Page 1 of 4 123 ... LastLast
Results 1 to 10 of 34

Thread: KDevelop With Clang Greatly Improves Its C++ Handling

  1. #1
    Join Date
    Jan 2007
    Posts
    13,431

    Default KDevelop With Clang Greatly Improves Its C++ Handling

    Phoronix: KDevelop With Clang Greatly Improves Its C++ Handling

    The KDevelop Clang plug-in has been greatly improved and is on its way to replacing the KDE's integrated development environment existing C++ language support with this Clang-based solution...

    http://www.phoronix.com/vr.php?view=MTU5MTg

  2. #2
    Join Date
    Sep 2012
    Posts
    568

    Default

    So I'll just quote what has been written today, just to highlight how some posters debate in good faith:
    Quote Originally Posted by Ericg
    4) Better integration into IDEs for those that like to use IDEs.
    Quote Originally Posted by brosis
    4) Only valid for proprietary IDE

  3. #3
    Join Date
    Jan 2014
    Posts
    2

    Default Hope

    I hope all of these clang/llvm articles try to push gcc people.

    Ofcourse, i know this site bias against GNU and mr larrabel is against rms.

    this war show gcc would never die with these cheap advertises from apple.
    Last edited by amirsdream; 02-04-2014 at 01:11 PM.

  4. #4
    Join Date
    Jan 2014
    Posts
    22

    Default

    I don't understand why GNU's GCC doesn't allow such features easily. After all, shouldn't GCC have embraced "do one thing and do it well" approach and should have been modular and therefore easily expose its code validation to outside? I know GCC has a code testing without actual compilation but apparently it is not really what it should be since all these good news about clang. Maybe phoronix is just feeding FUD.

    Hoping they make GCC to have a better infrastructure in short time.

  5. #5
    Join Date
    Jan 2013
    Posts
    26

    Default

    Quote Originally Posted by the303 View Post
    I don't understand why GNU's GCC doesn't allow such features easily. After all, shouldn't GCC have embraced "do one thing and do it well" approach and should have been modular and therefore easily expose its code validation to outside? I know GCC has a code testing without actual compilation but apparently it is not really what it should be since all these good news about clang. Maybe phoronix is just feeding FUD.

    Hoping they make GCC to have a better infrastructure in short time.
    GCC is compiler, not static analyser of image processor or web server. Compiler get human readable text, output binary target machine codes.

    Static analyser get text, output human readable related things. Maybe replace each sentence of input file by perfect poetry variant, who know.

    JIT emulate some abstract hardware on-fly.

    Why GGC team MUST inclute unrelated features? Just because some one else do same? To entertain audience? Why not tetris game, of hot chicks pictures? imagine, you start compling and same time screen loads of dosens of kittens!

  6. #6
    Join Date
    Jun 2011
    Posts
    624

    Default

    Quote Originally Posted by the303 View Post
    I don't understand why GNU's GCC doesn't allow such features easily. After all, shouldn't GCC have embraced "do one thing and do it well" approach and should have been modular and therefore easily expose its code validation to outside? I know GCC has a code testing without actual compilation but apparently it is not really what it should be since all these good news about clang. Maybe phoronix is just feeding FUD.

    Hoping they make GCC to have a better infrastructure in short time.
    Well basically the reason for this amounts to some political decisions in the past when the effectiveness of the GNU/GPL had not as of yet been established and FOSS was still really an experiment, and so in order to prevent proprietary developers from taking the code base and not releasing their modifications to the code they decided to go for a monolithic architecture and have been dealing with fixing that design decision ever since. The other problem is that now that the code based has been there and set up for forever... which is to say now that they have a product that is basically everywhere they can't just stop the presses and do a redesign they have to work slowly in that direction, while continuing to release on the current stack, and so that will take a long time if they decide to do it. Or at least that's my understanding of the related history anyway.

  7. #7
    Join Date
    Sep 2008
    Posts
    112

    Default

    Quote Originally Posted by storm_st View Post
    GCC is compiler, not static analyser of image processor or web server. Compiler get human readable text, output binary target machine codes.

    Static analyser get text, output human readable related things. Maybe replace each sentence of input file by perfect poetry variant, who know.

    JIT emulate some abstract hardware on-fly.

    Why GGC team MUST inclute unrelated features? Just because some one else do same? To entertain audience? Why not tetris game, of hot chicks pictures? imagine, you start compling and same time screen loads of dosens of kittens!
    Are you trying to twist around what the303 said to make it sound like he's AGAINST the unix philosophy? He's not suggesting new features. He's suggesting a complex task ought to have been cut up into modular parts, which is exactly what is at the core of the unix philosophy.

    Just because you can sum up any program into one big "it does X" doesn't mean the hugely complex X isn't a bundle of separate parts that ought to be well seperated. The ENORMOUSLY complex task of converting human readable text to machine code might be;
    0. Preprocessor
    1. Construct the AST
    2. Construct intermediate representation
    3. Optimizer
    4. Code generator

    And without having ever looked the sources for GCC, I'd bet it's essentially the components listed above. It's just not exposed as API-stable libraries as far as I know.
    Instead we have editors that are now unable to making full use of the compilers capabilities, and as a result has to build their own, often buggy, re-implementations.

  8. #8
    Join Date
    Aug 2012
    Location
    Pennsylvania, United States
    Posts
    1,750

    Default

    Quote Originally Posted by erendorn View Post
    So I'll just quote what has been written today, just to highlight how some posters debate in good faith:
    I love how I give Michael a shout-out to that blogpost last night and then first comment on the story I get quoted and vindicated... thank you Erendorn for making me smile haha

  9. #9
    Join Date
    Jan 2013
    Posts
    26

    Default

    So what? LLVM is algorithm and mess with text and binaries. Tetris is algorithm and mess with text and binaries. Emacs operating system ...

    Imagine there will be another project that work with text and graph operations in some modern way. Sublime text editor as example. Cool features. Why LLVM do not integrate it? Lets start paid massive PR actions over Phoronix to say that Sublime can rename one word in place and LLVM+clang can not. And Shim+Grub can parse text comfig and boot computer, but LLVM can not do it even after 10+years of development.

  10. #10
    Join Date
    Jan 2013
    Posts
    96

    Default

    Quote Originally Posted by storm_st View Post
    So what? LLVM is algorithm and mess with text and binaries. Tetris is algorithm and mess with text and binaries. Emacs operating system ...

    Imagine there will be another project that work with text and graph operations in some modern way. Sublime text editor as example. Cool features. Why LLVM do not integrate it? Lets start paid massive PR actions over Phoronix to say that Sublime can rename one word in place and LLVM+clang can not. And Shim+Grub can parse text comfig and boot computer, but LLVM can not do it even after 10+years of development.
    What. Is this an actual attempt to communicate?

    This makes sense because LLVM is a C++ parser and static analyser (and compiler, but that's not important). KDevelop has a C++ parser, but could use a better one. Assuming (with good reason) that LLVM will continue to be developed, it'll have every relevant C++-parsing feature at no cost to the KDevelop devs. It won't be able to boot computers, because that's not a useful feature in a C++ parser.

    This is definitely good news - I get my code parsed faster (even on my tiny projects the parser lag is noticeable) and the devs can work on other awesome features to make my coding life easier.

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •