• Stars
    star
    10
  • Rank 1,807,489 (Top 36 %)
  • Language
    Scala
  • License
    MIT License
  • Created about 13 years ago
  • Updated almost 5 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Scala classpath scanner

Sclasner is a classpath scanner written in Scala.

It is intended as a replacement of Annovention and mainly used for standalone JVM applications. If you want a more complex solution, please see Reflections.

With Sclasner, you can:

  • Scan all .class files (including those inside .jar files in classpath), then use Javassist or ASM to extract annotations
  • Load all .po files
  • etc.

Scan

See Scaladoc.

For example, if you want to load all .txt files:

import java.io.File
import sclasner.{FileEntry, Scanner}

// We define a callback to process each FileEntry:
// - The 1st argument is an accumulator to gather process results for each entry.
// - The 2nd argument is each entry.
// - The result of this callback will be passed to as the accumulator (the
//   1st argument) to the next call.
// - When all entries have been visited, the accumulator will be returned.
def entryProcessor(acc: Seq[(String, String)], entry: FileEntry): Seq[(String, String)] = {
  if (entry.relPath.endsWith(".txt")) {
    val fileName = entry.relPath.split(File.pathSeparator).last
    val body     = new String(entry.bytes)
    acc :+ (fileName, body)
  } else {
    acc
  }
}

// We actually do the scan:
// - The 1st argument is the initial value of the accumulator.
// - The 2nd argument is the callback above.
val acc = Scanner.foldLeft(Seq.empty, entryProcessor)

Things in FileEntry:

  • container: File, may be a directory or a JAR file in classpath. You may call container.isDirectory or container.isFile. Inside each container, there may be multiple items, represented by the two below.
  • relPath: String, path to the file you want to check, relative to the container above.
  • bytes: Array[Byte], body of the file the relPath above points to. This is a lazy val, accessing the first time will actually read the file from disk. Because reading from disk is slow, you should avoid accessing bytes if you don't have to.

Signature of Scanner.foldLeft:

foldLeft[T](acc: T, entryProcessor: (T, FileEntry) => T): T

Cache

One scan may take 10-15 seconds, depending things in your classpath and your computer spec etc. Fortunately, because things in classpath do not change frequently, you may cache the result to a file and load it later.

You provide the cache file name to foldLeft:

// You can use File instead of file name
val acc = Scanner.foldLeft("sclasner.cache", Seq.empty, entryProcessor)

If sclasner.cache exists, entryProcessor will not be run. Otherwise, entryProcessor will be run and the result will be serialized to the file. If you want to force entryProcessor to run, just delete the cache file.

If the cache file cannot be successfully deserialized (for example, serialized classes are older than the current version of the classes), it will be automatically deleted and updated (entryProcessor will be run).

For the result of entryProcessor to be written to file, it must be serializable.

Cache in development mode

Suppose you are using SBT, Maven, or Gradle.

While developing, you normally do not want to cache the result of processing the directory target (SBT, Maven) or build (Gradle) in the current working directory.

Sclasner's behavior:

  • If container is a subdirectory of target or build, the result of processing that container will not be cached.
  • When loading the cache file, if a container is a subdirectory of target or build, entryProcessor will be run for that container.

Use with SBT

Supported Scala versions: 2.13, 2.12

libraryDependencies += "tv.cntt" %% "sclasner" % "1.8.0"

Sclasner is used in Xitrum.

More Repositories

1

xitrum

Async and clustered Scala web framework and HTTP(S) server
Scala
446
star
2

scalive

Connect a Scala REPL to running JVM processes without any prior setup
Java
196
star
3

glokka

Library to register and lookup actors by names in an Akka cluster
Scala
56
star
4

scaposer

GNU Gettext .po file loader for Scala
Scala
38
star
5

scala-xgettext

Scala compiler plugin that acts like GNU xgettext command to extract i18n strings in Scala source code files to Gettext .po file
Scala
25
star
6

xitrum-new

Empty Xitrum project skeleton, like the one created by "rails new"
Scala
21
star
7

xitrum-demos

Demos for Xitrum
Scala
19
star
8

xitrum-package

SBT plugin for collecting dependency .jar files for standalone Scala programs
Scala
18
star
9

comy

Simple URL shortener using Xitrum and MongoDB
Scala
16
star
10

xitrum-doc

Xitrum Guide Book and presentations about Xitrum
CSS
8
star
11

RhinoCoffeeScript

coffee-script.js precompiled to JVM .class file by Rhino
Java
7
star
12

agent7

Java agent to reload .class files; it uses file watch API available in Java 7+
Java
5
star
13

xitrum-scalate

Template engine for Xitrum based on Scalate
Scala
4
star
14

xitrum-hazelcast

Cache and server side session store for Xitrum, based on Hazelcast
Scala
3
star
15

xitrum-multimodule-demo

http://groups.google.com/group/xitrum-framework/browse_thread/thread/7588995934854a56
Scala
2
star
16

xitrum-ko

Knockout.js plugin for Xitrum
Scala
1
star
17

xitrum-scalatags

Template engine for Xitrum based on ScalaTags
Scala
1
star
18

xitrum-framework.github.io

Source code of the home page of Xitrum framework
HTML
1
star
19

xitrum-tictactoe

Tictactoe web game - Demo for Akka and Xitrum
Scala
1
star