RE: Can synopsis (be made to) follow Python impo rts?
Stefan Seefeld <[email protected]>
| Newsgroups | gmane.comp.documentation.synopsis |
|---|---|
| Message-ID | <20DCDD8F0FCED411AC4D001083CF504502380D43@MTL-EXCHANGE> |
> From: Vicki Brown [mailto:[email protected]] > Subject: [Synopsis-devel] Can synopsis (be made to) follow Python > imports? > > > (or C++ #includes, or...) For C++ a file traversal using #include statements is definitely possible (though currently not implemented, please open a feature request on http://synopsis.fresco.org/issues/bug for that). Python is an entirely different story, as name lookup is done at runtime, and so a static analyzer can't predict what symbols will be found when the python processor actually runs the code. > I have (at my workplace) a set of Python source files that will > become a single executable process. The code is made up of many .py > files and assorted subdirectories. It is NOT a package. > > The "main" file imports 50 additional files which, in turn, > import others. > > What I want to see is the source, classes, functions, etc for the > entire executable program. > > Is there a way I can get this information from synopsis? you can process all the python code that is involved. But you won't be able to construct something like a callgraph from that, for example, as for that you'd have to execute the code (with a modified interpreter). Regards, Stefan