Add NuGet build config files#2
Conversation
| my_bin: { | ||
| #destination:${d_bin}/.; | ||
| #add-each-file : ::nuget.[default]targets.[${condition}].CopyToOutput; | ||
| "build/release/mlpack.dll" |
There was a problem hiding this comment.
Shouldn't this be mlpack.lib? Also, it looks like destination and add-each-file are unused, so maybe we can remove the parameters?
There was a problem hiding this comment.
CoApp handles *.lib automatically but for bin it creates extra package called mlpack-redist. So, I had to use this simple hack trick. destination specifies content should be in bin path and add-each-file specifies bin file should be copied to output folder during build time.
| } | ||
| files{ | ||
| nestedInclude:{ | ||
| #destination = ${d_include}mlpack; |
There was a problem hiding this comment.
If we remove destination then include folder in NuGet package doesn't get proper structure!
rcurtin
left a comment
rcurtin
left a comment
There was a problem hiding this comment.
Looks like there are a few more issues to address but I think this is a good start. Can you tell me how you tested this?
If I am understanding correctly, this autopkg file is only for packaging, not for compilation. So the compilation of mlpack on Windows is a separate step. Is that correct?
| description: "mlpack is an intuitive, fast, scalable C++ machine learning library, meant to be a machine learning analog to LAPACK. It aims to implement a wide array of machine learning methods and functions as a "swiss army knife" for machine learning researchers."; | ||
| summary: "Bugs Fixed"; | ||
| releaseNotes: ""; | ||
| authors: {Ryan Curtin}; |
There was a problem hiding this comment.
I'm definitely not the only contributor, maybe we should wrote see http://www.mlpack.org/about.html here or something.
There was a problem hiding this comment.
But you are the owner 😃 and the list is too long 😆.
| projectUrl: "https://github.com/mlpack/mlpack"; | ||
| title: "mlpack: a scalable C++ machine learning library"; | ||
| description: "mlpack is an intuitive, fast, scalable C++ machine learning library, meant to be a machine learning analog to LAPACK. It aims to implement a wide array of machine learning methods and functions as a "swiss army knife" for machine learning researchers."; | ||
| summary: "Bugs Fixed"; |
There was a problem hiding this comment.
Not sure what this means, maybe just put 'mlpack is a scalable C++ machine learning library' here and just use 'mlpack' as the title?
|
Definitely, it will only be used to pack NuGet package after AppVeyor build is complete!
and Done! |
Also add more search tags
|
Ok, sounds good. Can you fix the summary and description? Once this is merged, what will the next step be? We need to hook up AppVeyor to build the tags of mlpack, then push that to build the NuGet package. Do you have a plan for how that can be done? We don't have any Windows build slaves currently, so if we can use AppVeyor for this it would be best. |
3 participants
mlpack.autopkg is config file for NuGet build. During build, we can wget this file and use it to build and publish NuGet for mlpack.
Please, make any changes required in given file.