https://github.com/jponge/blobstore

A simple blob store in Java.
https://github.com/jponge/blobstore

Last synced: 5 months ago
JSON representation

A simple blob store in Java.

• Host: GitHub
• URL: https://github.com/jponge/blobstore
• Owner: jponge
• License: mit
• Created: 2012-03-19T17:21:47.000Z (almost 14 years ago)
• Default Branch: master
• Last Pushed: 2012-05-09T17:26:27.000Z (almost 14 years ago)
• Last Synced: 2024-10-20T22:24:44.902Z (over 1 year ago)
• Language: Java
• Size: 145 KB
• Stars: 5
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# BlobStore

A local, non thread-safe implementation of a key / value blobs storage in Java.

This is not mean to be bulletproof: this project purpose is to illustrate

the benefits of using [JBoss Byteman](http://www.jboss.org/byteman/) to

harness unit and functional test suites by injecting faults.

There are many cases where exceptional behaviour is hard to reproduce in tests,

especially when I/O and multiple threads are involved. JBoss Byteman manipulates

application bytecode through event/condition/action rules, and can make it very

easy to inject faults, wait/pause threads and so on.

## Discussion

The `master` branch contains the latest iterations over the code base and

tests suite.

The `without-byteman` branch contains the initial code base that was developed

with traditional unit tests, without fault injection.

By comparing both branches, you can see how introducing faults with Byteman

helped in:

* vastly increasing code coverage, and

* detecting a few bugs.

One may argue that the `without-byteman` branch could still be improved

without Byteman, and that the few bugs fixed afterwards in the `master`

branch could still have been spotted through peer-review.

This is true.

The point is that the `without-byteman` branch still had a reasonable

code coverage to begin with, with the missing spots lying in I/O errors

that would be very hard to introduce through even with object stubs/mocks.

Forcing a filesystem to refuse renaming a file, or forcing it to become

full all of a sudden is definitely non-trivial.

If you have experience with other techniques than leveraging a bytecode

manipulation tool like Byteman then I am very interested in hearing from

you!

## License

Copyright (c) 2012 [Julien Ponge](http://julien.ponge.info/).

    Permission is hereby granted, free of charge, to any person obtaining a copy

    of this software and associated documentation files (the "Software"), to deal

    in the Software without restriction, including without limitation the rights

    to use, copy, modify, merge, publish, distribute, sublicense, and/or sell

    copies of the Software, and to permit persons to whom the Software is

    furnished to do so, subject to the following conditions:

    The above copyright notice and this permission notice shall be included in

    all copies or substantial portions of the Software.

    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR

    IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,

    FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL THE

    AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER

    LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,

    OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN

    THE SOFTWARE.

## Building

The JBoss repositories can be boring to configure right with Maven, especially

if you want to avoid putting explicit repository references in your POMs, which

is bad practice anyway.

[Read on this JBoss documentation](https://community.jboss.org/wiki/MavenGettingStarted-Users).

## You can contribute!

Although this project is a use-case rather than a real attempt at making a rock-solid

blob storage engine, feel-free to propose enhancements.