/trunk/plugins/msdeploy- maven-plugin/src/main/java/npanday/plugin/msdeploy/create/CreateConten tPackageMojo.java?r1=1333788r2=1332286pathrev=1333788 Also, there is a problem using iisApp with resolve web roles, since it expects the contentPath destination: http://svn.apache.org/viewvc/incubator/npanday/trunk/plugins/msdeploy- maven-plugin/src/main/java/npanday/plugin/msdeploy/create/CreateConten tPackageMojo.java?r1=1333788r2=1332286pathrev=1333788 Modified: incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src /main/csharp/Converter/Algorithms/ASPNetPomConverter.cs incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src /test/resource/MvcApplication1/MvcApplication1/pom.test incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src /test/resource/NPANDAY_480_AzureSupportOneWebRole/HelloWorld_WebRole/pom.tes t incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src /test/resource/NPANDAY_480_CloudServiceWithMultipleRoles/HelloWorld_WebRole/ pom.test incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src /test/resource/NPANDAY_488/NPANDAY_488_MSDeployPackageSimpleWebApp/pom.test incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src /test/resource/SilverlightApplication1/SilverlightApplication1.Web/pom.test incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src /test/resource/SilverlightApplication5/SilverlightApplication5.Web/pom.test incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engi ne/src/test/resource/WebAppWithCultureRes/WebAppWithCultureRes/pom.tes t Modified: incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engi ne/src/main/csharp/Converter/Algorithms/ASPNetPomConverter.cs URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/dotnet/assemblies /NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/A SPNetPomConverter.cs?rev=1573504r1=1573503r2=1573504view=diff == --- incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engi ne/src/main/csharp/Converter/Algorithms/ASPNetPomConverter.cs (original) +++ incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/ +++ Engine/src/main/csharp/Converter/Algorithms/ASPNetPomConverter.cs +++ Mon Mar 3 09:55:30 2014 @@ -54,7 +54,7 @@ namespace NPanday.ProjectImporter.Conver } Plugin msdeployPlugin = AddPlugin(org.apache.npanday.plugins, msdeploy-maven-plugin, null, false); -AddPluginExecution(msdeployPlugin, create-msdeploy-package, new string[] { create-iisApp-package }, null); +AddPluginExecution(msdeployPlugin, + create-msdeploy-package, new string[] { create-content-package + }, null); if (projectDigest.SilverlightApplicationList != null) { Modified: incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engi ne/src/test/resource/MvcApplication1/MvcApplication1/pom.test URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/dotnet/assemblies /NPanday.ProjectImporter/Engine/src/test/resource/MvcApplication1/MvcA pplication1/pom.test?rev=1573504r1=1573503r2=1573504view=diff == --- incubator/npanday/trunk/dotnet/assemblies
Re: Building NPanday
In integration tests we do that already. I think we have a couple of integration tests that test other local repos… Maybe it is easy to just set the local-repository for all of them. Brett should have a better overview… I'm coming from the other side - learned Java just a couple of years ago. Am 04.12.2013 um 09:39 schrieb christofer.d...@c-ware.de christofer.d...@c-ware.de: Hi Lars, I can immagine that not all design decisions we did in Flexmojos are applicable and valid on other platforms. It was just that we noticed a lot less fuss with this Approach. I guess you are right. Windows stuff was allways a lot more complex than Java stuff ;-) But still we should think about how it would be possible to make it easier to start working with NPanday. Even I as a Maven and IT professional had to do quite some Research getting things up and running. From my experiance with Flexmojos the Level of willingness to do so is relatively small with the majority of potential users. I could help modify the testsuite to run with the maven-invoker. For this I could simply duplicate the Setup used in Flexmojos. This shouldn't be that hard. After all every test is simply executing a maven build and checking the Output. So things should be quite similar. Chris Von: Lars Corneliussen [m...@lcorneliussen.de] Gesendet: Dienstag, 3. Dezember 2013 20:07 An: npanday-dev@incubator.apache.org Betreff: Re: Building NPanday inline… Am 03.12.2013 um 12:22 schrieb christofer.d...@c-ware.de christofer.d...@c-ware.de: Well in Flex the SDK also Comes with quite a lot of binaries. Even if most of them are Jar files, there still are quite a lot of executeable binaries (Flashplayer, Air-Runtime, Air Debugger) which are exe-files. I guess however that DLLs could be a Problem, if the names Change in a mavenized form. So if MyCool.dll would become mycool-1.3.4.dll I guess this would mess up things, but I have to admit that I'm still pretty new to the .Net coding. I've got tons of Java experiance and am currently confronted of having to create .Net Clients for my Java Server applications ... and therefore wanting to do this in one maven build (Just to give you a bit more on my Background) Ok. In earlier versions of NPanday we had a separate repository that kept the files with correct names in version folders. But that introduced a lot of other problems. But I'd be fine with creating temporary execution-directories in the reactors parent target-folder. Still I'm not sure, if we should mavenize all of .NETs SDKs. I'm pretty sure they require changes to the registry ++ Also the current vendor detection is more sophisticated than just resolving a maven artifact (version, vendor, frameworkVersion, ++). I think the npanday-settings.xml-approach is ok, but in order to avoid the bootstrapping problem, the vendor detection could run live in JAVA code. Still I see the need to resolve executables from maven (which again could be imported from nuget). That is basically what the .NET plugins do. Just that they aren't normal executables, but they get passed the mojo configuration... But what do you think about my Suggestion to have the Testsuite run in a separate local test maven repo? I'm still cleaning up my main local maven repo after my last execution of NPandays Testsuite. Totally agree. Not sure, what needs to be done to have that running, though... Chris Von: Lars Corneliussen [m...@lcorneliussen.de] Gesendet: Montag, 2. Dezember 2013 22:29 An: npanday-dev@incubator.apache.org Betreff: Re: Building NPanday got you! but the npanday-settings.xml is more about the SDK binaries than about maven artifacts. i don't think it is easy to copy them all around - but who knows.… For nunit, ++ it shouldn't be a problem at all! but I agree, that we should be able to have maven-artifacts (zip files, e.g.) as prereq for plugins. We could for example use the application-packaging (uses maven archiver). Am 26.11.2013 um 08:48 schrieb christofer.d...@c-ware.de: Well currently there are different Library packages in different versions (.Net 2.0, .Net 3.5, .Net X.X. NUnit, Lots more ...). All of These have their own internal Directory structure and contain sets of libraries (DLLs). As far as I undertstood NPanday it depends on These Libs being installed on the host running a NPanday build and NPanday depends on These libraries as the dlls are not copied to the maven repository, but instead the npanday-settings.xml tells it in what Directory things are located. My Suggestion was to mavenize the libraries (Making Maven artifacts out of them and copying them to the local repo instead of maintaining the npanday-settings.xml). This way you could deploy the artifacts to a Company Maven repo and another user wanting to do a build
Re: Building NPanday
do still not really understand… what libs are you talking about? nunit, ++? adana-repo? ... Am 22.11.2013 um 11:42 schrieb christofer.d...@c-ware.de: Well I guess in the case of NPanday the local maven repostitory contains the NPanday Plugin(s) in the maven local repository and a npanday-settings.xml that is located somewhere. The Libs and Binaries used for compiling are located in the place where they were installed. Assume an new colleague joins a Team already working on a Project. He has to install all sorts of libs first and hopefully get them in the right Version. In all other cases the build will Fail. I was suggesting not to use a npanday-settings.xml at all, but to create a mavenizer that copies all the libs and binaries needed into a Maven form located in the maven local repository. This way These files can also be shared using a companies Maven repository and used in a Project by adding a simple Maven dependency. Assumings this Approach, as soon as the new colleague joins the Team, the first build, would download all needed libs and binaries from the companies Maven repository and he could start working immediately. When I first started to get the NPanday Unit Test Suite to execute, I needed quite some time to find out which Libraries were missing, After installing some of them I noticed that the new Versions were incompatable with the ones the Unit Test required and it was very hard to get Access to the old Versions. This was all making it harder for me to get started and I think it will also prevent others from doing the same. Chris Von: Lars Corneliussen [m...@lcorneliussen.de] Gesendet: Freitag, 22. November 2013 10:24 An: npanday-dev@incubator.apache.org Betreff: Re: Building NPanday so, instead of running the mavenizer (dotnet-plugins/settings-generator) through maven it would just run as a standalone program and update the npanday-settings.xml? Am 21.11.2013 um 13:53 schrieb christofer.d...@c-ware.de christofer.d...@c-ware.de: Hi Lars, And what about the idea of creating a .Net mavenizer that creates maven artifacts for all needed components and libraries. Then you could rely on this maven structure and new .Net Versions or lib Versions would simply be a Thing of updating the Mavenizer. This is the path we took for Flexmojos which relys on a Flex SDK, Air SDK and Flashplayer libs. Ryling on libs in a certain structure on a System was a far to fragile construct and keept on breaking things whenever a new SDK changed the structure a Little. Chris Von: Lars Corneliussen [m...@lcorneliussen.de] Gesendet: Donnerstag, 21. November 2013 11:07 An: npanday-dev@incubator.apache.org Betreff: Re: Building NPanday Hi Christopher, for Azure we read the SDK paths from the registry (configured in embedded xml-files). Some of the paths detection still runs in dotnet-plugins; I think it should be moved to Java code in order to have it accessible directly (live) from all plugins. (Now we generate a file npanday-settings.xml that contains the information about installed .NET SDKs) (https://issues.apache.org/jira/browse/NPANDAY-505) But it is not easy to get everything to run in all environments. The nuget-importer also needs to have nuget on the PATH - which I don't like at all! But there is no default place to get it from… Same with NUnit. It would be great if we could bootstrap things through nuget… Maybe embedding the nuget-commandline bottstrapper… _ Lars Am 13.11.2013 um 20:42 schrieb christofer.d...@c-ware.de christofer.d...@c-ware.de: Hi, Ok so I had another look at the bootstrap thing and it seems to work ... think I should start looking at the stuff in the root directory and not trusting the documentation on the website ... this sort of never works ;-) So I don't want to change a running system ;-) Now I'm trying to run the integration tests. Here I was having a little trouble: - I installed Azure SDK 2.2 (Sort of couldn't install 1.7) and linked that directory to C:\Programms\Windows Azure SDK\v1.6 and C:\Programms\Microsoft SDKs\Windows Azure\.NET SDK\2012-06 and it seemed to have worked a little :-) - I have Visual Studio 2013 installed and some tests are skipped because of me not having 2010 installed. - Still there are 18 Test failing (It seems some tests are currently not running ... which ones are known not to run?) One other thing I found a little annoying in the integration-tests suite, was that it pollutes my local maven repository. I am the lead developer of Flexmojos and here we use the maven-invoker-plugin to populate a test local repo located in the test-harness' target directory and invoke child maven builds, resulting in the normal local maven repo staying untouched. After a mvn clean all of this is cleared keeping
Re: Incubator
Hi guys, just my two cents on the project. We (Faktum Software GmbH) are still interested in bringing NPanday forward. We use it actively on a big project - but the project consumes all my time. So I have hardly more time to do more then just make it run. Still, we are actively making changes (PDB support, Nuget-importer, …) - but we haven't had the energy to push the release of 1.5 forward. I'm also happy to help others getting involved. Just skype me on lcorneliussen (I should maybe start to actively watch this list … :-)) Currently _ Lars Am 16.11.2013 um 20:38 schrieb Dennis Lundberg dennisl.apa...@gmail.com: Hi Raphael On Sun, Nov 10, 2013 at 9:16 PM, Raphael Bircher r.birc...@gmx.ch wrote: Hi David Am 10.11.13 20:41, schrieb David Akehurst: Hi, I know that Brett Porter used to be very active here (perhaps he is on holiday), he helped a college of mine get a local version of the plugin built, and has answered some of my questions before. Yea, I saw Brett Porter and know him from the infra list. Nice for him to have Holiday ;-) Anyway, I think, there are more committers here who are still sleeping. As you know, it needs time to wake up after a long sleep ;-) Anyway, I asked now the Incubator for new Mentors for this project. since the old Mentor Dennis Lundberg is inactive for over a year. And one Mentor is not enough anyway That's not entirely true, I signed the report back in February 2013, but yes I have been largely occupied elsewhere. In September I contacted Brett off-list to give a heads up that I planned to resign as a mentor. I'll take that discussion over at incubator when I can find the time. I too would be happy to make some contributions, but need some guidance, and can only build on linux at present. Unfortunaly I can't tell you much here. I comming from a compleet different direction (OpenOffice) and have no experiance with this kind of code. I can help with community building, and probabily a bit testing. But to be honest. I'm not planing to contribute for this project. Just help to restart. Btw. maybe a mail to the users would be good. Greetings from Switzerland Raphael -- Dennis Lundberg
Re: Building NPanday
Hi Christopher, for Azure we read the SDK paths from the registry (configured in embedded xml-files). Some of the paths detection still runs in dotnet-plugins; I think it should be moved to Java code in order to have it accessible directly (live) from all plugins. (Now we generate a file npanday-settings.xml that contains the information about installed .NET SDKs) (https://issues.apache.org/jira/browse/NPANDAY-505) But it is not easy to get everything to run in all environments. The nuget-importer also needs to have nuget on the PATH - which I don't like at all! But there is no default place to get it from… Same with NUnit. It would be great if we could bootstrap things through nuget… Maybe embedding the nuget-commandline bottstrapper… _ Lars Am 13.11.2013 um 20:42 schrieb christofer.d...@c-ware.de christofer.d...@c-ware.de: Hi, Ok so I had another look at the bootstrap thing and it seems to work ... think I should start looking at the stuff in the root directory and not trusting the documentation on the website ... this sort of never works ;-) So I don't want to change a running system ;-) Now I'm trying to run the integration tests. Here I was having a little trouble: - I installed Azure SDK 2.2 (Sort of couldn't install 1.7) and linked that directory to C:\Programms\Windows Azure SDK\v1.6 and C:\Programms\Microsoft SDKs\Windows Azure\.NET SDK\2012-06 and it seemed to have worked a little :-) - I have Visual Studio 2013 installed and some tests are skipped because of me not having 2010 installed. - Still there are 18 Test failing (It seems some tests are currently not running ... which ones are known not to run?) One other thing I found a little annoying in the integration-tests suite, was that it pollutes my local maven repository. I am the lead developer of Flexmojos and here we use the maven-invoker-plugin to populate a test local repo located in the test-harness' target directory and invoke child maven builds, resulting in the normal local maven repo staying untouched. After a mvn clean all of this is cleared keeping the normal local repo nice and clean. Yet another improvement proposal would be the way the resources in the MS SDKs are handled. I have seen a lot of systemPath stuff in the test poms. In Flex we too have different versions of SDKs that are installed to different places, which need to be accessed by the maven plugins. Instead of somehow accessing the files in their native locations, I created a mavenizer application, which creates Maven artifacts from the files in the SDKs and allows deploying these locally and in remote repositories. What do you think about creating a MS SDK mavenizer, which makes everything maveny and allows reducing the complexity of the plugins greatly? Chris -Ursprüngliche Nachricht- Von: christofer.d...@c-ware.de [mailto:christofer.d...@c-ware.de] Gesendet: Mittwoch, 13. November 2013 16:15 An: npanday-dev@incubator.apache.org Betreff: AW: Building NPanday Sure, I'll create an issue and attach a patch as soon as I'm home. Chris Von: Brett Porter [br...@porterclan.net] im Auftrag von Brett Porter [br...@apache.org] Gesendet: Mittwoch, 13. November 2013 13:25 An: npanday-dev@incubator.apache.org Betreff: Re: Building NPanday On 13 Nov 2013, at 6:44 pm, christofer.d...@c-ware.de wrote: Hi Brett, Well in my checkout I added two profiles default and minimal each containing only a modules section. Minimal only referencing the compiler-maven-plugin and the default containing the normal modules-section. I then disabled the Profile which automatically disables itself as soon as the bootstrap property is set. At least this way is buildable using mvn clean install -Pminimal mvn clean install without having to have any Prior Version available in any repo. I would much favour this Approach and it would be much more like other maven plugin Projects are Setup. Yep, sounds good to me - would you like to contribute that as a patch in JIRA? - Brett -- Brett Porter @brettporter http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Re: PPMC/Committer who is still active
I am a committer. Now walking through the list and find old mails. :-) Am 13.11.2013 um 17:13 schrieb Raphael Bircher r.birc...@gmx.ch: Hi at all I have a important question about the PPMC/Committers. Who is in the PPMC and who is committer. And how many of them are still active. We have to create a PPMC wich is able to manage the project. For this we need minimum three people. On the committer list I see 10 People with access to the SVN. Are they all sucked in a black hole or they waiting for better time ;-) Greetings Raphael
Re: Status of the project?
I like coding. But i hate mailing lists. :-) Am 13.11.2013 um 08:00 schrieb Brett Porter br...@apache.org: Hi Raphael, Thanks for checking in. You were right in another message, I was on holidays. There are people here willing to develop on the project, but time comes and goes for each. When we first brought the project into Apache, there was interest from several others, including some Apache committers, but it waned - and though we get occasional patches from new people, we've had trouble getting them on as committers. I'm glad to see the interest sparked by your message, which I'll respond to separately. In terms of mentors, we've gradually lost them over time, and I was unsuccessful in recruiting new ones from general@. We very much need an active mentor to keep us honest :) I hope you'll volunteer... For the overall status: We need to ship 1.5.0. Unfortunately, while the Windows build slave was broken for a while some changes were made that broke compatibility. They are good changes, but the regressions need to be fixed to ship a release. Aside from that, the project is reasonably stable. There are additions here and there for new releases that come out, and some potential architectural changes to take advantage of recent developments in .NET and Maven and cruft to clear out, but otherwise it pretty much does what you need it to do. That can make it hard to attract new contributors. One thing that could be improved is Mono support. While some things exist, the existing developers don't use it - so that's an area for new contributors to get involved. Another challenge I've found is that contributors tend to interact purely over JIRA or even reach out over Skype. It has been difficult to get more happening on this list as it should. I'll put this in as a report for this month if it is not too late. Thanks again for following up. Cheers, Brett On 8 Nov 2013, at 2:03 pm, Raphael Bircher r.birc...@gmx.ch wrote: Hi all My name is Raphael Bircher. I'm from the Incubator and doing shepherding for this project this month. In my review i saw nearly no activity in the last three months. There is also only one mentor who is inactive since one year, (according dev ml). The big question is: Is here still some willing and energy to develop this project, or not. You have not filled the bord report now, pleas do it. I will do my shepherd report anyway, and bring the topic to the incubator list. Greetings Raphael -- Brett Porter @brettporter http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Re: Status of the project?
good idea to make send jira-comments to dev! thats where I live - if so. Am 13.11.2013 um 13:28 schrieb Brett Porter br...@apache.org: On 13 Nov 2013, at 6:49 pm, christofer.d...@c-ware.de wrote: Hi Brett, Well I would strongly suggest to bring those discussions back here to the list, cause this is the usual way Apache Projects communicate ... in the Flex Project they always say: If it's not on the list, it didn't happen.. This too would give other People more an Impression that this Project is not dead. Yes, certainly agree. Additionally you could Setup Jira to mirror stuff Happening in Jira to this list (We had this in the Flex Project, but switched that off as soon as Speed increased). It currently goes to -commits - but I think in this case -dev makes more sense. I'll look into changing that. - Brett -- Brett Porter @brettporter http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Re: trunk
developing npanday on linux/mac doesnt really work. try on windows... _ Mobil versendet. Am 01.07.2013 um 21:00 schrieb David Akehurst d...@akehurst.net: Hi, I have just checked out the latest npanday trunk from https://svn.apache.org/repos/asf/incubator/npanday/trunk The build fails at NPanday :: Executable [INFO] NPanday :: Model :: Settings .. SUCCESS [1.498s] [INFO] NPanday :: Vendor . SUCCESS [1.463s] [INFO] NPanday :: Executable . FAILURE [2.460s] [INFO] NPanday :: Model :: Configuration Appenders ... SKIPPED there seems to be some test failures Failed tests: testCommandArgWithSpaces[0](npanday.executable.execution.CommandExecutorTest): (cmd.standardOut == expected). Values: expected = a b testCommandArgWithSpaces[1](npanday.executable.execution.CommandExecutorTest): (cmd.standardOut == expected). Values: expected = a b Tests in error: testCommandArgWithEmbeddedSingleQuotes_middle[0](npanday.executable.execution.CommandExecutorTest): NPANDAY-040-001: Could not execute: Command = /bin/sh -c echo 'a \' b', Result = 2 testCommandArgWithEmbeddedSingleQuotes_middle[1](npanday.executable.execution.CommandExecutorTest): NPANDAY-040-001: Could not execute: Command = /bin/sh -c echo 'a \' b', Result = 2 I am building on a linux machine, not windows, would this be why, or is there a problem with the current trunk? cheers
Re: [jira] [Commented] (NPANDAY-587) Support for .NET Framework 4.5
4.5 is an in-place-upgrade for 4.0. Will depend on if the systems has it installed or not. But maybe, if stated 4.5 we can somehow validate, that 4.5 is really installed. [ https://issues.apache.org/jira/browse/NPANDAY-587?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13667497#comment-13667497 ] Brett Porter commented on NPANDAY-587: -- New vendor information needs to be aded Support for .NET Framework 4.5 -- Key: NPANDAY-587 URL: https://issues.apache.org/jira/browse/NPANDAY-587 Project: NPanday Issue Type: Improvement Components: Maven Plugins Reporter: Brett Porter Assignee: Brett Porter Along with NPANDAY-586, check that projects using the new version of the framework build successfully. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
Re: building from SVN
I have tried to do npanday development on mono a couple of times, but given up every time. in this case the implementation does the quoting escaping wrong - but i think if you fix that, the next problem will arise: http://stackoverflow.com/questions/7254509/how-to-escape-single-quotes-in-bash-grep Am 25.03.2013 um 17:59 schrieb David Durham david.durham...@gmail.com: Mono JIT compiler version 2.10.8 (tarball Sat Oct 6 23:22:30 UTC 2012) Copyright (C) 2002-2011 Novell, Inc, Xamarin, Inc and Contributors. www.mono-project.com TLS: __thread SIGSEGV: altstack Notifications: epoll Architecture: amd64 Disabled: none Misc: softdebug LLVM: supported, not enabled. GC:Included Boehm (with typed GC and Parallel Mark) Linux 3.7.10-1-ARCH #1 SMP PREEMPT Thu Feb 28 09:50:17 CET 2013 x86_64 GNU/Linux Thanks, Dave On Fri, Mar 22, 2013 at 8:30 PM, Brett Porter br...@apache.org wrote: What operating system and version of Mono are you using? Others may have more input, as most of my NPanday work is done on a Windows virtual machine. Regards, Brett On 22/03/2013, at 3:05 PM, David Durham david.durham...@gmail.com wrote: Hi all, I'm getting the following error when I checkout from SVN and run bootstrap.sh. Anyone available to help? Thanks. Running npanday.executable.execution.switches.SwitchFormatTest Tests run: 4, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.581 sec Running npanday.executable.execution.QuotingAndEscapingBehaviourTest Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.003 sec Running npanday.executable.execution.CommandExecutorTest Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting [INFO] +--[ RUNNING: /bin/sh -c asdfasdf] [ERROR] | /bin/sh: asdfasdf: command not found [INFO] +--[ FAILED, result = 127, error output = true] Result is 127 Executing with x_unified_simple_quoting [INFO] +--[ RUNNING: /bin/sh -c mkdir /home/ddurham/work/npanday/npanday/components/dotnet-executable/target/test-resources/sampledirectory] [INFO] +--[ DONE ] Executing with x_unified_simple_quoting [INFO] +--[ RUNNING: /bin/sh -c mkdir '/home/ddurham/work/npanday/npanday/components/dotnet-executable/target/test-resources/sample directory'] [INFO] +--[ DONE ] Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting [INFO] +--[ RUNNING: /bin/sh -c echo ] [INFO] | [INFO] +--[ DONE ] npanday.executable.execution.UnifiedShellCommandExecutor@69a0bc8a Executing with x_unified_simple_quoting [INFO] +--[ RUNNING: /bin/sh -c echo x] [INFO] | x [INFO] +--[ DONE ] Executing with x_unified_simple_quoting [INFO] +--[ RUNNING: /bin/sh -c echo 'a b'] [INFO] | a b [INFO] +--[ DONE ] Executing with x_unified_simple_quoting [INFO] +--[ RUNNING: /bin/sh -c echo 'a \' b'] [ERROR] | /bin/sh: -c: line 0: unexpected EOF while looking for matching `'' [ERROR] | /bin/sh: -c: line 1: syntax error: unexpected end of file [INFO] +--[ FAILED, result = 1, error output = true] Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_simple_quoting Executing with x_unified_custom_quoting Executing with x_unified_custom_quoting Executing with x_unified_custom_quoting Executing with x_unified_custom_quoting Executing with x_unified_custom_quoting Executing with x_unified_custom_quoting Executing with x_unified_custom_quoting Executing with x_unified_custom_quoting Executing with x_unified_custom_quoting Executing with x_unified_custom_quoting Executing with x_unified_custom_quoting [INFO] +--[ RUNNING: /bin/sh -c asdfasdf] [ERROR] | /bin/sh: asdfasdf: command not found [INFO] +--[ FAILED, result = 127, error output = true] Result is 127 Executing with x_unified_custom_quoting [INFO] +--[ RUNNING: /bin/sh -c mkdir
Re: NPanday 1.5 reease
Hi, Brett and I had some trys to release NPanday 1.5 - but couldn't wrap it up yet. We are still actively developing it - but it goes in rushes. @Brett: from your point, what is in the way for releasing? Failing ITs? But you can configure the release plugin to accept snapshot plugin versions. _ Lars Am 28.12.2012 um 13:33 schrieb Adrián Boimvaser adrianboimva...@gmail.com: Dear NPanday developers, Like a year ago I contributed a couple patches for NPANDAY-510 NPANDAY-533 and I've been waiting for a the next release since then. Now that I'm about to release a my project I'm being forced to create a private release on my corporate repository or replace wix-maven-plugin with exec-maven-plugin. Are there any plans to release NPanday 1.5? Is development still active? Thanks in advance. Best regards, Adrián Boimvaser
Re: svn commit: r1390287 - in /incubator/npanday/npanday-its/trunk/src/test/java/npanday/its: AbstractNPandayIntegrationTestCase.java NPandayIT0036InstalledArtifactsVerificationTest.java
that is the problem, when not trusting the build :) i made the its fail, but i didn't think it was my fault, since builds fail often :-) Am 26.09.2012 um 07:49 schrieb br...@apache.org: Author: brett Date: Wed Sep 26 05:49:15 2012 New Revision: 1390287 URL: http://svn.apache.org/viewvc?rev=1390287view=rev Log: update test now that sources are not necessarily copied to target/build-sources Modified: incubator/npanday/npanday-its/trunk/src/test/java/npanday/its/AbstractNPandayIntegrationTestCase.java incubator/npanday/npanday-its/trunk/src/test/java/npanday/its/NPandayIT0036InstalledArtifactsVerificationTest.java Modified: incubator/npanday/npanday-its/trunk/src/test/java/npanday/its/AbstractNPandayIntegrationTestCase.java URL: http://svn.apache.org/viewvc/incubator/npanday/npanday-its/trunk/src/test/java/npanday/its/AbstractNPandayIntegrationTestCase.java?rev=1390287r1=1390286r2=1390287view=diff == --- incubator/npanday/npanday-its/trunk/src/test/java/npanday/its/AbstractNPandayIntegrationTestCase.java (original) +++ incubator/npanday/npanday-its/trunk/src/test/java/npanday/its/AbstractNPandayIntegrationTestCase.java Wed Sep 26 05:49:15 2012 @@ -310,24 +310,9 @@ public abstract class AbstractNPandayInt return target/comments.xml; } -protected String getBuildSourcesMain( String fileName ) -{ -return getBuildFile( build-sources, fileName ); -} - protected String getBuildSourcesGenerated( String fileName ) { -return getBuildSourcesMain( fileName ); -} - -protected String getTestSourcesMain( String fileName ) -{ -return getBuildFile( build-test-sources, fileName ); -} - -protected String getTestSourcesGenerated( String fileName ) -{ -return getTestSourcesMain( fileName ); +return getBuildFile( build-sources, fileName ); } protected String getBuildFile( String buildDirectory, String fileName ) Modified: incubator/npanday/npanday-its/trunk/src/test/java/npanday/its/NPandayIT0036InstalledArtifactsVerificationTest.java URL: http://svn.apache.org/viewvc/incubator/npanday/npanday-its/trunk/src/test/java/npanday/its/NPandayIT0036InstalledArtifactsVerificationTest.java?rev=1390287r1=1390286r2=1390287view=diff == --- incubator/npanday/npanday-its/trunk/src/test/java/npanday/its/NPandayIT0036InstalledArtifactsVerificationTest.java (original) +++ incubator/npanday/npanday-its/trunk/src/test/java/npanday/its/NPandayIT0036InstalledArtifactsVerificationTest.java Wed Sep 26 05:49:15 2012 @@ -38,8 +38,9 @@ public class NPandayIT0036InstalledArtif File testDir = ResourceExtractor.simpleExtractResources( getClass(), /NPandayIT0036InstalledArtifactsVerificationTest ); Verifier verifier = getVerifier( testDir ); verifier.executeGoal( install ); -verifier.assertFilePresent( -new File( testDir, getAssemblyFile( NPandayIT0036, 1.0.0.0, exe ) ).getAbsolutePath() ); +String exe = +new File( testDir, getAssemblyFile( NPandayIT0036, 1.0.0.0, exe ) ).getAbsolutePath(); +verifier.assertFilePresent( exe ); verifier.assertFilePresent( new File( testDir, getAssemblyFile( NPandayIT0036-test, 1.0.0.0, dll ) ).getAbsolutePath() ); verifier.assertFilePresent( new File( testDir, getAssemblyFile( test-assemblies/NPandayIT0036, 1.0.0.0, @@ -49,27 +50,8 @@ public class NPandayIT0036InstalledArtif verifier.assertFilePresent( new File( testDir, getAssemblyFile( test-assemblies/NUnit.Framework, 1.0.0.0, dll ) ).getAbsolutePath() ); -verifier.assertFilePresent( new File( testDir, getBuildSourcesMain( Module1.vb ) ).getAbsolutePath() ); -verifier.assertFilePresent( new File( testDir, getBuildSourcesMain( folder/Module2.vb ) ).getAbsolutePath() ); -verifier.assertFileNotPresent( -new File( testDir, getBuildSourcesMain( should-not-be-copied.txt ) ).getAbsolutePath() ); -verifier.assertFileNotPresent( -new File( testDir, getBuildSourcesMain( should-not-be-copied.xml ) ).getAbsolutePath() ); -verifier.assertFileNotPresent( -new File( testDir, getBuildSourcesMain( folder/should-not-be-copied-2.txt ) ).getAbsolutePath() ); -verifier.assertFileNotPresent( -new File( testDir, getBuildSourcesMain( folder/should-not-be-copied-2.xml ) ).getAbsolutePath() ); - -verifier.assertFilePresent( new File( testDir, getTestSourcesMain( Module1.vb ) ).getAbsolutePath() ); -verifier.assertFilePresent( new File( testDir, getTestSourcesMain( folder/Module2.vb ) ).getAbsolutePath() ); -
Re: moving npanday.org to the ASF
Hi Brett, sounds good to me! When starting to offer dotnet-librarys extracted from nuget-packages, we have to think through .NET CLR versions - since one nuget-package can contain multiple libs for multiple clr-versions Either we include the clr-version in the library version (e.g. 1.2.0-net40), or we offer multiple repositories for 2.0, 3.0, 3.5, 4.0, and so on… People can then define those in the right order to configure precedence. _ Lars Am 29.05.2012 um 12:44 schrieb Brett Porter: The ASF infra team are waiting on some sign of consensus before moving forward, so if anyone has an opinion - it would be great to hear it now. On 24/05/2012, at 11:49 AM, Brett Porter wrote: A couple of sounds good would be handy too if you don't have any concerns :) On 22/05/2012, at 5:12 PM, Brett Porter wrote: Hi, I have a long overdue note to transfer the domain name from its current location to the ASF. The current uses are: - repo.npanday.org points to repo.maestrodev.com, which houses the artifacts to build NPanday = 1.4 with Maven as a convenience (without having to construct your own repository), as well as the releases prior to entering the ASF. Pretty similar to how repo.maven.apache.org is handled. - www.npanday.org, which is redirected to http://incubator.apache.org/npanday I don't propose we change either of those at the moment. Hopefully with the changes Lars has done on trunk the 1.5.0 release will not require the repository to build any more, and we can have a separate discussion on how we might make it easier for users to obtain other artifacts (again, his work on NuGet integration will help here). Any concerns with this plan? Cheers, Brett -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter http://twitter.com/brettporter -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter http://twitter.com/brettporter -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter http://twitter.com/brettporter
Re: exe archiver for Maven assembly plugin?
Hallo Christopher, seems like you find out things … Can't really grasp what you are trying to do. _ Lars - auch deutsch :-) - Am 02.03.2012 um 10:39 schrieb christofer.d...@c-ware.de: Hi, I am currently trying to setup a Maven build consisting of a mixed configuration (I am using 1.5-incubator-SNAPSHOT): 1. Java Library (jar packaging) 2. Windows Executable (winexe packaging) The release should be a jar containing the Java library with an embedded exe (So I can load the binary from within my Java code) When using the Maven assembly plugin, all I get is this: [ERROR] Failed to execute goal org.apache.maven.plugins:maven-assembly-plugin:2.2.1:single (make-assembly) on project lib-psexec-assembly: Failed to create assembly: Error adding file-set for 'de.cware.utils:lib-psexec-service:winexe:1.0.0-SNAPSHOT' to archive: Error adding archived file-set. PlexusIoResourceCollection not found for: C:\Maven-Repository\de\cware\utils\lib-psexec-service\1.0.0-SNAPSHOT\lib-psexec-service-1.0.0-SNAPSHOT.exe: No such archiver: 'exe'. - [Help 1] So it seems there is no archiver available to handle exe resources. So my question now is: How do I configure the assembly to do this or is the such an assembler simply missing from NPanday and one should be implemented. I would volunteer implementing one, if this would be the way to go, I just don't want to implement something that isn't needed. Chris [ C h r i s t o f e r D u t z ] C-Ware IT-Service Inhaber Dipl. Inf. Christofer Dutz Karlstraße. 104, 64285 Darmstadt fon: 0 61 51 / 27315 - 61 fax: 0 61 51 / 27315 - 64 mobil: 0171 / 7 444 2 33 email: christofer.d...@c-ware.demailto:christofer.d...@c-ware.de http://www.c-ware.dehttp://www.c-ware.de/ UStId-Nr. DE195700962
Handling configuration when packaging with NPanday
Hi guys, I have some thoughts to share around configuration for the new packaging implementations in NPanday 1.5.0: cloud service, dotnet application, msdeploy/web applications. Sorry for the long e-mail, but that really is important for me as dev, coach, npanday contributor, and user :-) First, lets have a look at how Visual Studio handles the documentation for the various types: web.config / app.config Before Visual Studio 2008 one had to completely manually handle how the configuration is deployed onto other environments. I usually then deployed the config file initially only; after that i handled all changes to the configuration manually. In 2008 three new features are added that make this easier. 1) configuration can be split up in many files by using theSection configurationFile=../ on any section. Then you can deploy most of the config every time and just handle environment-specific configs in separate files like connectionStrings.config++ 2) Config transforms (XDT); with right click on a web.config you can choose Add Config Transforms which will add a sub-file for each configuration in your Configuration Mangager, hence Web.Debug.config, Web.Release.config; those files are quite intelligent patch definitions applied to the main web.config on publish. By default this is only available for the main web.config, but not for sublevel web.configs nor for the app.config. 3) Parameters. Here you can define XPath-replacement-rules for various files; when unpacking/deploying a package with MSDeploy, by providing your environment-specific values for the defined parameters, the replacement will be applied to the files contained in the package. Sadly this is not integrated in Visual Studio, but has to be done manually. Azure Azure configurations consist of two parts: Cloud Service Definition (csdef) and Cloud Service Configuration (cscfg). The latter is meant to be environment-specific and Visual Studio allows you to configure multiple copies (Cloud, Local, ...). Csdef on the other hand is not (meant to be) environment specific and hence makes it into the Cloud Service Package (cspkg). Sadly in the UI it is not easy to see what config setting is stored where; however when turning on a specifc configuration (! ALL) some options will be disabled; these are the ones that go into the csdef: endpoints, vm size, ... NPanday In NPanday we would like to unify how configuration is handled. The main goal must be to create and distribute environment-unspecific packages. NPanday should prevent the developer from accidentally including sensitive configuration elements like connection strings e.g. in (public) packages. This is why we do not want to just xcopy all config files into the packages. But rather we'd like to find a good generic approach to handling config files. Wanted behaviour: 1) Instead of Debug and Release, Cloud and Local we use Package all over. 2) For Azure, by default, ServiceConfiguration.Azure.cscfg is distributed; it is a COPY, not a transformation. Should we support transformation here? XDT, XSL? 3) For Webs and Apps we use web. or app.package.config and copy it over as web. or app.config. Currently this also is a copy, but the plan is to make it a XDT transformation that then is applied to the main web. or app.config. If the *.package.config file doesn't exist, it prints a warning. I think it should actually FAIL!! But should it rather fallback to web/app.config with the risk of including sensitive settings in the packages? 4) An application can have multiple configs (sublevel web.configs or splitted-out configs) which is not supported by the current implementation (can add them through a component descriptor though). I suggest we handle those in the same way? XDT applyable? Fallback? Happy for any thoughts/answers around this issue. _ Lars
Re: NPanday future direction for multi-platform
Question of design and naming. I certainly can make XDT pluggable, using the MSBuild version on Windows and make it fail, if it isn't available. . With MSBuild I can try to make it easy to integrate xbuild later on. _ Mobil versendet. Am 05.01.2012 um 11:53 schrieb Brett Porter br...@apache.org: How much of this can be handled through the execution framework that is now in place, so that a default implementation that works with Windows can be built, but then someone inclined to use it could (now or later) implement an alternative for Mono? - Brett On 05/01/2012, at 8:12 PM, Lars Corneliussen wrote: Hi, we have now integrated azure and msdeploy in 1.5.0; functionality that certainly isn't and probably won't be available on other platforms than Microsoft.NET/Windows. But still we have surrounding things. dotnet-application, for example, supports packaging of runnables (config + exe/dll + deps) in a zip; that is available to all platforms. https://issues.apache.org/jira/browse/NPANDAY-493 Now, I want to create an XDT-plugin (XmlDocumentTransform) that enables transformation of config-files in the packaging process (usually you wouldn't like your local web.config/app.config) to be packaged unchanged. XDT is provided by MSBuild, but there is also an alternative implementation available, that maybe runs, or at least could be changed to run on mono, too (http://code.google.com/p/xdt/). How much extra effort would it be worth to choose that implementation instead of Microsofts, in order to support it on !win? Alternatively we could also offer XSL-transformations only. Functional, though not as comfortable as XDT. Anybody willing to help making XDT work on Mono? https://issues.apache.org/jira/browse/NPANDAY-486 Then, we also think to implement MSBuild. Again, we have the option to support Xbuild as well; does anybody want to support, doing this? any comments? Disclaimer: I'm not (yet) using Mono at all. Not on Windows, neither on my Mac. _ Lars -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter http://twitter.com/brettporter
Re: svn commit: r1226446 - in /incubator/npanday/trunk/components/dotnet-core/src/main: groovy/npanday/lifecycle/LifecycleConfigurationGenerator.groovy java/npanday/ArtifactType.java
Same should be done for dotnet-application _ Mobil versendet. Am 02.01.2012 um 16:49 schrieb br...@apache.org: Author: brett Date: Mon Jan 2 15:49:48 2012 New Revision: 1226446 URL: http://svn.apache.org/viewvc?rev=1226446view=rev Log: [NPANDAY-488] add includeDependencies to the artifact type for MSDeploy packages, so that Azure projects don't have a problem trying to resolve the dependencies Modified: incubator/npanday/trunk/components/dotnet-core/src/main/groovy/npanday/lifecycle/LifecycleConfigurationGenerator.groovy incubator/npanday/trunk/components/dotnet-core/src/main/java/npanday/ArtifactType.java Modified: incubator/npanday/trunk/components/dotnet-core/src/main/groovy/npanday/lifecycle/LifecycleConfigurationGenerator.groovy URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/components/dotnet-core/src/main/groovy/npanday/lifecycle/LifecycleConfigurationGenerator.groovy?rev=1226446r1=1226445r2=1226446view=diff == --- incubator/npanday/trunk/components/dotnet-core/src/main/groovy/npanday/lifecycle/LifecycleConfigurationGenerator.groovy (original) +++ incubator/npanday/trunk/components/dotnet-core/src/main/groovy/npanday/lifecycle/LifecycleConfigurationGenerator.groovy Mon Jan 2 15:49:48 2012 @@ -59,6 +59,10 @@ class LifecycleConfigurationGenerator { configuration{ extension typeDef.extension type typeDef.packagingType +if (typeDef.includesDependencies) +{ + includesDependencies typeDef.includesDependencies +} } } } Modified: incubator/npanday/trunk/components/dotnet-core/src/main/java/npanday/ArtifactType.java URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/components/dotnet-core/src/main/java/npanday/ArtifactType.java?rev=1226446r1=1226445r2=1226446view=diff == --- incubator/npanday/trunk/components/dotnet-core/src/main/java/npanday/ArtifactType.java (original) +++ incubator/npanday/trunk/components/dotnet-core/src/main/java/npanday/ArtifactType.java Mon Jan 2 15:49:48 2012 @@ -176,7 +176,7 @@ public enum ArtifactType AZURE_CLOUD_SERVICE_CONFIGURATION (azure-cloud-service-configuration, null, cscfg), /* MSDeploy support */ -MSDEPLOY_PACKAGE (msdeploy-package, null, msdeploy.zip); +MSDEPLOY_PACKAGE (msdeploy-package, null, msdeploy.zip, true); /** * The extension used for the artifact(netmodule, dll, exe) @@ -194,6 +194,12 @@ public enum ArtifactType private String targetCompileType; /** + * Whether to flag the artifact type as including its dependencies (do not pass them on transitively), or not (pass + * them on transitively) + */ +private boolean includesDependencies = false; + +/** * Constructor */ ArtifactType( String packagingType, String targetCompileType, String extension ) @@ -204,6 +210,17 @@ public enum ArtifactType } /** + * Constructor + */ +ArtifactType( String packagingType, String targetCompileType, String extension, boolean includesDependencies ) +{ +this.packagingType = packagingType; +this.targetCompileType = targetCompileType; +this.extension = extension; +this.includesDependencies = includesDependencies; +} + +/** * Returns extension used for the artifact(netmodule, dll, exe). * * @return Extension used for the artifact(netmodule, dll, exe). @@ -233,6 +250,11 @@ public enum ArtifactType return targetCompileType; } +public boolean isIncludesDependencies() +{ +return includesDependencies; +} + /** * Returns artifact type for the specified packaging name *
Re: svn commit: r1226489 - in /incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src: main/csharp/Converter/Algorithms/ test/resource/MvcApplication1/MvcApplication1/ test/resou
XDT will then also require a new file to be added. _ Mobil versendet. Am 02.01.2012 um 18:35 schrieb br...@apache.org: Author: brett Date: Mon Jan 2 17:35:35 2012 New Revision: 1226489 URL: http://svn.apache.org/viewvc?rev=1226489view=rev Log: [NPANDAY-488] adjust default config behaviour until XDT is in place, so that a reasonable Web/app.config is Modified: incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/ASPNetPomConverter.cs incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/AzureWorkerPomConverter.cs incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/test/resource/MvcApplication1/MvcApplication1/pom.test incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/test/resource/NPANDAY_480_AzureSupportOneWebRole/HelloWorld_WebRole/pom.test incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/test/resource/NPANDAY_480_CloudServiceWithMultipleRoles/HelloWorld_WebRole/pom.test incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/test/resource/NPANDAY_480_CloudServiceWithMultipleRoles/HelloWorld_WorkerRole/pom.test incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/test/resource/NPANDAY_480_CloudServiceWithWorkerRole/HelloWorld_WorkerRole/pom.test incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/test/resource/NPANDAY_488/NPANDAY_488_MSDeployPackageSimpleWebApp/pom.test Modified: incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/ASPNetPomConverter.cs URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/ASPNetPomConverter.cs?rev=1226489r1=1226488r2=1226489view=diff == --- incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/ASPNetPomConverter.cs (original) +++ incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/ASPNetPomConverter.cs Mon Jan 2 17:35:35 2012 @@ -40,17 +40,14 @@ namespace NPanday.ProjectImporter.Conver Liststring goals = new Liststring(); goals.Add(assemble-package-files); -foreach (Content content in projectDigest.Contents) -{ -if (content.IncludePath.Equals(web.package.config, System.StringComparison.InvariantCultureIgnoreCase)) -{ -goals.Add(process-web-config); -} -} +goals.Add(process-web-config); Plugin aspnetPlugin = AddPlugin(org.apache.npanday.plugins, aspnet-maven-plugin, null, false); AddPluginExecution(aspnetPlugin, prepare-package, goals.ToArray(), null); +// TODO: until XDT works, just use Web.config itself +AddPluginConfiguration(aspnetPlugin, webConfig, Web.config); + Plugin msdeployPlugin = AddPlugin(org.apache.npanday.plugins, msdeploy-maven-plugin, null, false); AddPluginExecution(msdeployPlugin, create-msdeploy-package, new string[] { create-package }, null); Modified: incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/AzureWorkerPomConverter.cs URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/AzureWorkerPomConverter.cs?rev=1226489r1=1226488r2=1226489view=diff == --- incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/AzureWorkerPomConverter.cs (original) +++ incubator/npanday/trunk/dotnet/assemblies/NPanday.ProjectImporter/Engine/src/main/csharp/Converter/Algorithms/AzureWorkerPomConverter.cs Mon Jan 2 17:35:35 2012 @@ -40,18 +40,15 @@ namespace NPanday.ProjectImporter.Conver Liststring goals = new Liststring(); goals.Add(assemble-package-files); -foreach (Content content in projectDigest.Contents) -{ -if (content.IncludePath.Equals(app.package.config, System.StringComparison.InvariantCultureIgnoreCase)) -{ -goals.Add(process-app-config); -} -} +goals.Add(process-app-config); goals.Add(package); Plugin plugin = AddPlugin(org.apache.npanday.plugins, application-maven-plugin, null, false); AddPluginExecution(plugin, package-application, goals.ToArray(),
Re: Contributing to wix-maven-plugin
Hi Adrián, I'm a little bit christmas-offline now. So Sit and wait can happen sometimes :) There is a way to become an NPanday developer, but it'll require a couple of good patches that drive the project in the right direction. It also usually takes some time until patches are available in a new release. When I originally had a series of patches for NPanday, I had a private git repository based on https://github.com/apache/npanday to keep a fork for a while. I also then wrote a utility that helped me creating SVN-compatible patches from git feature branches here: http://lcorneliussen.de/utils/git-svn/github.com _ Lars Am 22.12.2011 um 15:53 schrieb Adrián Boimvaser: Hi Lars? What's next? Sit and wait? I made some refactorings and improvements to wix-maven plugin I would also like to contribute. Is there a way I can become a NPanday developer? Thanks. Adrián. 2011/12/21 Adrián Boimvaser adrianboimva...@gmail.com Thanks Lars, I created the issue and attached the patch. https://issues.apache.org/jira/browse/NPANDAY-510 Adrián. 2011/12/21 Lars Corneliussen m...@lcorneliussen.de Hi Adrián, happy to hear that. The best is, if you create an issue and attach the patch here: https://issues.apache.org/jira/secure/CreateIssue.jspa?pid=12311182issuetype=4Create=Create Is that ok for you? _ Lars Am 21.12.2011 um 13:35 schrieb Adrián Boimvaser: Hi, I'd like to make a contribution to wix-maven-plugin. Where should I send the patch? Thanks. Adrián Boimvaser
Re: svn commit: r1221092 - /incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java
why is it skipped on !Windows? _ Mobil versendet. Am 20.12.2011 um 05:06 schrieb br...@apache.org: Author: brett Date: Tue Dec 20 04:06:05 2011 New Revision: 1221092 URL: http://svn.apache.org/viewvc?rev=1221092view=rev Log: show more information for settings generator, so there's less confusion when it is skipped Modified: incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java Modified: incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java?rev=1221092r1=1221091r2=1221092view=diff == --- incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java (original) +++ incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java Tue Dec 20 04:06:05 2011 @@ -157,6 +157,7 @@ public class SettingsGeneratorMojo { if ( !System.getProperty( os.name ).contains( Windows ) ) { +getLog().info( NPANDAY-119-001: Skipping settings generation, only valid on Windows operating systems ); return false; } @@ -175,11 +176,13 @@ public class SettingsGeneratorMojo { if ( isFrameworkVersionExisting( frameworkVersion ) ) { +getLog().info( NPANDAY-119-002: Skipping settings generation, requested framework + frameworkVersion + is already configured ); return false; } } catch ( IOException e ) { +getLog().info( NPANDAY-119-003: Skipping settings generation, error reading existing file: + e.getLocalizedMessage() ); return false; } }
Re: svn commit: r1221092 - /incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java
Also an error reading the current file should just be thrown; it is an error that the user has to fix. most certainly he edited it himself. NPanday should then just fail. _ Mobil versendet. Am 20.12.2011 um 05:06 schrieb br...@apache.org: Author: brett Date: Tue Dec 20 04:06:05 2011 New Revision: 1221092 URL: http://svn.apache.org/viewvc?rev=1221092view=rev Log: show more information for settings generator, so there's less confusion when it is skipped Modified: incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java Modified: incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java?rev=1221092r1=1221091r2=1221092view=diff == --- incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java (original) +++ incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/javabinding/src/main/java/NPanday/Plugin/Settings/SettingsGeneratorMojo.java Tue Dec 20 04:06:05 2011 @@ -157,6 +157,7 @@ public class SettingsGeneratorMojo { if ( !System.getProperty( os.name ).contains( Windows ) ) { +getLog().info( NPANDAY-119-001: Skipping settings generation, only valid on Windows operating systems ); return false; } @@ -175,11 +176,13 @@ public class SettingsGeneratorMojo { if ( isFrameworkVersionExisting( frameworkVersion ) ) { +getLog().info( NPANDAY-119-002: Skipping settings generation, requested framework + frameworkVersion + is already configured ); return false; } } catch ( IOException e ) { +getLog().info( NPANDAY-119-003: Skipping settings generation, error reading existing file: + e.getLocalizedMessage() ); return false; } }
Re: Jenkins CI is up
agree. _ Mobil versendet. Am 08.12.2011 um 17:53 schrieb Brett Porter br...@apache.org: On 08/12/2011, at 6:53 PM, Lars Corneliussen wrote: Am 06.12.2011 um 22:27 schrieb Brett Porter: On 07/12/2011, at 5:15 AM, Lars Corneliussen wrote: That is good news!!! Some questions: 1) what about embedded integration tests? should those run in the normal build? guess it takes too long? what do you mean by embedded? The wix-like ones, or running the current ITs without forking Maven? The wix-like ones. I'm writing some of these for new plugins, too. That's currently an extra 3 minutes, so I imagine with more tests it would take too long for quick feedback. It would also be helpful to get a separate build history for the main build and the one with ITs on, IMO. - Brett -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Draft: Incubator PMC/Board report for Dec 2011
Hi. Here is my draft for the PMC/Board report. Any comments? NPanday NPanday allows projects using the .NET framework to be built with Apache Maven. NPanday allows .NET projects to be converted into Maven projects thus allowing them to fully utilize the other technologies driven by Maven. NPanday has been incubating since August 2010. We are now in the process of releasing NPanday-1.4.1-incubating and at the same time developing NPanday-1.5.0-incubating, which will come with azure- and silverlight support as well as quite extensive core refactorings removing historical burden. We are also happy to report that our community continues to gain new people in terms of patch contributors and reporters on issues. The top priorities towards graduation are: * regular release * guide regular patch contributors towards becoming committers * improve engagement of existing committers There are no issues for the Incubator PMC or board at this time.
Re: What do you use MSBuild-plugin for? Rethinking MSBuild / Xbuild integration
I think we can offer both versions in two separate MSBuild lifecycles: integrated and delegated Details on this idea are all documented on the issue, but let me give a brief summary of the plan: The integrated lifecycle would still leave use NPanday to do the actual compile, but generate sources and resources which then will automatically be added to pom/build/resources and sources (in-memory) The delegating lifecycle would leave the artifact generation to MSBuild. But NPanday would surround it and ensure MSBuild to respect dependencies correctly and use correct output paths, e.g. If people then just want to execute MSBuild as-is, I'd point them to the custom-lifecycle and maven-exec-plugin - IMHO we shouldn't support that requirement. _ Lars Am 22.11.2011 um 05:44 schrieb Brett Porter: Hi Lars, Those requirements look good to me. While I prefer that approach, I wonder if it might become too complicated in some projects. I'm still curious if there is a usecase for NPANDAY-244 to provide a simpler (though less rich) build mechanism that effectively delegates to msbuild, then allows NPanday users to augment it with some dependency/distribution management and the execution of additional plugins in the lifecycle. Cheers, Brett (BTW, dropping -users CC - to avoid cross-posting it might be better just to point them to the -dev@ thread in future) On 22/11/2011, at 3:34 AM, Lars Corneliussen wrote: Hello dear NPanday users and developers, we think of reimplementing the integration with MSbuild since the current solution is quite fragile and erroneous. I have started to document my thoughts here and i'd be really happy to get feedback from you all! Especially I'd like to hear if, and how, you use the current MSBuild plugin. Please anwer to this Mail, or even better, write a comment on the issue. https://issues.apache.org/jira/browse/NPANDAY-486 _ Lars Apache NPanday Committer -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Discontinue or update the vsinstaller plugin !?
Hi again, the maven-vsinstaller-plugin is outdated. On my machine it even fails, because it can't find a component descriptor for dotnet-module - vsinstaller has it's own Components.xml in order to be able to run standalone. If we want to keep it up to date, we would have to: 1) Remove the Components.xml and instead create a package (Assembly-plugin) in VisualStudio.Addin that contains needed DLLs and the .AddIn-file and then reuse the same for both the wix-based Windows installer and vsinstaller-plugin (see https://issues.apache.org/jira/browse/NPANDAY-293). Then, on install, only override the bin-path for the DLL. 2) Update the documentation of the plugin (usage). 3) Update the documentation of how to debug NPanday Addin (because then the developer can just use the vsinstaller) _ Lars
Re: Discontinue or update the vsinstaller plugin !?
Actually I was wrong - it's just that you can't run the VS installer plugin while you are located in some Maven Project that contains dotnet-module packaging. Am 21.11.2011 um 12:26 schrieb Lars Corneliussen: Hi again, the maven-vsinstaller-plugin is outdated. On my machine it even fails, because it can't find a component descriptor for dotnet-module - vsinstaller has it's own Components.xml in order to be able to run standalone. If we want to keep it up to date, we would have to: 1) Remove the Components.xml and instead create a package (Assembly-plugin) in VisualStudio.Addin that contains needed DLLs and the .AddIn-file and then reuse the same for both the wix-based Windows installer and vsinstaller-plugin (see https://issues.apache.org/jira/browse/NPANDAY-293). Then, on install, only override the bin-path for the DLL. 2) Update the documentation of the plugin (usage). 3) Update the documentation of how to debug NPanday Addin (because then the developer can just use the vsinstaller) _ Lars
What do you use MSBuild-plugin for? Rethinking MSBuild / Xbuild integration
Hello dear NPanday users and developers, we think of reimplementing the integration with MSbuild since the current solution is quite fragile and erroneous. I have started to document my thoughts here and i'd be really happy to get feedback from you all! Especially I'd like to hear if, and how, you use the current MSBuild plugin. Please anwer to this Mail, or even better, write a comment on the issue. https://issues.apache.org/jira/browse/NPANDAY-486 _ Lars Apache NPanday Committer
Re: Remove repitetive namespace folder segments
Then we have to drop src/main/csharp, too. Tests could be in a subfolder 'Test' or 'Tests', resources are kept together with normal sources. That would be most common while adding the possibility to have tests together with the code, but omit it in the compiled assembly. Most .NET-Developers will fi d that to be good news and it is optional anyway. _ Lars _ Mobil versendet. Am 15.11.2011 um 09:57 schrieb Brett Porter br...@apache.org: Sounds good to me. The conventional directory layout should be what most .NET developers expect it to be (but still configurable). - Brett On 14/11/2011, at 11:25 PM, Lars Corneliussen wrote: Hi guys, in java it is common to have package folders inside 'src/main/java', as for example 'npanday/plugins/compile' - in .NET on the other hand we just configure a root namespace in the project file and then start with folders from there. I'd like to do that in NPanday, too. That means remove the path-segments that repeat the default namespace per project. Any objections? _ Lars -- Brett Porter br...@apache.org http://brettporter.wordpress.com/
Re: NPanday Visibility
Agree! Java dev's will still find it :) NPanday / Maven is about much more than dependency management; but still, without a good repository containing everything (Maven Central / Nuget Gallery), it will be hard to reach the masses. Integration with Nuget Gallery, and a new fresher Website will help us gaining momentum. Currently, when you visit http://incubator.apache.org/npanday/ you need 5 minutes to figure out what it is about. Try http://ninject.org/ ; after 10 seconds, you'll be able to tell me what it is about. The same with http://nuget.org/ _ Lars Am 13.11.2011 um 01:02 schrieb Brett Porter: Currently the site is a bit oriented towards Maven users who want to build .NET projects. Perhaps we need some HOWTO's titled .NET dependency management, oriented more at bringing NPanday's capability to the conventional .NET user? On 09/11/2011, at 5:05 AM, Lars Corneliussen wrote: Hi guys, I got a comment on my blog: Hi, I think, it’s a great project! Please, make this project more visible, we are using Maven for Java and we searched for “.net dependency management”. All we found was some alpha projects at codplex (refix and crude have around 30 downloads). After an extensive search we have found NuGet and OpenWrap. I have found this project completely accidentally!!! It would be really useful to get least some google hits to npanday on the first page…. Any concrete ideas on how to make NPanday more visible? _ Lars -- Brett Porter br...@apache.org http://brettporter.wordpress.com/
Remove repitetive namespace folder segments
Hi guys, in java it is common to have package folders inside 'src/main/java', as for example 'npanday/plugins/compile' - in .NET on the other hand we just configure a root namespace in the project file and then start with folders from there. I'd like to do that in NPanday, too. That means remove the path-segments that repeat the default namespace per project. Any objections? _ Lars
[vote] NPanday Source: Move on to VS 2010, but still target CLR 2.0 + .NET 3.5
Hi, it would be great, if we could use the many new language features introduced in C# 3.0 and 4.0. If I implement multi-targeting (https://issues.apache.org/jira/browse/NPANDAY-404), is it then OK to require .NET 4.0 tools and Visual Studio 2010 for all NPanday developers? _ Lars
Re: How can I contribute? (and a bug feature request)
Hi Stoyan, welcome to the list! We have some Get Involved documentation here: http://incubator.apache.org/npanday/docs/1.4.0-incubating/developers/index.html Preamble: I'm a .NET-guy, as you easily can see on my blog (http://lcorneliussen.de/blog) I agree with you; .NET is way behind Java in terms of build tools. It was sad for me that the guys at Microsoft that built Nuget (and Guys that built the preceeding Nu) didn't even know about Apache Maven - and neither about NPanday. The .NET-Community struggles to learn from other communities. I joined the team in October 2010 because I think NPanday is absolutely the right horse to bet on. But still, there is loads of work to do. One more thing. No offense but the current code base (I have only looked at the .NET code so far) needs some cleaning up (TODOs, XML parsing, etc.). Let me know if you're interested in a code review and to who I should e-mail it, also I'll be more than glad to do most of the clean up myself if I'd be allowed to push/propose changes. I guess that about 50% of the Java code is simply dispensable. And the .NET code is quite badly organized. I think we rather need concrete improvement suggestions + patches - in small steps, than a code review :-) A year ago I started some work on the VS plugin, refactoring to command pattern, but I didn't manage to finish it: https://github.com/lcorneliussen/npanday/commit/5ed038f9bb38a957d40d14bba8895ac9337e8aaa I also commented on the two issues you mentioned/submitted. _ Lars https://github.com/lcorneliussen/npanday/commit/5ed038f9bb38a957d40d14bba8895ac9337e8aaa Am 01.11.2011 um 21:44 schrieb Stoyan Damov: Hi guys, First, sorry if the I'm not following the protocol for sending a single thing in one e-mail (e.g. this is a question, a bug report, and a feature request) but I couldn't find a How to send e-mails to this list link. Our company is evaluating NPanDay for use by our the .NET team (the Java team is already using maven), mainly to get rid of NuGet as a dependency management system with which we're not at all happy, but also to have a common toolset (.NET tools are years behind the ones for Java). From my little experience with NPanDay I believe it's the cure we were looking for (we were interested in NPanDay for quite some time but because we're using .NET framework 4 VS 2010 previous versions were not worth evaluating). I found some problems (described below) in NPanDay.VisualStudio.Plugin (mainly in ReferenceManager) which I QD fixed. However, I'd like to actually do a more thorough code review make the best possible fix, and then propose it. I couldn't find what's the ceremony of submitting a patch, hence I'm asking on this list. Sorry if that's not the right place for that. The first bug is in Resync references which calls ReferenceManager.CopyArtifact: if (!artifact.FileInfo.Exists || artifact.Version.EndsWith(SNAPSHOT)) { if (!NPanday.ProjectImporter.Digest.Model.Reference.DownloadArtifact(artifact,l ogger)) { ReferenceErrorEventArgs e = new ReferenceErrorEventArgs(); e.Message = string.Format(Unable to get the artifact {0} from any of your repositories., artifact.ArtifactId); onError(e); return; } } The check for SNAPSHOT above forces Resync references to try and download the artifact off the remote repository. If the artifact is not yet installed (which is common if I'm developing a new module but don't want to commit/push it yet), what happens is that: 1. The resync fails (which is bad) and 2. If the artifact is installed in the local repo, it gets deleted (which is worse) What should happen (for SNAPSHOTs) instead is this: 1. Try to get timestamp of artifact in remote repo. If that fails, it shouldn't be a problem, the artifact might not be deployed (yet). 2. Try to get timestamp of artifact in local repo. If that fails, then this is an error and should be reported. 3. Get the timestamp of the local file (in .references/...) 4. If there's a remote timestamp and it's greater than the local one, update the local repo's artifact. 5. If the local repo's timestamp (possibly updated at point 4) is greater than the timestamp of the local file, update the file in .references/... For steps 1 and 2 use the timestamp in maven-metadata.xml (in ReferenceManager.copyToReferenceFolder there's already a TODO which suggests to use metadata/versioning/lastUpdated for the timestamp). Currently, the local repo's timestamp is considered to be the file's last write time (not even .LastWriteTimeUtc, which is another bug). To make it clear, since we're all developers, here's the algo in pseudo code (this might not compile, I'm writing it in Outlook): DateTime remoteRepoTimestamp, localRepoTimestamp, localFileTimestamp; bool hasRemoteTimestamp =TryGetArtifactRemoteRepositoryTimestamp(artifact, logger, out remoteRepoTimestamp); bool hasLocalRepoTimestamp =
Re: How can I contribute? (and a bug feature request)
I'll do my best to help NPanday kill NuGet ;) I think we could rather use the momentum behind the NuGet Gallery to speed up the adoption of NPanday... Maven encourages lazyness... who will ever download a jar manually? Everything, in every version, is on Maven Central. But with NPanday you still have to deploy every single .NET-Dependency yourself. If we could leverage Nuget Gallery (groupId 'nuget' :-), or what ever is before the first '.' in the nuget name) that would be a huge benefit: https://issues.apache.org/jira/browse/NPANDAY-372 I recently spoke to a guy (to remain nameless) at Microsoft (and works with Phil Haack) who said our company is one of the 5 in the world having a .NET Java division, Every company with more than 20 developers does Java, .NET and PHP; even Microsoft does Java and PHP. Sad to hear such words from the lonely isle of DevDiv. I simply don't buy that, hence I'm more than happy to contribute to NPanday. He didn't know about Apache Maven and NPanday as well, which as you said is sad. Me neither; and again welcome! The road to become a committer isn't hard either - after a few good patches, that drive the project in the correct direction, we'll set up you up for a vote :-) _ Lars Am 02.11.2011 um 09:42 schrieb Stoyan Damov: Thanks for the warm welcome! I'll do my best to help NPanday kill NuGet ;) I recently spoke to a guy (to remain nameless) at Microsoft (and works with Phil Haack) who said our company is one of the 5 in the world having a .NET Java division, and that the things I want from NuGet simply don't match the needs of the 80% of NuGet's target group. I simply don't buy that, hence I'm more than happy to contribute to NPanday. He didn't know about Apache Maven and NPanday as well, which as you said is sad. Cheers, Stoyan On 11/2/11 9:45 AM, Lars Corneliussen m...@lcorneliussen.de wrote: Hi Stoyan, welcome to the list! We have some Get Involved documentation here: http://incubator.apache.org/npanday/docs/1.4.0-incubating/developers/index .html Preamble: I'm a .NET-guy, as you easily can see on my blog (http://lcorneliussen.de/blog) I agree with you; .NET is way behind Java in terms of build tools. It was sad for me that the guys at Microsoft that built Nuget (and Guys that built the preceeding Nu) didn't even know about Apache Maven - and neither about NPanday. The .NET-Community struggles to learn from other communities. I joined the team in October 2010 because I think NPanday is absolutely the right horse to bet on. But still, there is loads of work to do. One more thing. No offense but the current code base (I have only looked at the .NET code so far) needs some cleaning up (TODOs, XML parsing, etc.). Let me know if you're interested in a code review and to who I should e-mail it, also I'll be more than glad to do most of the clean up myself if I'd be allowed to push/propose changes. I guess that about 50% of the Java code is simply dispensable. And the .NET code is quite badly organized. I think we rather need concrete improvement suggestions + patches - in small steps, than a code review :-) A year ago I started some work on the VS plugin, refactoring to command pattern, but I didn't manage to finish it: https://github.com/lcorneliussen/npanday/commit/5ed038f9bb38a957d40d14bba8 895ac9337e8aaa I also commented on the two issues you mentioned/submitted. _ Lars https://github.com/lcorneliussen/npanday/commit/5ed038f9bb38a957d40d14bba8 895ac9337e8aaa Am 01.11.2011 um 21:44 schrieb Stoyan Damov: Hi guys, First, sorry if the I'm not following the protocol for sending a single thing in one e-mail (e.g. this is a question, a bug report, and a feature request) but I couldn't find a How to send e-mails to this list link. Our company is evaluating NPanDay for use by our the .NET team (the Java team is already using maven), mainly to get rid of NuGet as a dependency management system with which we're not at all happy, but also to have a common toolset (.NET tools are years behind the ones for Java). From my little experience with NPanDay I believe it's the cure we were looking for (we were interested in NPanDay for quite some time but because we're using .NET framework 4 VS 2010 previous versions were not worth evaluating). I found some problems (described below) in NPanDay.VisualStudio.Plugin (mainly in ReferenceManager) which I QD fixed. However, I'd like to actually do a more thorough code review make the best possible fix, and then propose it. I couldn't find what's the ceremony of submitting a patch, hence I'm asking on this list. Sorry if that's not the right place for that. The first bug is in Resync references which calls ReferenceManager.CopyArtifact: if (!artifact.FileInfo.Exists || artifact.Version.EndsWith(SNAPSHOT)) { if (!NPanday.ProjectImporter.Digest.Model.Reference.DownloadArtifact(artifac t,l ogger
Re: How can I contribute? (and a bug feature request)
Correct list; no rules violated! :) Regarding the state of the .NET-Code: I could't agree more. And it is not much better on the Java side; but still the right horse to bet on!! I'll answer more in detail tomorrow! cheers, Lars _ Mobil versendet. Am 01.11.2011 um 21:44 schrieb Stoyan Damov stoyan.da...@pirinsoft.com: Hi guys, First, sorry if the I'm not following the protocol for sending a single thing in one e-mail (e.g. this is a question, a bug report, and a feature request) but I couldn't find a How to send e-mails to this list link. Our company is evaluating NPanDay for use by our the .NET team (the Java team is already using maven), mainly to get rid of NuGet as a dependency management system with which we're not at all happy, but also to have a common toolset (.NET tools are years behind the ones for Java). From my little experience with NPanDay I believe it's the cure we were looking for (we were interested in NPanDay for quite some time but because we're using .NET framework 4 VS 2010 previous versions were not worth evaluating). I found some problems (described below) in NPanDay.VisualStudio.Plugin (mainly in ReferenceManager) which I QD fixed. However, I'd like to actually do a more thorough code review make the best possible fix, and then propose it. I couldn't find what's the ceremony of submitting a patch, hence I'm asking on this list. Sorry if that's not the right place for that. The first bug is in Resync references which calls ReferenceManager.CopyArtifact: if (!artifact.FileInfo.Exists || artifact.Version.EndsWith(SNAPSHOT)) { if (!NPanday.ProjectImporter.Digest.Model.Reference.DownloadArtifact(artifact,l ogger)) { ReferenceErrorEventArgs e = new ReferenceErrorEventArgs(); e.Message = string.Format(Unable to get the artifact {0} from any of your repositories., artifact.ArtifactId); onError(e); return; } } The check for SNAPSHOT above forces Resync references to try and download the artifact off the remote repository. If the artifact is not yet installed (which is common if I'm developing a new module but don't want to commit/push it yet), what happens is that: 1. The resync fails (which is bad) and 2. If the artifact is installed in the local repo, it gets deleted (which is worse) What should happen (for SNAPSHOTs) instead is this: 1. Try to get timestamp of artifact in remote repo. If that fails, it shouldn't be a problem, the artifact might not be deployed (yet). 2. Try to get timestamp of artifact in local repo. If that fails, then this is an error and should be reported. 3. Get the timestamp of the local file (in .references/...) 4. If there's a remote timestamp and it's greater than the local one, update the local repo's artifact. 5. If the local repo's timestamp (possibly updated at point 4) is greater than the timestamp of the local file, update the file in .references/... For steps 1 and 2 use the timestamp in maven-metadata.xml (in ReferenceManager.copyToReferenceFolder there's already a TODO which suggests to use metadata/versioning/lastUpdated for the timestamp). Currently, the local repo's timestamp is considered to be the file's last write time (not even .LastWriteTimeUtc, which is another bug). To make it clear, since we're all developers, here's the algo in pseudo code (this might not compile, I'm writing it in Outlook): DateTime remoteRepoTimestamp, localRepoTimestamp, localFileTimestamp; bool hasRemoteTimestamp =TryGetArtifactRemoteRepositoryTimestamp(artifact, logger, out remoteRepoTimestamp); bool hasLocalRepoTimestamp = TryGetArtifactLocalRepositoryTimestamp(artifact, logger, out localRepoTimestamp); localFileTimestamp = GetLocalFileTimestamp(artifact.FileInfo, logger); if (hasRemoteTimestamp hasLocalRepoTimestamp remoteRepoTimestamp localRepoTimestamp) { UpdateLocalRepoArtifact(); // == NPanday.ProjectImporter.Digest.Model.Reference.DownloadArtifact(artifact, logger) // ^^ error handling above omitted for brevity localRepoTimestamp = remoteRepoTimestamp; } if (hasLocalRepoTimestamp localRepoTimestamp localFileTimestamp) { copyToReferenceFolder(artifact, referenceFolder); } Something like this. The 2nd thing is that there's a pretty good case to have Resync references from *local* repo and Resync references (from remote, which is described in the bug above). Resync references from local repo should skip the remote repo. Here's the valid case: I'm working on X and Y. I make a change in X and maven install it (in the local repo). I don't push the changes to the SCM. Y has a dependency on X and I want to resync the references so that I get the updated X dependency. Currently I can't do that. I can't (don't want to) push X's changes and wait on the build server to deploy the new snapshot because I don't want to hurt the rest of the developers with something which might still have bugs in it.
Re: .NetPortable support analysis
So far we determine the tools by vendor, vendorVersion, frameworkVersion (npanday-settings.xml) and then we have profile within that version (compiler-plugins.xml), right? This could be: Microsoft;3.0;2.0+FULL Mono;2.6;2.0+FULL I don't know .NETPortable, but: a) if it is a different toolset, and not really part of .NET framework releases, we should maybe rephrase Vendor by Distribution or so - and then add .NETPortable b) if it is a part of .NET core (versioned within), a profile .NETPortable,Profile2 might be enough. But I rather think .NETPortable is comparable to silverlight, which is separately released and still not supported by NPanday. From here, what about respecifying the attributes: vendor: means, what it means today vendorVersion: this one, too runtime: CLR, Mono, silverlight, Portable? runtimeVersion: instead of frameworkVersion runtimeProfile: Profile2, FULL, CLIENT, ... What do you guys think? Am 06.07.11 04:35, schrieb John Fallows: Devs, Now that PCL has landed in trunk, I'd like to discuss some alternatives in how we might prefer to expose the feature. plugin groupIdorg.apache.npanday.plugins/groupId artifactIdmaven-compile-plugin/artifactId extensionstrue/extensions configuration profilePORTABLE/profile frameworkVersionv4.0/frameworkVersion frameworkName.NETPortable,Version=v4.0,Profile=Profile2/frameworkName frameworkDisplayNamePortable Library/frameworkDisplayName /configuration /plugin It turns out that the obscurely named Profile2 in the example from NPANDAY-450 actually has system-defined meaning and is not a byproduct of internal VS2010 internal project-specific naming as previously thought. As you can see from the directory listings below, the various profiles have meaning; Profile1, Profile2, Profile3, Profile4 - in terms of which specific combinations of platforms are supported. This is because some of the DLLs used to resolve APIs during compilation are not actually present on all platforms. (!) *Directory of c:\Program Files\Reference Assemblies\Microsoft\Framework\.NETPortable\v4.0\Profile\Profile1\SupportedFrameworks *06/02/2011 09:18 PM 254 .NETFramework,Version=v4.0+,Profile=All.xml 06/02/2011 09:18 PM 271 Silverlight,Version=v4.0+,Profile=WindowsPhone.xml 06/02/2011 09:18 PM 252 Silverlight,Version=v4.0+.xml 06/02/2011 09:18 PM 199 Xbox,Version=v4.0+,Profile=All.xml *Directory of c:\Program Files\Reference Assemblies\Microsoft\Framework\.NETPortable\v4.0\Profile\Profile2\SupportedFrameworks *06/02/2011 09:18 PM 254 .NETFramework,Version=v4.0+,Profile=All.xml 06/02/2011 09:18 PM 271 Silverlight,Version=v4.0+,Profile=WindowsPhone.xml 06/02/2011 09:18 PM 252 Silverlight,Version=v4.0+.xml *Directory of c:\Program Files\Reference Assemblies\Microsoft\Framework\.NETPortable\v4.0\Profile\Profile3\SupportedFrameworks *06/02/2011 09:18 PM 254 .NETFramework,Version=v4.0+,Profile=All.xml 06/02/2011 09:18 PM 252 Silverlight,Version=v4.0+.xml *Directory of c:\Program Files\Reference Assemblies\Microsoft\Framework\.NETPortable\v4.0\Profile\Profile4\SupportedFrameworks *06/02/2011 09:18 PM 271 Silverlight,Version=v4.0+,Profile=WindowsPhone.xml 06/02/2011 09:18 PM 252 Silverlight,Version=v4.0+.xml Therefore, it would seem appropriate to use theprofile configuration in the compile plugin to select the corresponding combination of libraries so that unsupported APIs would not be undesirably exposed during compilation with NPanday. For example: plugin groupIdorg.apache.npanday.plugins/groupId artifactIdmaven-compile-plugin/artifactId extensionstrue/extensions configuration profile.NETPortable;Profile2/profile frameworkVersionv4.0/frameworkVersion /configuration /plugin Using the more specificprofile name will allow us to precisely control which libraries are needed during compile time. Taking this approach would move responsibility of the TargetFrameworkAttribute value to the compiler-plugins.xml file, to obtain the value of .NETPortable,Version=v4.0,Profile=Profile2 to be used when generating the assembly attributes, rather than taking it from the frameworkName mojo property as it is today. Further, we could make the target framework display name such as Portable Library defined by compiler-plugins.xml as well, eliminating the frameworkDisplayName from the compiler mojo. What do you think? tc, -john.
Codeplex update
Hi, I updated the codeplex page. Added the 1.4.0-incubating release including the urls to Apache Incubator. http://npanday.codeplex.com/releases/view/67075 I also changed the Home and Documentation pages, to make the move even more clear. http://npanday.codeplex.com/ http://npanday.codeplex.com/documentation I'm now in the process of merging the documentation. _ Lars
1.4.0-incubation Blog Post
For those that doesn't follow my blog: http://startbigthinksmall.wordpress.com/2011/05/26/apache-npanday-1-4-0-incubating-released-net-for-maven/
Re: [ANNOUNCEMENT] Apache NPanday 1.4.0-incubating Released!
This is great!!! ;) should we remove 1.3 from the download archive? http://archive.apache.org/dist/incubator/npanday/binaries/ -- Message sent from mobile device Am 17.05.2011 um 04:59 schrieb Adelita Padilla apadi...@maestrodev.com: Hi Everyone, The NPanday development team is pleased to announce the release of Apache NPanday 1.4.0-incubating under the Apache Incubator. NPanday is a project to integrated Apache Maven into .NET development environments. It includes both a Visual Studio add-in to integrated Maven, and a set of plugins for Maven that can build .NET projects uniformly from the command line. The latest release can be downloaded from http://incubator.apache.org/npanday/docs/1.4.0-incubating/download.html These are the major changes included in this release: * Visual Studio 2010 and .Net 4.0 Framework support * Removal of UAC and PAB directories * Updated groupIds of all NPanday components including plugins to org.apache.npanday For a complete list of all the features and fixes included in this release, please refer to http://incubator.apache.org/npanday/docs/1.4.0-incubating/release-notes.html If you have any questions, please consult: the website: http://incubator.apache.org/npanday/ the npanday-user mailing list: npanday-us...@incubator.apache.org This is the first release of NPanday here in Apache. Thanks to the NPanday and Apache Incubating community for helping grow the project. Enjoy! -- liit
Re: [VOTE] NPanday-1.4.0-incubating
+1 - VS Addin installs and launches (not tested more) - Repo contains everything needed (rest is downloaded from central). - Most ITs pass (its-trunk). - My sample projects pass :-) Well done. Integration Test Errors http://pastie.org/1873091 Tests in error: test(npanday.its.NPANDAY_377_WithCustomNPandaySettingsFileTest) test(npanday.its.NPANDAY_377_WithCustomNPandaySettingsDirectoryTest) - what could that be? both fail with 'The plugin 'org.apache.npanday.plugins:maven-compile-plugin' does not exist or no valid version could be found testConflictingExtensions(npanday.its.NPANDAY_140_ConflictingExtensionsTest) - this never passed on my machine testLadderBuild(npanday.its.NPANDAY_268_TransitiveDependenciesTest) testTransitiveDependenciesNotOnCompile(npanday.its.NPANDAY_268_TransitiveDependenciesTest) - seems to be a bug in the tests. both fail with: The plugin 'npanday.plugin:maven-compile-plugin' does not exist or no valid version could be found Am 06.05.11 11:44, schrieb Adelita Padilla: Hi NPanday Users, I'm happy to announce that NPanday-1.4.0-incubating is available and ready for your testing and voting. You can start testing here: Repository Builder: http://vmbuild.apache.org/archiva/repository/staged-npanday/org/apache/npanday/npanday-repository-builder/1.4.0-incubating/ MS Installer: http://vmbuild.apache.org/archiva/repository/staged-npanday/org/apache/npanday/npanday-installer/1.4.0-incubating/ Source: http://vmbuild.apache.org/archiva/repository/staged-npanday/org/apache/npanday/npanday-project/1.4.0-incubating/ The docs for 1.4.0-incubating can also be found here http://incubator.apache.org/npanday/docs/1.4.0-incubating/index.html [] +1 1.4.0-incubating passes and is ready for release [] 0 Don't care [] -1 Objection and please state the reason why. Thanks again for the hardworking people who contributed patches and filed issues. Cheers, -- liit
Re: Different solutions for sources and tests
IMHO, everything should be refactored in to one solution. Just go ahead. I don't think this is intentional. _ Lars Am 02.05.11 11:57, schrieb Deng Ching: Hi, Currently, the sources and the tests are in different solutions. This makes it harder to do refactoring -- if there are changes in the API, the tests don't get updated and you'd have to explicitly update them. Was there a reason why it was done like this and would there be any problems if they are merged into one solution? Thanks, Deng
NPanday Conventions
Hi alltogether, Just wan't to discuss some of the documented and undocumented conventions: In the documentation file there already is a comment to keep in mind, when reading: ~~ TODO: set a title for identifiers, since there are other conventions. Adjust to simplify - one group for whole project, ~~ plus subgroup for plugins and one for visual studio. Take a note that some don't follow convention now and are ~~ retained for legacy reasons, but will be refactored away in future 1) Conventions for Group-IDs and Artifact IDs: *OLD: ** If the module does not contain children modules, the Group ID is the same as the artifact ID. *OLD:* * If a module contains children modules, the child module Group ID should either be equivalent to a pluralized parent module Group ID or be a deriviative of the parent module Group ID. Imho this doesn't make sense. As stated in the comment above we should change that to something like: NEW: Use the groupId to group a certain set of artifacts. If possible, do not use the same identifier for both the group and the artifact - this often leads to confusions. The Group-ID usually is lower case. In .NET-projects, the artifact id becomes the assembly name. This makes UpperCase more suitable for .NET artifact ids. Also this doesn't make sense: *OLD: *The directory structure of the source directory (typically src/main/csharp) will follow the same pattern as the group ID. It should be changed to: *NEW: *The directory structure of the source directory (typically src/main/csharp) will follow the same pattern as the root namespace of a project. This usually corresponds to both the artifactId and the assembly name. 2) Add SVN conventions - Copy from and link to http://maven.apache.org/developers/conventions/svn.html 3) Add Source-code conventions - Copy from and link to http://maven.apache.org/developers/committer-environment.html Any other suggestions? _ Lars
Re: Is this a different transitive dependency issue from NPANDAY-412?
Any project that want's to resolve references (transitive or not) need the compile-plugin as extension in order to map from packaging to file extension... dotnet-library - dll, dotnet-executable - exe Workarround: Instead of specifying the packaging, you can also specify the file extension as type in dependencies. This should work without the compile-plugin, too. But it is not really a nice way. Sorry for hinting here again, but how can we get beyond NPANDAY-412? It is one of the last issues to be resolved for 1.3.1-incubating. Am 02.05.11 22:25, schrieb John Fallows: When a non-NPanday project (in an independent Maven execution) depends on a dotnet-executable, and tries to resolve transitive dependencies for that project, such as dotnet-library, the build breaks for the non-NPanday project because it fails to resolve the transitive dotnet-library dependency. The original dependency was not for the dotnet-executable itself, but for a separate classifier on that project, so the NPanday plugin was not necessary in the non-NPanday project to resolve the immediate dependency. However, in order to avoid the breakage when attempting to resolve the dotnet-library transitive dependency, we had to add in the NPanday plugin as an extension in the non-NPanday project. Is this an expected requirement? Is it specific to NPanday, or a general requirement for Maven? One could imagine there being enough information in the dotnet-executable pom to alleviate the need for specifying the NPanday extension plugin again in the non-NPanday project to resolve the transitive dotnet-library dependency. tc, -john.
Status for 1.3.1-incubating
Hi, we have 6 open issues, still. 123 needs attention. Who has a old machine to test it on? We could also release RC, and have it tested there. - 412 relates to this, too 377 is in progress. Should be resolved tomorrow. 383, who can rename the group ids? 403 successful generation of poms after clicking on Cancel button, *Unassigned*, reported by Liit Liit, what is to be done to resolve this issue? 412 Execuable path for Windows 7 SDK is not auto-detected when npanday-settings.xml is generated, resgen.exe not found John, I need input from you here. 398 MojoGenerator can't generate .NET-Mojos referencing a newer NPanday.Plugin.dll Joe, you wanted to take care of this. What is the status? _ Lars Hi, we have only six todos left: 383 groupid to org.apache.npanday, *unassigned*, reported by liit 377 hardcoded path in registry-config.xml, *liit* 387 Improve checking of mirror repos, *Unassigned*, reported by Joe 389 MojoGenerator can't generate .NET-Mojos referencing a newer NPanday.Plugin.dll, *Joe*, reported by Lars 403 successful generation of poms after clicking on Cancel button, *Unassigned*, reported by Liit 123 building without path, *Brett* (testing and fixes?) 1) When should we do 383? After code/feature freeze as part of release preparation? 2) Who will take care of 403? 3) Which dates should we target for code/feature freeze? I'd suggest Apr. 27th - one week from now. 4) Who will take care of the release preparations? What date should we target for the first RC? I'd suggest April 29th or May 2nd. _ Lars
Re: svn commit: r1098424 - in /incubator/npanday/trunk: components/dotnet-vendor/src/main/java/npanday/vendor/ components/dotnet-vendor/src/main/java/npanday/vendor/impl/ plugins/netplugins/NPanday.Pl
-- Message sent from mobile device Am 02.05.2011 um 03:39 schrieb Brett Porter br...@apache.org: On 02/05/2011, at 8:19 AM, lcornelius...@apache.org wrote: Modified: incubator/npanday/trunk/components/dotnet-vendor/src/main/java/npanday/vendor/SettingsUtil.java URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/components/dotnet-vendor/src/main/java/npanday/vendor/SettingsUtil.java?rev=1098424r1=1098423r2=1098424view=diff == --- incubator/npanday/trunk/components/dotnet-vendor/src/main/java/npanday/vendor/SettingsUtil.java (original) +++ incubator/npanday/trunk/components/dotnet-vendor/src/main/java/npanday/vendor/SettingsUtil.java Sun May 1 22:19:50 2011 @@ -35,7 +35,7 @@ import java.util.Hashtable; public class SettingsUtil { /** - * Return the registered settings, or create from default settings file location (.m2/npanday-settings.xml) + * Return the registered settings, or create from configured (-Dnpanday-settings=...) or default settings file location (.m2/npanday-settings.xml) * @param repositoryRegistry The registry. * @return The current, or just created SettingsRepository * @throws SettingsException If anything goes wrong reading or registering the settings @@ -43,7 +43,13 @@ public class SettingsUtil public static SettingsRepository getOrPopulateSettingsRepository( RepositoryRegistry repositoryRegistry) throws SettingsException { - return getOrPopulateSettingsRepository(repositoryRegistry, PathUtil.getHomeM2Folder()); +String settingsFolder = PathUtil.getHomeM2Folder(); +String customFolder = System.getProperty( npanday.settings ); +if (customFolder != null customFolder != ) This won't necessarily work, as string comparisons aren't native. You want !.equals(customFolder), or use something like !StringUtils.isEmpty for the whole line. i'll fix that. +{ +settingsFolder = customFolder; +} +return getOrPopulateSettingsRepository(repositoryRegistry, settingsFolder ); } /** Modified: incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/src/main/csharp/NPanday/Plugin/Settings/PathUtil.cs URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/src/main/csharp/NPanday/Plugin/Settings/PathUtil.cs?rev=1098424r1=1098423r2=1098424view=diff == --- incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/src/main/csharp/NPanday/Plugin/Settings/PathUtil.cs (original) +++ incubator/npanday/trunk/plugins/netplugins/NPanday.Plugin.Settings/src/main/csharp/NPanday/Plugin/Settings/PathUtil.cs Sun May 1 22:19:50 2011 @@ -17,12 +17,29 @@ // under the License. // +using System; using System.IO; namespace NPanday.Plugin.Settings { +// TODO: Move this to common PathUtil, when NPANDAY-422 is fixed public static class PathUtil { +public static string GetHomeM2Folder() +{ +return Environment.GetEnvironmentVariable(USERPROFILE) + /.m2; +} + +public static string BuildSettingsFilePath( string settingsPathOrFile ) +{ +if (settingsPathOrFile.EndsWith( xml )) +{ +return settingsPathOrFile; +} Same comment as before - this would be more robust if there was a way to check isDirectory or isFile instead. when it doesn't exist, we still need to determine if it is a file or a directory. This might also have been liits thoughts when reqiring a dir, instead of a file - because npanday both writes, reads and overwrites the file. - Brett -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Re: Welcome John Fallows as new Committer for NPanday
Herzlich Willkommen John! - I'd say in German :-) Happy to see npanday grow. Cheers, Lars Am 20.04.11 09:30, schrieb Josimpson Ocaba: Hi Everyone, I'm very happy to welcome John as a new committer for NPanday, he has been active in the community for some time now and has been regularly contributing patches, discussions, and ideas on how to grow NPanday. We hope to see more people become committers soon! Thanks again John for all your hard work! Cheers! Joe
Get the release out the door
Hi, we have only six todos left: 383 groupid to org.apache.npanday, *unassigned*, reported by liit 377 hardcoded path in registry-config.xml, *liit* 387 Improve checking of mirror repos, *Unassigned*, reported by Joe 389 MojoGenerator can't generate .NET-Mojos referencing a newer NPanday.Plugin.dll, *Joe*, reported by Lars 403 successful generation of poms after clicking on Cancel button, *Unassigned*, reported by Liit 123 building without path, *Brett* (testing and fixes?) 1) When should we do 383? After code/feature freeze as part of release preparation? 2) Who will take care of 403? 3) Which dates should we target for code/feature freeze? I'd suggest Apr. 27th - one week from now. 4) Who will take care of the release preparations? What date should we target for the first RC? I'd suggest April 29th or May 2nd. _ Lars
Re: Test Failures
That is a regression. I tried to catch Command line to long for cmd.exe. This test should only run on windows. Brett, do we support categories by os? -- Message sent from mobile device Am 18.04.2011 um 15:47 schrieb Matthias Wessendorf mat...@apache.org: Hi, in my previous email, I reported to JUnit test failures. One I corrected already ([1]) - The patch contains a bit of documentation. (as stated in the bug - I am surprised that on Linux the return-code is 1 as well for 'command not found') The second failure: testTooLongCommandName_withSpace(npanday.executable.CommandExecutorTest) generates something like: /bin/bash -c echo x. Not sure I get why you are expecting a test failure here. Thanks! Matthias [1] https://issues.apache.org/jira/browse/NPANDAY-407 -- Matthias Wessendorf blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf twitter: http://twitter.com/mwessendorf
Re: other old commits
Am 12.04.11 04:45, schrieb Brett Porter: Modified: incubator/npanday/trunk/dotnet/assemblies/NPanday.Plugin.MojoGenerator/src/main/csharp/NPanday/Plugin/MojoGenerator/Generator.cs URL:http://svn.apache.org/viewvc/incubator/npanday/trunk/dotnet/assemblies/NPanday.Plugin.MojoGenerator/src/main/csharp/NPanday/Plugin/MojoGenerator/Generator.cs?rev=1025511r1=1025510r2=1025511view=diff == --- incubator/npanday/trunk/dotnet/assemblies/NPanday.Plugin.MojoGenerator/src/main/csharp/NPanday/Plugin/MojoGenerator/Generator.cs (original) +++ incubator/npanday/trunk/dotnet/assemblies/NPanday.Plugin.MojoGenerator/src/main/csharp/NPanday/Plugin/MojoGenerator/Generator.cs Wed Oct 20 11:37:11 2010 @@ -100,11 +100,9 @@ namespace NPanday.Plugin.MojoGenerator jcuLocal.unmarshall(javaClass, fileInfo); } - ResourceManager resourceManager = new ResourceManager(NPanday.Plugin.MojoGenerator, - Assembly.GetExecutingAssembly()); - String pomXml = (String) resourceManager.GetObject(pom-java.xml); - TextReader reader = new StringReader(pomXml); - XmlSerializer serializer = new XmlSerializer(typeof(NPanday.Model.Pom.Model)); +TextReader reader = new StreamReader(Assembly.GetExecutingAssembly(). + GetManifestResourceStream(Assembly.GetExecutingAssembly().GetManifestResourceNames()[0])); + XmlSerializer serializer = new XmlSerializer(typeof(NPanday.Model.Pom.Model)); NPanday.Model.Pom.Model model = (NPanday.Model.Pom.Model) serializer.Deserialize(reader); model.artifactId = artifactId + .JavaBinding; model.groupId = groupId; I'm not sure I understand this change. Is it reading the POM as the first resource from an asembly? I also stumbled over that change and filed a bug. Didn't know it was introduced lately: JavaBindings-Generator for .NET-Mojos is fragile (and not working for NPanday.Plugin.SettingsGenerator) https://issues.apache.org/jira/browse/NPANDAY-385 https://issues.apache.org/jira/browse/NPANDAY-385
Re: NPanday 1.3.1-incubating Branched from a certain revison from Trunk
Hi Joe, I found out, that NPanday can still be built on Maven 2.2.1 (didn't try below). So this should be ok. We could just release what we have now as 1.4-incubating - it is certainly better than 1.3-incubating was. The PATH issue should be good. And documentation I can stop at any time :-) Liit should have a look at the last open issue, and then we go for a fast release. _ Lars Am 06.04.11 10:02, schrieb Josimpson Ocaba: Hi Everyone, With the many changes in our trunk recently I feel that this is too much for a point release and could fall into a 1.4-incubating release in itself. Having NPanday being built on Maven 3.x is also a major change for me. So what do you guys think if we branch out from a certain revision and release that as 1.3.1-incubating instead? Of course we will make sure that the revision will be stable. With this branch out we can also lessen the time for preparation for the release instead of stabilizing the issues that are in trunk. I'm very much biased on getting a 1.3.1-incubating ASAP :D What do the others think? Cheers, Joe
Documentation TOC/Navigation cleanup...
Hi, for me it is hard to even find the stuff I documented myself. This is why I made some changes to the styles (for better readability) and updated the Navigation. This is what it looks like here now. What do you think?
Re: Documentation TOC/Navigation cleanup...
On Thu, Apr 7, 2011 at 8:13 AM, Josimpson Ocaba joc...@maestrodev.com wrote: I have a question on ALM, can we replace it with something that a new user can easily understand? I'm sorry I also don't have anything in mind that could replace it but maybe we can brainstorm something. Or.. maybe the ALM section can just be removed then move Release and Version Management inside Developer's Guide and Artifact Repositories and Automated Builds under Get Involved? Don't think it belongs there. Both have very little to do with NPanday and would just get in the way when somebody wants to learn NPanday. Maybe renaming it to Application Lifecycle Management helps. Or Beyond Building. Maven Ecosystem Maven Ecosystem Beyond Building I think Application Lifecycle Management is still the best of these choices. Beyond building sounds good, too.
NPanday Repo down...
http://repo.npanday.org/archiva/repository/npanday-group/ seems to be down (again). Service Temporarily Unavailable The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later. Apache/2.2.3 (CentOS) Server at repo.npanday.org Port 80
Re: Finding a date for NPanday 1.3.1 Finalizing Meetup
See y'all in about 16 hours, then. 8:00 Friday for me. Second option on doodle. -- Message sent from mobile device Am 30.03.2011 um 09:28 schrieb Lars Corneliussen m...@lcorneliussen.de: Hi, I created a doodle. Please add your name and check where you have time available. You can adjust to your timezone in the upper right corner. NPanday 1.3.1 Finalizing Meetup http://www.doodle.com/r8f5x2pzvdhye3wf _ Lars
Re: Weird dependency resolving issues
ok. then we have to fix that for 1.3.1 :) -- Message sent from mobile device Am 01.04.2011 um 07:25 schrieb Josimpson Ocaba joc...@maestrodev.com: Yes, the actual dll were also deployed because when building npanday and trying to resolve the dependencies it is looking for a .dll from the remote repository. As what Adelita mentioned if those dependencies are not yet installed it what cause a failure :(
Re: Set up Continuous Integration for NPanday
Hi, we had a discussion on that a while ago: http://mail-archives.apache.org/mod_mbox/incubator-npanday-dev/201101.mbox/%3c4d353360.70...@lcorneliussen.de%3E I think brett planned to work on getting continuum back working. Last execution over there was March 14th and it failed. http://ci.npanday.org/continuum/projectGroupSummary.action?projectGroupId=16 Brett? PD: No te abrumes. Casi todos aqui hablan el ingles como lengua extranjera :-) - Lars Am 30.03.11 00:33, schrieb Javier Mandiola: Hi Lars, Which CI Build Server do you want to use? Hudson, CruiseControl or Continnium? I have experience with Hudson, but I have little knowledge of maven, I'm learning recently PD:sorry for my english but my native language is Spanish Regards On Tue, Mar 29, 2011 at 5:43 PM, Lars Corneliussenm...@lcorneliussen.dewrote: How is the state here? Would be great to have that setup ASAP - as I'm trying to aquire new developers :-) - Lars
Finding a date for NPanday 1.3.1 Finalizing Meetup
Hi, I created a doodle. Please add your name and check where you have time available. You can adjust to your timezone in the upper right corner. NPanday 1.3.1 Finalizing Meetup http://www.doodle.com/r8f5x2pzvdhye3wf _ Lars
Re: Set up Continuous Integration for NPanday
1) some still fail 2) will it build on commit? Am 30.03.11 15:29, schrieb Brett Porter: On 30/03/2011, at 5:54 PM, Lars Corneliussen wrote: Hi, we had a discussion on that a while ago: http://mail-archives.apache.org/mod_mbox/incubator-npanday-dev/201101.mbox/%3c4d353360.70...@lcorneliussen.de%3E I think brett planned to work on getting continuum back working. Last execution over there was March 14th and it failed. http://ci.npanday.org/continuum/projectGroupSummary.action?projectGroupId=16 It got tripped on the initial version change and some server misconfigurations. With moving to Apache's Jenkins instance on the cards, some of these things haven't been fixed. That should be ready now if someone wants to proceed - I think Joe filed the initial ticket and was working on it. In the mean time, I am building it again on ci.npanday.org now. - Brett -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Bootstrapping NPanday
Hi again, when you first build with the new settings, do the following: 1) Build trunk using mvn 2.x and NPanday 1.3-incubating 2) Delete npanday-settings.xml 3) Run any integration test against NPanday 1.3.1-incubating (joe should have switched the trunk version) in order to create a fresh one 4) Remove (almost) everything .NET-related from the PATH 5) Just keep Nunit-binaries on the PATH :-) 6) Switch to Maven 3 7) Change the root pom to stable.npanday.version${project.version}/stable.npanday.version !--stable.npanday.version1.2.1/stable.npanday.version-- 8) Build with mvn clean install As an alternative to 1) and 3) you can do the switch, and then run the first build with in a Visual Studio Prompt. Who could run the integration tests and find, if there are any more issues here? - Lars
Set up Continuous Integration for NPanday
How is the state here? Would be great to have that setup ASAP - as I'm trying to aquire new developers :-) - Lars
Building NPanday using newest NPanday and Maven 3
Hi, I really find it an obstacle to build NPanday trunk itself with an old NPanday version. It feels like stone age :) But sadly, due to a bug in MVN 2.2.1, it is not possible to use plugin-versions that just built within the same reactor. So i suggest to move on to Maven 3.0.x for building NPanday trunk. Suggestion: 1) Bootstrapping NPanday with current NPanday. (not possible on MVN 2.2.x) 2) Require Maven = 3 for *building* NPanday = 1.3.1 We should still make sure, that NPanday does work with Maven 2.2.x, though. Integration tests should run both in Maven 2.2.1 and Maven 3.0.2 or 3.0.3 *TODO:* * Add a Jira-Issue * change npanday.stable.version to npanday.bootstrap.version * preconfigure npanday.bootstrap.version with ${project.version} * update docs * remove all .NET-PATHes from the PATH environment * remove most of the PATH-related docs *Helpers on Windows for rapid switching of maven versions:* *mvn3.bat* (somewhere on the PATH) set M2=C:\Program Files\Apache\apache-maven-3.0.2\bin set M2_HOME=C:\Program Files\Apache\apache-maven-3.0.2 set PATH=C:\Program Files\Apache\apache-maven-3.0.2\bin;%PATH% mvn --version *mvn2.bat* (somewhere on the PATH) set M2=C:\Program Files\Apache\apache-maven-2.2.1\bin set M2_HOME=C:\Program Files\Apache\apache-maven-2.2.1 set PATH=C:\Program Files\Apache\apache-maven-2.2.1\bin;%PATH% mvn --version
Moving on to .NET 3.5 for NPanday C#-Sources
Hi again, since .NET 3.5 targets CLR 2.0, it would not change anything for our users when we move on to .NET 3.5 for NPanday trunk. All of us have 3.5 installed anyway, right? Is a vote appropriate for such questions? lg, Lars
Re: Building NPanday using newest NPanday and Maven 3
What do the others think? Vote necessary? Am 28.03.11 16:51, schrieb Brett Porter: Definitely helpful in trying to ship a self-contained release. On 28/03/2011, at 5:30 PM, Lars Corneliussen wrote: Hi, I really find it an obstacle to build NPanday trunk itself with an old NPanday version. It feels like stone age :) But sadly, due to a bug in MVN 2.2.1, it is not possible to use plugin-versions that just built within the same reactor. So i suggest to move on to Maven 3.0.x for building NPanday trunk. Suggestion: 1) Bootstrapping NPanday with current NPanday. (not possible on MVN 2.2.x) 2) Require Maven= 3 for *building* NPanday= 1.3.1 We should still make sure, that NPanday does work with Maven 2.2.x, though. Integration tests should run both in Maven 2.2.1 and Maven 3.0.2 or 3.0.3 *TODO:* * Add a Jira-Issue * change npanday.stable.version to npanday.bootstrap.version * preconfigure npanday.bootstrap.version with ${project.version} * update docs * remove all .NET-PATHes from the PATH environment * remove most of the PATH-related docs *Helpers on Windows for rapid switching of maven versions:* *mvn3.bat* (somewhere on the PATH) set M2=C:\Program Files\Apache\apache-maven-3.0.2\bin set M2_HOME=C:\Program Files\Apache\apache-maven-3.0.2 set PATH=C:\Program Files\Apache\apache-maven-3.0.2\bin;%PATH% mvn --version *mvn2.bat* (somewhere on the PATH) set M2=C:\Program Files\Apache\apache-maven-2.2.1\bin set M2_HOME=C:\Program Files\Apache\apache-maven-2.2.1 set PATH=C:\Program Files\Apache\apache-maven-2.2.1\bin;%PATH% mvn --version -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Re: Moving on to .NET 3.5 for NPanday C#-Sources
Am 28.03.11 17:01, schrieb Brett Porter: On 28/03/2011, at 5:33 PM, Lars Corneliussen wrote: Hi again, since .NET 3.5 targets CLR 2.0, it would not change anything for our users when we move on to .NET 3.5 for NPanday trunk. All of us have 3.5 installed anyway, right? yup, as long as testing that it runs inside VS 2005 still exists (or we decide explicitly not to support that any more). It should. But maybe users would have to install .NET Fx 3.5 - when we reference System.Core. Does somebody have a clean machine with VS2005 only on? Is a vote appropriate for such questions? Not really necessary unless something is very hard to gain consensus on. Otherwise, voting is best kept for releases committers where there is a binary decision to be made. - Brett -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Re: Weird dependency resolving issues
Am 28.03.11 17:12, schrieb Brett Porter: On 28/03/2011, at 6:24 PM, Lars Corneliussen wrote: Hi, I get lots of these: 28.03.2011 09:21:38 npanday.dao.impl.ProjectDaoImpl storeProjectAndResolveDependencies WARNUNG: NPANDAY-180-018: Not found in local repository, now retrieving artifact from wagon:npanday.model:NPanday.Model.Settings:library:2.0-SNAPSHOT, Failed Path Check = C:\Workbench\NPanday\svn-trunk\target\NPanday.Model.Settings.dll Any idea? The ProjectDaoImpl is incredibly fragile for these sorts of things. I don't think the RDF repository will be happy with developing multiple versions at a time - I usually flush it before switching (as you got 2.0 here). I'll try again after deleting the rdf-repos. Also lately the build slowed down, because maven is trying to resolve dependencies for Visual Studio (but build is not failing): This is in the repository builder? It seems like something recent affected it - not sure how it was before, but those POMs aren't in the remote repo now. e.g. http://repo.npanday.org/archiva/repository/npanday-group/VsWebSite/Interop/VsWebSite.Interop/ No. In the normal build. mvn install on npanday trunk root. Anyone know whether they were ever there? - Brett -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Re: [NPanday] Update of groupId
I hate when group ids change :) But org.apache.npanday is a better one :) So i'm all 4 it. Could we also eliminate the extra groupids for plugins/netplugins i.e. -- Message sent from mobile device Am 25.03.2011 um 05:18 schrieb Brett Porter br...@apache.org: On 25/03/2011, at 2:50 PM, Adelita Padilla wrote: Hi Everyone, In the last release (1.3-incubating), the deployed artifacts were not synched to central and it's because of the groupId (which was npanday). So before we proceed we the next release, we are proposing to change the groupId from npanday to org.apache.npanday. Any comments/suggestions/reactions? I think it's important to do, we just need to be aware of the implications. Plugin versions will no longer be mediated. Initially I wanted to wait until 2.0 (as perhaps we could trim the list of plugins needed anyway). However, I don't have any confidence in being able to get the existing group ID into central and I think the additional repositories are being a pain. However, I'll also note that the RDF artifacts aren't in central either, so there is still configuration needed there anyway. If others feel strongly that we shouldn't change for this release (and just not sync to central), then I'm ok with that. - Brett -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ http://au.linkedin.com/in/brettporter
Re: svn commit: r1084495 - /incubator/npanday/trunk/components/dotnet-executable/src/main/java/npanday/executable/impl/NetExecutableFactoryImpl.java
wrong place!! this must go into the settingsgenerator (written in .NET)!! i'll come back with more details -- Message sent from mobile device Am 23.03.2011 um 09:02 schrieb joc...@apache.org: Author: jocaba Date: Wed Mar 23 08:02:35 2011 New Revision: 1084495 URL: http://svn.apache.org/viewvc?rev=1084495view=rev Log: [NPANDAY-123] PATH should not have to be set for both Visual Studio AddIn or mvn command-line to function Added a default setting of the netHome to the proper SDKs Modified: incubator/npanday/trunk/components/dotnet-executable/src/main/java/npanday/executable/impl/NetExecutableFactoryImpl.java Modified: incubator/npanday/trunk/components/dotnet-executable/src/main/java/npanday/executable/impl/NetExecutableFactoryImpl.java URL: http://svn.apache.org/viewvc/incubator/npanday/trunk/components/dotnet-executable/src/main/java/npanday/executable/impl/NetExecutableFactoryImpl.java?rev=1084495r1=1084494r2=1084495view=diff == --- incubator/npanday/trunk/components/dotnet-executable/src/main/java/npanday/executable/impl/NetExecutableFactoryImpl.java (original) +++ incubator/npanday/trunk/components/dotnet-executable/src/main/java/npanday/executable/impl/NetExecutableFactoryImpl.java Wed Mar 23 08:02:35 2011 @@ -419,11 +419,18 @@ public class NetExecutableFactoryImpl executableRequirement.setProfile( profile ); ExecutableConfig executableConfig = ExecutableConfig.Factory.createDefaultExecutableConfig(); -executableConfig.setCommands( commands ); + +executableConfig.setCommands( commands ); ListString executablePaths = ( executableConfig.getExecutionPaths() == null ) ? new ArrayListString() : executableConfig.getExecutionPaths(); -if ( netHome != null ) + +if (netHome == null) +{ +netHome = getNetHome(); +} + +if ( netHome != null ) { logger.info( NPANDAY-066-014: Found executable path in pom: Path = + netHome.getAbsolutePath() ); executablePaths.add( netHome.getAbsolutePath() ); @@ -455,4 +462,34 @@ public class NetExecutableFactoryImpl throw new PlatformUnsupportedException( NPANDAY-066-001: Unable to find net executable, e ); } } + +private File getNetHome() +{ +String nHome = ; + +// 32 bit system +if( new File(C:/Program Files/Microsoft SDKs/Windows/v6.0A/bin/gacutil.exe).exists() new File(C:/Program Files/Microsoft SDKs/Windows/v6.0A/bin/xsd.exe).exists() ) +{ +nHome = C:/Program Files/Microsoft SDKs/Windows/v6.0A/bin; +} +if( new File(C:/Program Files/Microsoft SDKs/Windows/v7.0A/bin/gacutil.exe).exists() new File(C:/Program Files/Microsoft SDKs/Windows/v7.0A/bin/xsd.exe).exists() ) +{ +nHome = C:/Program Files/Microsoft SDKs/Windows/v7.0A/bin; +} + +// 64 bit system +if ( nHome != ) +{ +if( new File(C:/Program Files (x86)/Microsoft SDKs/Windows/v6.0A/bin/gacutil.exe).exists() new File(C:/Program Files (x86)/Microsoft SDKs/Windows/v6.0A/bin/xsd.exe).exists() ) +{ +nHome = C:/Program Files/Microsoft SDKs/Windows/v6.0A/bin; +} +if( new File(C:/Program Files (x86)/Microsoft SDKs/Windows/v7.0A/bin/gacutil.exe).exists() new File(C:/Program Files (x86)/Microsoft SDKs/Windows/v7.0A/bin/xsd.exe).exists() ) +{ +nHome = C:/Program Files (x86)/Microsoft SDKs/Windows/v7.0A/bin; +} +} + +return new File(nHome); +} }
Re: NPanday release without the Incubator PMC approval
Hi, and wow. I did not know they were that strict. How long will it take before 1.3-incubating is back out again? - Lars Am 14.03.11 06:20, schrieb Adelita Padilla: Hi Niall, I already removed the binaries and sources from the distribution in www.apache.org. I'll be sending out an email to general-incubator list for the Vote. Thanks, -- liit - Niall Pembertonniall.pember...@gmail.com wrote: This needs to be sorted out. Niall -- Forwarded message -- From: Niall Pembertonniall.pember...@gmail.com Date: Fri, Mar 11, 2011 at 11:58 PM Subject: Re: [ANNOUNCEMENT] Apache NPanday 1.3-incubating Released To: gene...@incubator.apache.org I didn't recall a release vote on this list and searching the NPanday list I see the following which confirms that: http://markmail.org/message/exvh6mjwqi6kpcjp Releases require the Incubator PMC approval - not just your mentor's votes and for that you should have held a vote on this list. This release is therefore not sanctioned by the ASF and should be removed until it is. Niall On Thu, Mar 10, 2011 at 1:22 PM, Josimpson Ocabajoc...@g2ix.net wrote: Hi Everyone, The NPanday development team is pleased to announce the release of Apache NPanday 1.3-incubating under the Apache Incubator! NPanday is a project to integrate Apache Maven into .NET development environments. It includes both a Visual Studio add-in to integrate Maven, and a set of plugins for Maven that can build .NET projects uniformly from the command line. The latest release can be downloaded from: http://incubator.apache.org/npanday/docs/1.3-incubating/download The major features for this release are: * Visual Studio 2010 and .Net 4.0 Framework support * Removal of UAC and PAB directories For a complete list of all the features and fixes included in this release, please refer to http://incubator.apache.org/npanday/docs/1.3-incubating/release-notes.html If you have any questions, please consult: the website: http://incubator.apache.org/npanday/ the npanday-user mailing list: npanday-us...@incubator.apache.org This is the first release of NPanday here in Apache. Thanks to the NPanday and Apache Incubating community for helping grow the project. Enjoy! -- Joe Ocaba
Re: NPanday documentation on Confluence?
Hi, looks good. Wysiwyg is what I want :-) Could we try it? Capacity to set it up? Funny architecture though. But why not using svn as a backend for CMS. _ Lars Am 03.03.11 17:17, schrieb Brett Porter: http://wiki.apache.org/general/ApacheCMSFAQ :) On 28/02/2011, at 10:02 PM, Lars Corneliussen wrote: both. :) -- Message sent from mobile device Am 28.02.2011 um 21:56 schrieb Brett Porterbr...@apache.org: The wiki isn't going to be supported for deployment of new websites, but there is a CMS that can be used we could investigate when it's ready. What's the part that slows you down at the moment - the user interface to editing, or getting it live quickly? On 27/02/2011, at 11:05 PM, Lars Corneliussen wrote: Hi, what about moving the docs out to Apache's Confluence? I think I had documented more, if it wouldn't be in wikifiles - and if it would be available faster. I also think it's more worth to have one up-to-date documentation, than to have it versioned along with npanday source. - Lars -- Brett Porter br...@apache.org http://brettporter.wordpress.com/ -- Brett Porter br...@apache.org http://brettporter.wordpress.com/
Re: [NPanday] - Proposal for NPanday-361
Should it be created if it doesn't exist? I suggest only automatic creation, when it is set to default location. Else it can be scaffolded using the generate-settings goal. What does maven do, if settings location is overridden? -- Message sent from mobile device Am 01.03.2011 um 18:09 schrieb Brett Porter br...@apache.org: On 28/02/2011, at 9:27 AM, Prins, drs. M.C. (Mark) wrote: Before you invent a property, does Maven already have one it knows about for the regular settings.xml file? There are two well known locations; one in the .m2 directory that is in the users %HOMEDRIVE% or %HOME% directory (or user profile directory), this depends on), the other is a system level in %M2_HOME%\conf The system level one is always there, the user level one doesn't have to be there, I think... When you run mvn w/ -X you can see it 1st tries to load the system file and then th euser file. Yep. We don't need the global location as npanday-settings doesn't currently merge those and it shouldn't write anything into the installation. You can lookup where -s pointed to, but I wouldn't rely on magic to write to the same directory. Since everything in NPanday originates from a mojo, I think the best thing to do is to always have a mojo property for expression ${npanday.settings}. That will honour both a property in a profile and on the command line, as well as being able to set it in the plugin configuration on a per-execution basis. If not set, it can default to ${user.home}/.m2/npanday-settings.xml. From the mojo, this value needs to be passed into method calls that require it (or lookup a central component and set the value). - Brett -- Brett Porter br...@apache.org http://brettporter.wordpress.com/