Watch, Follow, &
Connect with Us

Please visit our new home
community.embarcadero.com.


Welcome, Guest
Guest Settings
Help

Thread: C++ code completion project "Xerias"


This question is not answered. Helpful answers available: 2. Correct answers available: 1.


Permlink Replies: 1 - Last Post: May 19, 2017 1:11 AM Last Post By: Vladimir Zhuchko Threads: [ Previous | Next ]
Florian Koch

Posts: 3
Registered: 5/7/10
C++ code completion project "Xerias"  
Click to report abuse...   Click to reply to this thread Reply
  Posted: May 16, 2017 8:56 PM
Greetings.

For years I'm very dissatisfied with the performance of C++Builder's code completion feature (especially since the introduction of Clang).

From time to time there were statements by Embarcadero, like "We are working on it...", but I still have crashed IDEs multiple times a day when invoking code completion.

Also Embarcadero does only mention EITHER an improvement of code completion OR (finally) the implementation of C++ refactoring functions in the current roadmap scheduled for Tokyo Update 2.

As I call myself a "heavy user", who uses C++Builder at work and also sometimes in his spare-time, I'm getting a little jealous when I see things like IntelliSense in Visual C ++.

Since I no longer wanted to "suffer", I started a private Open Tools API project in C++ called "Xerias" earlier this year,
whose main objectives are to create a better and compilation-independent C++ code completion based on
SQLite and Universal Ctags.

In the last four months I think that I 've made quite a good progress (also with many nice hints from here).
However, I've also noticed that my spare-time would be not enough to complete the wizard alone in the medium term.

I am now thinking about putting my work under an open source license, in order to create a nice solution together with others.

Tagging the files and creating the tag database already works quite well.
All "basic structures" stand.

I think my sources are sufficient and understandable. Also, the code should be relatively easy to understand even for non-C++ gurus (VCL, a little STL and boost).

__published
public
protected
private

order and PascalCase style.

An example for my coding style can be found here:
[https://gist.github.com/FlKo/6a55a1e45291522c3ad3f5229e5cc533]

So, my question now is:
If I'll publish the wizard's project in it's current unfinished state under the GPLv3, is there any interest in completing it jointly?

Originally I've asked this in the C++Builder community on Google+, where I already got some responses: [https://plus.google.com/105641319933437744493/posts/aYSPWhjkCFx]

Vladimir Zhuchko

Posts: 107
Registered: 1/22/98
Re: C++ code completion project "Xerias"  
Click to report abuse...   Click to reply to this thread Reply
  Posted: May 19, 2017 1:11 AM   in response to: Florian Koch in response to: Florian Koch
Florian Koch wrote:
Greetings.

For years I'm very dissatisfied with the performance of C++Builder's code completion feature (especially since the introduction of Clang).

From time to time there were statements by Embarcadero, like "We are working on it...", but I still have crashed IDEs multiple times a day when invoking code completion.

Also Embarcadero does only mention EITHER an improvement of code completion OR (finally) the implementation of C++ refactoring functions in the current roadmap scheduled for Tokyo Update 2.

As I call myself a "heavy user", who uses C++Builder at work and also sometimes in his spare-time, I'm getting a little jealous when I see things like IntelliSense in Visual C ++.

Since I no longer wanted to "suffer", I started a private Open Tools API project in C++ called "Xerias" earlier this year,
whose main objectives are to create a better and compilation-independent C++ code completion based on
SQLite and Universal Ctags.

In the last four months I think that I 've made quite a good progress (also with many nice hints from here).
However, I've also noticed that my spare-time would be not enough to complete the wizard alone in the medium term.

I am now thinking about putting my work under an open source license, in order to create a nice solution together with others.

Tagging the files and creating the tag database already works quite well.
All "basic structures" stand.

I think my sources are sufficient and understandable. Also, the code should be relatively easy to understand even for non-C++ gurus (VCL, a little STL and boost).

__published
public
protected
private

order and PascalCase style.

An example for my coding style can be found here:
[https://gist.github.com/FlKo/6a55a1e45291522c3ad3f5229e5cc533]

So, my question now is:
If I'll publish the wizard's project in it's current unfinished state under the GPLv3, is there any interest in completing it jointly?

Originally I've asked this in the C++Builder community on Google+, where I already got some responses: [https://plus.google.com/105641319933437744493/posts/aYSPWhjkCFx]


I like this idea, the last working code completion was in the C++Builder6.
--
With best regards
Dr. Vladimir Zhuchko
Legend
Helpful Answer (5 pts)
Correct Answer (10 pts)

Server Response from: ETNAJIVE02