- Project tools
-
-
- How do I...
-
| Category |
Featured projects |
| scm |
Subversion,
Subclipse,
TortoiseSVN,
RapidSVN
|
| issuetrack |
Scarab |
| requirements |
xmlbasedsrs |
| design |
ArgoUML |
| techcomm |
SubEtha,
eyebrowse,
midgard,
cowiki |
| construction |
antelope,
scons,
frameworx,
build-interceptor,
propel,
phing
|
| testing |
maxq,
aut
|
| deployment |
current |
| process |
ReadySET |
| libraries |
GEF,
Axion,
Style,
SSTree
|
| Over 500 more tools... |
|
Ephedra - Frequently Asked Questions
- Q: How is Ephedra different from C2J++?
A:
C2J++ converts nice C++ classes. It is of little help
as soon as pointers are involved. It does not provide a standard
library replacement.
- Q: How is Ephedra different from C2J?
A:
C2J generates Java code that is hard to read and difficult to
integrate with mainstream Java code. It circumvents Java's type
checking and storage management.
- Q: Where are the sources?
A:
Ephedra is now part of the open source
Ovid project. The sources can be downloaded there. Right now, they
only compile with IBM VisualAge C++ V5.0 for AIX only (or IBM VisualAge C++
V4.0 on NT and IBM's confidential CodeStore API).
- Q: Can I help?
A:
Yes:
- You are welcome to try the tool and report problems. I am
particularly interested in some performance data: How does
the generated Java code compare to the original C/C++ code
and possibly hand-optimized Java code?
- You are welcome to implement parts of the C standard library.
Take a look at the existing Ephedra library for examples.
- Q: Where can I find more information on Ephedra?
A:
There are several ways:
- Read the papers mentioned above on the
main Ephedra page.
- Subscribe to the
Ovid mailing lists.
- Email me (jmartin _AT_ tigris _DOT_ org). Please do so only after
making sure your question hasn't been answered on one of the mailing
lists, and only if you think it is too stupid or too confidential to
be posted on the mailing list (such as: I want to give you $1 million,
where shall I send the cheque).
- Q: I would like to port Ephedra to [insert your favourite OS].
A:
So would I (probably). The problem is that Ephedra depends on
IBM VisualAge C++ V5.0. Rather than writing my own buggy C++ parser,
I chose to use VisualAge's C++ parser. That saved me a couple of
years of coding and debugging. If you want to port Ephedra to platform
xy, you will also have to port VisualAge, which should be
rather difficult (especially since you won't get the sources).
- Q: So why don't you use GNU GCC to parse C++ instead?
A:
This is one of the things that will be done in the
Ovid project.
- Q: Is it possible to implement custom optimisations and transformations
in Ephedra?
A:
Yes. At the moment, you will need the Ephedra sources to do that. Ovid
should make this easier.
- Q: Why do your email addresses look so strange?
A:
I hate junk mail. Replace the _AT_ by @ and the _DOT_ by . if you really
want to email me.
|