When you generate a IBM® Rational® Rhapsody® model
from legacy code in the reverse engineering process, further edits
to the model or to the code become synchronized in the roundtripping
process thereafter. Code centric mode is the default for the reverse
engineering process.
Macro collecting, the first stage of the reverse engineering process,
collects all the macros from the files before analyzing the files
and building the model.
The results of reverse engineering are as follows:
- Recognized and supported constructs are added to
the model.
- Existing features in a model are updated from the
source file to match the source file definition. For example, if the
type of an attribute differs in an existing model and a source file
being imported, it is changed in the model to match the source file.
- With the ability to preserve code structure ("code
respect"), the reverse engineered code in the Rational Rhapsody model
respects the structure of the original code and preserves this structure
when code is regenerated from the Rational Rhapsody model.
The reverse engineered C and C++ code preserves the order, location,
and dependencies of the global elements in the original code. For
more information, see Preserving code structure during generation.
- Unresolved elements that are not resolved by the
import process remain unresolved.
- Existing diagrams or statecharts are not synthesized
using imported elements.
- New model elements found in a source file are added
to the browser, but not to existing diagrams.
Use the roundtrip feature to update an existing model. See Roundtripping in code-centric mode for more information.