Does anyone know how to use column mode for editing source code within uVision3 IDE??? I am used to using Jens where I could insert text in a column format defined by the start and stop of the drag of my cursor. I could then right click and insert defined text on all lines simultaneously starting where my cursor orignated to the end of the highlighted segment area in a column format. Can anyone help with this??? I know how to use the Alt key and mouse to highlight text in a column format, but I can find a way to enter text on multiple lines in a column format. This is very annoying.
Alt + drag
http://www.keil.com/support/man/docs/uv4/uv4_ui_seltext.htm
Thank you for the reply, but I am not talking about just selecting text. I want to insert text in a similar format. However, I would like to define the location of the column to start on, then drag to the line where I want to stop at. Then I would like to somehow insert text at this defined location on all lines within the drag highlighted area starting from where the cursor began and stopped.
"It would be nice for one if a company took into consideration all of the editing features that are used by 99% of the software programming community"
that's precisely what the ide does not offer the column editing mode you are looking for: 99% of the software programming community does not use it.
that doesn't mean that nobody uses it. just that there isn't sufficient number of people using it to justify its existence in the ide.
I don't know the actual number, but the consensus here certainly does seem to be that it is not a "mainstream" or "widely-used" feature.
"there isn't sufficient number of people using it to justify its existence in the ide"
It could actually be that it's a sufficiently esoteric feature that the uVision developers hadn't even heard of it - let alone made any specific decision about whether or not to implement it!
I think uVision has many much more pressing deficiencies that ought to be addressed before this one! And these are things that really are common in other tools!
Eclipse seems to me to be a perfect illustration that "It would be huge to the point of being unusable"
You'll note I offered Eclipse as an example to support my assertion that the world has too many bad IDEs already, and could benefit from concentrating that effort some. I never claimed it was lightweight.
Eclipse is indeed huge. Don't even bother trying it in eanest on less than 1 GB. But it offers a lot of features for that price --- and the main advantage is that it's a framework, i.e. if you don't need a particular feature, you don't have to install it.
Eclipse is for IDEs what EMACS is among editors: their common property is that whatever feature you come up with that (you thing) you need, odds are high in the 90% range it's either already in the core package, or somebody has already written a plugin/module to do it.
Eclipse is so versatile and so configurable that, it seems to me, it's (almost) impossible for any non-specialist to understand!
It's not really all that bad. The default installation plus CDT plugin works fine as a generic IDE for C programs. It'll blow quite a number of vendor's self-made IDEs clean out of the water already. I've seen enough crappy IDEs in my time to have learned this: the world is better off with compiler writers not putting their time into user interface programming. The skillsets are just too different.
Specialist knowledge only gets involved if you were to make your own Eclipse plugins to, e.g., make an equivalent of uVision's devicse database and simulator engine part of Eclipse. And I rather much doubt that connecting such an engine into Eclipse is really harder than building your own IDE, project management, compiler option management, etc., from scratch.
Oh, and Eclipse does not do any form of column editing - it doesn't even do what uVision does!
Release documentation of Eclipse 3.5 seems to contradict you on that.
I appreciate all of the help and constructive criticism that I have received as feedback for my post. It seems that of the few people that I have got feedback from do NOT use column editing features. I don't know how you deal with that when you have to write hundreds of lines of code that are similar to the definition of several variables all the way to the level of functions. I would much rather write one line than one hundred, but I guess that's just me. You are right however, I could just copy the text that I want to manipulate and move it to the program that I know can execute column editing and paste it in. It's really not that tough, but it would be nice to have that available within the uVision IDE. In any event, it's still pretty frustrating to buy a software package and get no documentation that identifies all of the editing posibilities. In this case I had to go on a rabbit chase on the web to find out if something was available within the IDE instead of being able to find it within the IDE's documentation.
If you post an example of what you start with, and what editing you need to do, we might be able to give you suggestions what editing we do for similar problems.
If many things needs to work on a large set of variables, I normally use a struct. Then I can create a single variable of that struct at many places, instead of having to create a large number of variables again, and again.
Often, when I see repetitions, I have a strong reason to repackage the code into reusable modules instead of having multiple sets of similar code.
I use column editing quite a bit too. My solution is to edit and build within SlickEdit. The build button in SlickEdit can be configured to call uVision with the -b option to build a project. Errors are redirected to the build window so the error parser can see them. The only time I use uVision is to setup or modify a project. Yes, this is a workaround but it works pretty well for me and sure beats using the uVision editor. I prefer using eclipse with the SlickEdit core plug-in. Hopefully someday Keil will see the light and switch to the eclipse framework like many others have.
-Walt
I'm quite happy Keil is _not_ using the eclipse framework.
"It seems that of the few people that I have got feedback from do NOT use column editing features"
Not the specific "column editing feature" that you're speaking of - but I do quite often use the feature as found in uVision.
"I don't know how you deal with that when you have to write hundreds of lines of code that are similar to the definition of several variables"
I would write the first one, copy, paste it as many times as required, then overtype the bits that differ. Or start with a "skeleton", copy-&-paste, then fill in the blanks.
"I could just copy the text that I want to manipulate and move it to the program that I know can execute column editing and paste it in"
That's one way to do it. Alternatively, just open that file in the other editor - no need for copy/paste between applications.
As Per says, if you're routinely getting into that situation, maybe you should be looking at a better way to structure the code...?
some comments: I do not know about "99% of the software programming community" but I am a heavy user of column editiing: a) /* some inactive code */ b) //inactive ... //inactive ... //inactive ... b) is dead easy with the colun editing in CodeWright. b) has the HUGE advantage that the disabled lines show up as disabled on global searches.
The editor in the IDE of any small micro Compiler maker I know of STINKS. I am not suggesting they should redirect their efforts from compiler/linker to the editor, they should accept the fact and make an interface so that e.g. CodeWright could be hooked in.
the comment above about "going outside the IDE" is ridiculous, how big an effort is it to click on the taskbar instead of on "edit"
Erik
"but it would be nice to have that available within the uVision IDE."
it would be nice. but writing an ide is very much writing a piece of code: there are lots of nice things that we could but don't do in our code.
As a matter of fact, the greatest code we write has left out many nice features. and probably because of that, it is the greatest code we have written.
"Often, when I see repetitions, I have a strong reason to repackage the code into reusable modules instead of having multiple sets of similar code."
agree with that.
"make an interface so that e.g. CodeWright could be hooked in."
alternatively, it is not that difficult to embed the compiler in your own choice of editors (SI, CB, etc.).
As I've already mentioned, it's easy too hook Codewright in to edit the selelcted file from the uVision Project view, or the current file in the uVision editor - including going to the current line.
And there are documents here for hooking in to the debugger: http://www.keil.com/appnotes/list/uvision.htm
Note that a number of programmers editors have special features for comment away code, so you can do your alternative "b" without any column editing involved.
CodeWright has that feature but, as far as I know, doesn't have a correesponding un-comment feature.
So you still need the column edit (or, at least, column select & delete) for that...
A bit sloppy to add a "comment" but not a "uncomment" command. A column-delete command is hard, always removing a fixed number of characters. A uncomment command would be able to check each individual line and don't destroy a line that doesn't start with //. No problem if the block is 10-20 lines, but if it is 500 lines, it can be easy to miss that there is some code in the middle that wasn't commented away, when marking the block.
Yes - unless I've just missed it?
I guess the intention must've been just to make a block of text into a comment - rather than to enable/disable sections of code...?
View all questions in Keil forum