A file browser and editor for the Infinity Engine. You can find out more in the Near Infinity Wiki.
Required tools:
If you wish to compile Near Infinity from source, the easiest way to do so is
by using the included Ant build script. The default target builds the NearInfinity.jar
.
cd NearInfinity
ant
The sources also contain configuration files for the Eclipse IDE.
It is preferred to use Apache Ant
for compiling Near Infinity, however.
NI's BCS compiler uses code that is generated by the parser generator JavaCC. The file src/org/infinity/resource/bcs/parser/BafParser.jjt
is used to build the parse tree.
Use the following commands if you need to recreate it:
cd NearInfinity
ant parser-clean
ant parser-generate
The JavaCC executable is already included in the Near Infinity sources.
The unit tests (that currently have far from complete coverage) can be run by
passing test
as the target parameter to ant:
cd NearInfinity
ant test
(This will compile the tests and run the tests on the last compiled build)
This section contains information intended for those who contribute code to Near Infinity (NI). It contains some information on style and explains how the code is organised and added to.
Code style is intended to improve code readability and to reduce "diff noise" (meaningless changes).
Simply put, match the existing style. In particular:
-
Use an indentation level of 2 spaces, not tab stops.
-
Pay attention to how the existing code is indented and try to adhere to the same pattern.
-
Do not leave trailing white space. Most decent editors have tools to help with this.
-
End files with a single newline. Again, decent editors...
-
No wildcard imports. While wildcarded import declarations are easy to use, they are also a source of error if you want to use classes of the same name from different packages. Modern editors or IDEs should take care of it automatically.
Additionally, try to avoid overly long lines. Breaking lines at column 80 or 100 is very standard, but since this is ReallyRidiculouslyLongNames Java it's not always practical. Do try to limit yourself to less than column 120, however, or it becomes difficult to read the code on GitHub. The existing code is not always well-behaved in this regard, however.
When in doubt, refer to the official Java conventions (outdated) or Google's Java Style.
The project consists of two active branches:
-
master
- code that reflects the latest stable release. It is only updated by a change of the version number which happens very rarely. -
devel
- code that is ready to go into the next work-in-progress (WIP) release with little to no adjustment. That is, it should be complete and behave well locally, even if it has not been exhaustively tested under a wide variety of conditions.
If you work on something you should generally do so on a feature branch based off devel. Once the feature is complete, tested and preferably reviewed, it can be merged back into devel. While you are working on your branch, be sure to keep it up to date with devel, in order to avoid large divergences (and the resulting messy merges). Small, straightforward commits can be made directly to devel. Working off a feature branch also has the advantage of letting others easily check out your work (provided you push it, naturally), since they are able to pull your changes and simply check out your branch (compared to needing to merge your devel branch into their own, or clone a new local repository for your code).
The branch gh-pages
is primarily used for hosting the update definition
file which is used to notify users when a new Near Infinity version has been
released.
Other branches, if any, are either used for testing purposes or are not actively maintained anymore.
To maximise compatibility with available systems, NI's code base is using the
feature set of a specific Java version. You can find the currently supported
version by inspecting the variable JAVA_VERSION
in src/org/infinity/NearInfinity.java
.