| 1 |
On Tue, Jan 25, 2011 at 5:15 PM, Michael Orlitzky <michael@××××××××.com> wrote: |
| 2 |
> On 01/24/2011 05:34 PM, Mark Knecht wrote: |
| 3 |
>> Hello, |
| 4 |
>> I'm wondering if there are any generic sorts of code translation |
| 5 |
>> tools in portage wherein I could translate from an 'uncommon' language |
| 6 |
>> no one here is likely to use (EasyLanguage) into C? |
| 7 |
>> |
| 8 |
>> As an example I've attached a little EL function that takes |
| 9 |
>> buy/sell command data an puts it away in an array for safe keeping. |
| 10 |
>> What tools are out there, if any, that might allow me to describe how |
| 11 |
>> EL works and then the tool does the conversion? |
| 12 |
> |
| 13 |
> Since no one else has given you the bad news, this is basically |
| 14 |
> impossible if you care that the two programs behave the same. |
| 15 |
> |
| 16 |
> For any particular program, the best you can do is rewrite it by hand |
| 17 |
> after creating a battery of unit tests. The alternative is to compile |
| 18 |
> your source language to a common low-level language, and then decompile |
| 19 |
> back to your target language. |
| 20 |
> |
| 21 |
> Unfortunately, unless your common low-level language is some sort of |
| 22 |
> bytecode with additional metadata (you can translate between .NET |
| 23 |
> languages for example), the output from the decompiler is going to look |
| 24 |
> like garbage. |
| 25 |
|
| 26 |
Michael, |
| 27 |
Thanks for the inputs. It gives me more to think about. |
| 28 |
|
| 29 |
In this case the input language is interpreted, not compiled. The |
| 30 |
trading platform interprets the program and internally turns it into |
| 31 |
buy & sell operations. (Not the piece of code I supplied - that was |
| 32 |
just a small function.) Unfortunately, as the language is proprietary |
| 33 |
to the trading platform there isn't a way to go to any common |
| 34 |
low-level language. |
| 35 |
|
| 36 |
The 'battery of tests' would be, I think, the trading program being |
| 37 |
executed on a certain market, producing a certainly list of buy & sell |
| 38 |
operations and a specific gain or loss. It would be quite easy to |
| 39 |
compare the outcome because it's nothing more than a list of trades. |
| 40 |
If the translated code generates the same list then it works. If not, |
| 41 |
I dig into why. This part of the task seems relatively straight |
| 42 |
forward to me. |
| 43 |
|
| 44 |
I was mainly hoping to find a tool that might generate _reasonable_ |
| 45 |
C code, even if it's not perfect. If the C code compiles and runs then |
| 46 |
I could determine what works, what doesn't, and start fixing things. |
| 47 |
I'm not a C programmer and haven't touched that language in at least |
| 48 |
15 years so anything that moves me forward would be helpful. |
| 49 |
|
| 50 |
Again, I do appreciate your inputs. If this extra info gives you |
| 51 |
any new ideas please let me know. |
| 52 |
|
| 53 |
Cheers, |
| 54 |
Mark |
Archives