I work usually with valgrind+kcachegrind to profile C++ codes. A new code I am working with uses very long function names, so that the graphical results are a mess. I wonder how can one shorten the function names so that they fit in a small box.
This is sort of a dirty workaround... you could be to write a script that does a find and replace for a list of function names for the files generated by Valgrind and then use kcachegrind to visualize the data.
Maybe you could shorten the function names in the code by using namespaces?
Related
I have a list of functions in a text file that I'd like to expose to LLVM for its execution engine at run time, I'm wondering if its possible to find pointers to the functions at runtime rather than hard code in all the GlobalMappings by hand as I'd probably like to add in more later. For example:
// File: InternalFunctions.txt
PushScreen
PopScreen
TopScreen
// File: ExposeEngine.cpp
// Somehow figure out the address of the function specified in a string
void* addy = magicAddress("PushScreen");
jit->addGlobalMapping(llvmfunction, addy);
If this is possible I love to know how to do it, as I am trying to write my game engine by jit-ing c++. I was able to create some results earlier, but I had to hard-code in the mappings. I noticed that Gtk uses something along the lines of what I'm asking. When you use glade and provide a signal handler, the program you build in c will automatically find the function in your executable referenced by the string provided in the glade file. If getting results requires me to look into this Gtk thing I'd be more than happy to, but I don't know what feature or part of the api deals with that - I've already tried to look it up. I'd love to hear suggestions or advice.
Yes, you can do this. Look at the man pages for dlopen() and dlsym(): these functions are standard on *nix systems and let you look up symbols (functions or variables) by name. There is one significant issue, which is that C++ function names are usually "mangled" to encode type information. A typical way around this is to define a set of wrapper functions in an extern "C" {} block: these will be non-member, C-style functions which can then call into your C++ code. Their names will not be mangled, making them easy to look up using dlsym().
This is a pretty standard way that some plugin architectures work. Or at least used to work, before everyone started using interpreted languages!
How does std.conv.to!string(enum.member) work? How is it possible that a function takes an enum member and returns its name? Does it use a compiler extension or something similar? It's a bit usual to me since I came from C/C++ world.
What it does is use compile time reflection on the enum type to get a list of members (the names as strings) and their values. It constructs a switch statement out of this information for a fast lookup to get the name from a value. to!SomeEnum("a_string") uses the same principle, just in the other direction.
The compile time reflection info is accessed with __traits(allMembers, TheEnumType), which returns a list of strings that can be looped over to build the switch statement. Then __traits(getMember, TheEnumType, memberName) is used to fetch the body.
Traits can be seen more of here: http://dlang.org/traits.html#allMembers
That allMembers one works on many types, not just classes as seen in the example, but also structs, enums, and more, even modules.
The phobos source code has some examples like EnumMembers in std.traits: https://github.com/D-Programming-Language/phobos/blob/master/std/traits.d#L3360
though the phobos source is kinda hard to read, but on line 3399, at the bottom of that function, you can see it using __traits(allMembers) as its data source. std.conv.to is implemented in terms of many std.traits functions.
You can also check out the sample chapter tab to get the Reflection chapter out of my D cookbook which discusses this stuff too:
http://www.packtpub.com/discover-advantages-of-programming-in-d-cookbook/book
The final example in that chapter shows how to use several of the reflection capabilities to build a little function dispatcher based on strings. The following chapter (not available for free though) shows how to build a switch out of it for better efficiency too. It's actually pretty easy: just put the case statements inside a foreach over the compile time data and the D compiler will unroll then optimize the lookup table for you!
A little something that could be borrowed from IDEs. So the idea would be to highlight function arguments (and maybe scoped variable names) inside function bodies. This is the default behaviour for some C:
Well, if I were to place the cursor inside func I would like to see the arguments foo and bar highlighted to follow the algorithm logic better. Notice that the similarly named foo in func2 wouldn't get highlit. This luxury could be omitted though...
Using locally scoped variables, I would also like have locally initialized variables highlit:
Finally to redemonstrate the luxury:
Not so trivial to write this. I used the C to give a general idea. Really I could use this for Scheme/Clojure programming better:
This should recognize let, loop, for, doseq bindings for instance.
My vimscript-fu isn't that strong; I suspect we would need to
Parse (non-regexply?) the arguments from the function definition under the cursor. This would be language specific of course. My priority would be Clojure.
define a syntax region to cover the given function/scope only
give the required syntax matches
As a function this could be mapped to a key (if very resource intensive) or CursorMoved if not so slow.
Okay, now. Has anyone written/found something like this? Do the vimscript gurus have an idea on how to actually start writing such a script?
Sorry about slight offtopicness and bad formatting. Feel free to edit/format. Or vote to close.
This is much harder than it sounds, and borderline-impossible with the vimscript API as it stands, because you don't just need to parse the file; if you want it to work well, you need to parse the file incrementally. That's why regular syntax files are limited to what you can do with regexes - when you change a few characters, vim can figure out what's changed in the syntax highlighting, without redoing the whole file.
The vim syntax highlighter is limited to dealing with regexes, but if you're hellbent on doing this, you can roll your own parser in vimscript, and have it generate a buffer-local syntax that refers to tokens in the file by line and column, using the \%l and \%c atoms in a regex. This would have to be rerun after every change. Unfortunately there's no autocmd for "file changed", but there is the CursorHold autocmd, which runs when you've been idle for a configurable duration.
One possible solution can be found here. Not the best way because it highlights every occurrence in the whole file and you have to give the command every time (probably the second one can be avoided, don't know about the first). Give it a look though.
This question already has answers here:
Closed 12 years ago.
Possible Duplicate:
Dynamic source code in C++
is it possible to let the user type in a function and then run that function without using a lot of if's or a huge switch?
It is not possible to execute arbitrary c++ code in your program, since you than need a c++ compiler inside your program. But you could try to embed Python to your program. Boost python makes this relatively easy. The user can than write a python function that is executed and can interact with the classes and functions of your program. You need to make your functions explicitely visible to python.
What ever a user types in will be text, or a string. The only way I know to have it get mapped to a function is to use if/else or switch statements. That or the cringe inducing option of mapping each of your functions to a UI widget.
The end of the story, is it's your code. You have to write, and live with it. Just be careful, your program may be wildly successful, and you may not write code anymore, and then someone else will have to maintain your code. So be nice to the maintenance programmer who may follow you, and write code that isn't too tricky to figure out.
I assume you want something like eval from php.
You can try to play with command design pattern, but I doubt it will be an easy task. Basically you need to write simple C++ interpreter.
What type of function do you mean? A C++ function? If so, then you will have to either (1)interpret it or (2)compile and execute it. Interpretation would be the more likely choice here. I'm not sure if there are libraries out there already to do this but I'd assume there are.
If you don't like mega-if's or huge switches, you may be SoL on any solution for anything ever, but then again there is seldom one perfect way to do things. Consider looking in to various logic structures and algorithms to see how to do something that would normally be the job of a 23-case switch could be done another way. Like I said initially, however, sometimes you really do just need a million nested if's to do what you want to.
No, in C++ this is not possible. C++ is a compiled language. When the program runs, the compiler doesn't need to be accessible, or even installed on the machine that runs the program.
If you want to do this in C++, you need to write your own interpreter that parses whatever the user enters.
Here is my best idea, but it is a tad memory intensive.
First, create a class, lets call it MyFuncPtr to store a union of several different types of pointers to functions and an integer to tell which type it is. Overload the () operator to call the function stored with a variable length argument list. Make sure to include some sort of run-time argument checking.
Finally create a map of strings to MyFuncPtrs. Store your functions in this map along with their names. Then all you need to do is feed the name into the [] command to get a function that can be easily called. Templates could probably be used to aid in the making of MyFuncPtr instances.
This would be the easiest if it were plain C functions and no name mangling is performed on the symbols (use extern "C" { ... })
With some platform-specific code you can get the address of a function by its name. Then you cast the address as a function pointer which you can use to call the function.
On windows you must be using GetProcAddress and dlsym on Posix compliant platforms.
So you know off the bat, this is a project I've been assigned. I'm not looking for an answer in code, but more a direction.
What I've been told to do is go through a file and count the actual lines of code while at the same time recording the function names and individual lines of code for the functions. The problem I am having is determining a way when reading from the file to determine if the line is the start of a function.
So far, I can only think of maybe having a string array of data types (int, double, char, etc), search for that in the line and then search for the parenthesis, and then search for the absence of the semicolon (so i know it isn't just the declaration of the function).
So my question is, is this how I should go about this, or are there other methods in which you would recommend?
The code in which I will be counting will be in C++.
Three approaches come to mind.
Use regular expressions. This is fairly similar to what you're thinking of. Look for lines that look like function definitions. This is fairly quick to do, but can go wrong in many ways.
char *s = "int main() {"
is not a function definition, but sure looks like one.
char
* /* eh? */
s
(
int /* comment? // */ a
)
// hello, world /* of confusion
{
is a function definition, but doesn't look like one.
Good: quick to write, can work even in the face of syntax errors; bad: can easily misfire on things that look like (or fail to look like) the "normal" case.
Variant: First run the code through, e.g., GNU indent. This will take care of some (but not all) of the misfires.
Use a proper lexer and parser. This is a much more thorough approach, but you may be able to re-use an open source lexer/parsed (e.g., from gcc).
Good: Will be 100% accurate (will never misfire). Bad: One missing semicolon and it spews errors.
See if your compiler has some debug output that might help. This is a variant of (2), but using your compiler's lexer/parser instead of your own.
Your idea can work in 99% (or more) of the cases. Only a real C++ compiler can do 100%, in which case I'd compile in debug mode (g++ -S prog.cpp), and get the function names and line numbers from the debug information of the assembly output (prog.s).
My thoughts for the 99% solution:
Ignore comments and strings.
Document that you ignore preprocessor directives (#include, #define, #if).
Anything between a toplevel { and } is a function body, except after typedef, class, struct, union, namespace and enum.
If you have a class, struct or union, you should be looking for method bodies inside it.
The function name is sometimes tricky to find, e.g. in long(*)(char) f(int); .
Make sure your parser works with template functions and template classes.
For recording function names I use PCRE and the regex
"(?<=[\\s:~])(\\w+)\\s*\\([\\w\\s,<>\\[\\].=&':/*]*?\\)\\s*(const)?\\s*{"
and then filter out names like "if", "while", "do", "for", "switch". Note that the function name is (\w+), group 1.
Of course it's not a perfect solution but a good one.
I feel manually doing the parsing is going to be a quite a difficult task. I would probably use a existing tool such as RSM redirect the output to a csv file (assuming you are on windows) and then parse the csv file to gather the required information.
Find a decent SLOC count program, eg, SLOCCounter. Not only can you count SLOC, but you have something against which to compare your results. (Update: here's a long list of them.)
Interestingly, the number of non-comment semicolons in a C/C++ program is a decent SLOC count.
How about writing a shell script to do this? An AWK program perhaps.