On Tue, Apr 10, 2001 at 03:27:40PM -0500, David Huttleston Jr wrote: | On Tue, Apr 10, 2001 at 01:34:05PM -0600, Dave Brueck wrote: | > > | Can gVim do the following? | > > | | > > | a) Have autocompletion / intellisense in the editor. | > > | > > no, this is one feature that would be really cool. OTOH, it makes for | > > a lazy coder who can't (doesn't, rather) remember the | > > function/variable names. (This is from experience using JBuilder for | > > a while) | > | > Hi D-Man, | > | > Lazer coder? How about "makes for a more productive coder". Just as Python Well, it depends on how you evaluate it. I found that when I was relying on JBuilder's auto-complete I didn't really know the function names, and I couldn't describe it very well to others without first looking it up. (Admittedly, I wasn't very experienced with the Java libs yet, and hadn't used java for about a year) I didn't have to spend the time to look it up when I typed it, but I had to search through the possible completions each time I type it. I've been using gvim much more heavily (like, completely -- it is just too powerful to leave behind when modifing stuff) for quite a while and I find that while I must look at the docs in netscape, I have to look less since I remember what the function names are. I'm not saying that auto-completion is a bad feature, but too much dependence on it is. Maybe it is just the way the comletion worked in JBuilder and/or my experience level. I'm thinking of bash now, and how I _always_ use auto-completion, yet I know the command/program names pretty well. Whenever I am at a csh or DOS shell, I get lots of whitespace, then I complain and type the whole command or filename <wink>. | > Lack of autocompletion tends to _encourage_ laziness and _lower_ | > productivity. One example is that because the coder has to type out the full | > variable and function names he ends up naming them something short and less | > descriptive, making the code less clear and harder to maintain. Worse, the | > coder then adds a useless comment to tell the purpose of the function, the | > same information that a decent function name would have conveyed. I tend to like short _and_ descriptive names, sacrificing shortness to improve effectiveness. If method/variable names are consistently very long, they won't fit on a line very well, and can make it harder to read. It's all about balance. | use ctags (aka Exuberant CTags http://ctags.sourceforge.net/) to provide | most of what you want from code-completion. In addition, ctags allows | quick navigation through source code using mouse or keyboard to jump to | the definition of a class or function and back. CTags supports python, | java, c, c++, eiffel, and many more languages. Oh ..., yeah ..., ctags. Something I've been meaning to learn about. I'll have to remember what they do, and go learn them soon! | Current I use Emacs because I find the Object Browser and the JDE package | provide a more complete enviroment for java and jython which we use heavily. The JDE package is pretty cool. I used emacs for a while (~ one quarter of class, then I forgot all the commands, and learned how to make vim behave nicely (real vi doesn't have all the comforts needed for coding, IMO)). JDE is quite close to what I would want in an IDE -- a browser on the side to view the files/modules/classes and jump to it in your editor. Perhaps scripting vim with python, or using python to drive (vim | <your favorite editor with python bindings attached>) would be the solution. -D
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4