github.com/phobos182/packer@v0.2.3-0.20130819023704-c84d2aeffc68/website/source/docs/extend/post-processor.html.markdown (about) 1 --- 2 layout: "docs" 3 --- 4 5 # Custom Post-Processor Development 6 7 Post-processors are the components of Packer that transform one artifact 8 into another, for example by compressing files, or uploading them. 9 10 In the compression example, the transformation would be taking an artifact 11 with a set of files, compressing those files, and returning a new 12 artifact with only a single file (the compressed archive). For the 13 upload example, the transformation would be taking an artifact with 14 some set of files, uploading those files, and returning an artifact 15 with a single ID: the URL of the upload. 16 17 Prior to reading this page, it is assumed you have read the page on 18 [plugin development basics](/docs/extend/developing-plugins.html). 19 20 Post-processor plugins implement the `packer.PostProcessor` interface and 21 are served using the `plugin.ServePostProcessor` function. 22 23 <div class="alert alert-block"> 24 <strong>Warning!</strong> This is an advanced topic. If you're new to Packer, 25 we recommend getting a bit more comfortable before you dive into writing 26 plugins. 27 </div> 28 29 30 ## The Interface 31 32 The interface that must be implemented for a post-processor is the 33 `packer.PostProcessor` interface. It is reproduced below for easy reference. 34 The actual interface in the source code contains some basic documentation as well explaining 35 what each method should do. 36 37 38 <pre class="prettyprint"> 39 type PostProcessor interface { 40 Configure(interface{}) error 41 PostProcess(Ui, Artifact) (a Artifact, keep bool, err error) 42 } 43 </pre> 44 45 ### The "Configure" Method 46 47 The `Configure` method for each post-processor is called early in the 48 build process to configure the post-processor. The configuration is passed 49 in as a raw `interface{]`. The configure method is responsible for translating 50 this configuration into an internal structure, validating it, and returning 51 any errors. 52 53 For decoding the `interface{]` into a meaningful structure, the 54 [mapstructure](https://github.com/mitchellh/mapstructure) library is 55 recommended. Mapstructure will take an `interface{}` and decode it into an 56 arbitrarily complex struct. If there are any errors, it generates very 57 human-friendly errors that can be returned directly from the configure 58 method. 59 60 While it is not actively enforced, **no side effects** should occur from 61 running the `Configure` method. Specifically, don't create files, don't 62 create network connections, etc. Configure's purpose is solely to setup 63 internal state and validate the configuration as much as possible. 64 65 `Configure` being run is not an indication that `PostProcess` will ever 66 run. For example, `packer validate` will run `Configure` to verify the 67 configuration validates, but will never actually run the build. 68 69 ### The "PostProcess" Method 70 71 The `PostProcess` method is where the real work goes. PostProcess is 72 responsible for taking one `packer.Artifact` implementation, and transforming 73 it into another. 74 75 When we say "transform," we don't mean actually modifying the existing 76 `packer.Artifact` value itself. We mean taking the contents of the artifact 77 and creating a new artifact from that. For example, if we were creating 78 a "compress" post-processor that is responsible for compressing files, 79 the transformation would be taking the `Files()` from the original artifact, 80 compressing them, and creating a new artifact with a single file: the 81 compressed archive. 82 83 The result signature of this method is `(Artifact, bool, error)`. Each 84 return value is explained below: 85 86 * `Artifact` - The newly created artifact if no errors occurred. 87 * `bool` - If true, the input artifact will forcefully be kept. By default, 88 Packer typically deletes all input artifacts, since the user doesn't generally 89 want intermediary artifacts. However, some post-processors depend on the 90 previous artifact existing. If this is `true`, it forces packer to keep the 91 artifact around. 92 * `error` - Non-nil if there was an error in any way. If this is the case, 93 the other two return values are ignored.