This turned out to be so useful for code reviews (we do a lot of those) that I "ported" it to a web app that lets us set hilocks that follow us from buffer to buffer (along with xrefs, class/method/fn definition sites, bookmarks, and a lot of other junk).
No (reverse) call trees, no high-level views, no help on how data propagates through the program. We could do so much better, especially in static languages. Sometimes it seems like reverse engineering tools that work without source code (such as IDA) do a better job of cross-referencing than tools that have the actual source code.
Is there any good/usable point and click code browser / code comprehension tool these days? Or is the state of the art of 1998 still that of today?
Also, I really wish the myth that "point and click" is inevitably superior would go away. It may be easier to learn, but my experience is that with some practice keyboard-based programming tools are almost invariably faster.
My point is that much more intelligent, useful and efficient queries could be possible by making use of the fact that code is a data structure, and not just text. Various things could be inferred automatically, which would be especially handy in legacy C source code where it's really hard to make head or tail of (I see a lot of that in my job).
It should be able to answer questions like "where does this value come from", "how is it computed" and show a tree/chain of statements. Or "where is this structure accessed" ... and so on.
As clang makes these things easier, I hope there will be a new resurgence in intelligent code comprehension tools. Those don't necessarily have to be point and click, but could have different UIs...
The Roslyn CTP for Visual Studio is an API for the C#/VB compilers. Visual Studio itself uses this for e.g. refactorings. You can write VS plugins that examine the abstract syntax trees of the files in the project, instead of working with the flat text.
The tools therefore need a fully working build environment, and the ability to temporarily override aspects of that build environment.
DXR is a source code search engine. It appears to be cscope with a database. That's not nothing, but it's not exactly taking us out of the opaque blobs of text and into semantics.
DXR is currently a pain to get up and running. There's been talk on the mailing list on streamlining the install process but it would still require a lot intervention depending the compilation process. https://groups.google.com/forum/#!topic/mozilla.dev.static-a...
That's a lot to ask for code that the editor may not know the compile and link options for, which may not even compile using the right options, or which may only be understood by a compiler that isn't installed on the system where you're browsing the code.
Basically what you want is an IDE with a project set up telling it exactly how to compile the code or what the runtime setup will be for a dynamic language. Then you'll have decent code navigation, if the IDE is integrated with a compiler that can handle the code, and if there aren't any bugs! I can tell you that for Java the problem is thoroughly solved. The language is simple, and the IDEs are mature. For C++ you're SOL. Scala is getting there with Eclipse and IntelliJ. I don't know about other languages.
I for one enjoy knowing how to handle unstructured text, because you can never count on having tools that will understand everything you come across. XML configuration files, Java projects that have Ruby build scripts for unknown reasons, log files, there's an unlimited number of text formats that some tool somewhere knows how to parse, but it isn't worth your while to spend half an hour setting up an obscure dev environment for a one-off job that you can do in five minutes in emacs.
It goes way beyond "find references" and "go to definition", too. For instance, marking a variable indicates all the usage points near the scroll bar, and ctrl-,/. moves between them. Ctrl-T opens an index of all classes. Ctrl-O has an index of the current file. Shift-Alt-Up/Down selects increasingly large expressions (across lines, as well).
I guess it all sounds trivial, but taken together I navigate within a Java project in an entirely different way than I navigate source code for other languages or unstructured files.
 Grep and its ilk are obviously still common tools for things like log files.
It supports slicing operations: Which downstream changes happen if I change this variable? What upstream control structures influence the value of this variable?
A lot of my coworkers swear by Understand, but I haven't tried it myself.
ideally, it would be really nice if cscope/xcscope provided some means of creating (and saving) bunch of call-stacks, which you can navigate in forwards/reverse direction (not sure if i am making much sense here...)
It worked well for my own usage, but it probably needs some love. I'm not sure how *cscope works though, and if it would be possible to add it as another backend.
Also, I've had a lot of luck with pycscope.py for python code + cscope tools.
GNU Global needs some love though. It'd be awesome to see it handle more stuff.
What's the advantage to acscope vs. the others? I personally use xcscope, I don't even remember why I initially chose it...
ascope is an improvement over xcscope that runs all queries through a single cscope process, instead of starting a new process and reloading the database for each query. It was made by Staton Sun as a merge between xscope and another single-process interface, bscope
cscope, etags, etc can't really handle this task. Eclipse and Visual Studio can.
I am really disappointed that emacs and vim who spout productivity over learning curve allow important productivity tasks such as this to remain more than a decade behind the state of the art.
Find files in DIR containing a regexp REGEXP and start Dired on output.
In fact it seems the pure C version is free now; it claims full indexing of 3 million lines of Linux source code in 15 minutes.
I used to keep track of it many years back back while it was being developed, but then C++ starting making up 1% of my projects rather than 99% (nowadays I use the JetBrains PyCharm for those 99% which does a reasonable effort).
It seems the tool hasn't seen updates in a while (forget about C++11 support), but I remember it working much better than just tags, cscope last time I tried it many years ago.
(global-set-key "\M-/" 'tags-search)
(global-set-key "\M-'" 'tags-apropos)
1. Generate a file index, every now and then(read cron).
> find . > fileList
3. If the previous routine was written in bash written a elisp function that calls the bash script.
That's mostly it.
I could use cscope but it sucks donkey balls with java. When I ask for definition it finds a declaration.
Ack plugins for Emacs: