This has been discussed before, but not recently.... Has there been any progess in making uv3 not open all its windows when using the -b command line option? It is rather annoying when all you want to do is compile from within emacs and it insists on waisting all that time opening 10s if windows for no reason at all....
FWIW, I use SlickEdit and use uv3.exe -b to build my projects by attaching the command to the build button. I direct the results to a file which SlickEdit parses and takes me to the line(s) containing errors in the source code. Works pretty good. I'm also irritated by waiting for the GUI to open but I don't believe there is a way around it unless you move away from the uv3.exe -b command. UV3 does use an option file to remember which source files were open last so if you close all files and exit out of UV3, the next time it opens to build, no source files will open. It's a little quicker that way. -Walt
so if you close all files and exit out of UV3, the next time it opens... Isn't it wonderful: you need to find inconvenient ways to make what supposedly makes it convenient work better. Erik
You may also minimize µVision. Then close it by pressing Close from the context menu of its task bar item. µVision will save that state an start minimized the next time. You will only see the splash screen. Generally I would build the project only once using µVision and then use the batch file, that µVision can generate automatically.
I'm going to come out in support of Keil: I use the IDE and I think it's excellent. Is the problem not just that you guys are used to other editors and are too old to learn new tricks?
Is the problem not just that you guys are used to other editors and are too old to learn new tricks? Sure, partially it is. But its also hard to forget tricks which are no longer tricks but are everyday parts of life, but don't work with this IDE. There are lots of little things it does badly. eg just one example.... When you build it puts the output from the compiler into a window. But it always shows the last output lines by default. This is not what im interested in, i want to see the first bit of output since there is the first error and what i need to fix. What comes after that is less important, its often a cascade of errors caused by the first error. OK you then press F4 to goto the first error. Thats OK. But when the compiler has produced N errors for one line of code i have to press F4 N times in order to move onto the next line with an error Why? I can see from the compiler output there was multiple errors so i don't need to be taken to the same line N times. I want to goto the next line with an error. OK, you fix the error which involves inserting a few lines of code. Pressing F4 takes you to the old line number, not the new line number becasue of the inserted lines... These are all things that a more mature editor would do better. The editor i use, emacs, is over 20 years old and does all these little things how i like. There are many more little things that the IDE does not so well. And i get no choice. I have to do horrible tricks like we are dicussing here in order to use a better editor. As i said before, we are getting away from the subject and more into philisophy and editor wars.....
"As i said before, we are getting away from the subject and more into philisophy and editor wars....." Don't worry, I'm not after an editor war, and I have little doubt that your editor is better in many ways than the IDE. I was just trying to defend what I find to be an excellent tool. I also admit to trolling a bit with the 'too old to learn new tricks' part. I unfortunately fit quite well into that category...
Matthias, Nice trick, thanks! Stefan, You don't need to defend Keil's IDE from me, I think it's fine. I use it maintain project build info and debugging. I don't use the editor simply because it doesn't have the features I'm used to like brief emulation, auto listing structure members, auto listing function parameters, context tag completion and differencing directories just to name a few. -Walt
re "defending the Keil IDE" As I read the thread, I see that all that do not use the Imbecile Developer Environment say "for applying a wrong concept, Keil has done a fair job" so there should be no reason to defend Keil in this respect. ANY IDE can only be developed by making assumptions about what the user wants and thus it must end up as an insult to thinking individuals that know what they want, since ANY IDE will support what the dveloper of the IDE think the user wants, not necessarily what the user actually wants. So, if I use an IDE i will have to change my wants to the wants the developer of the IDE thought I had. You know something, writing the above I just got it. Here is the skinny: if you have not been in a situation where you could develop your own wants, the assumed wants of yours will fit. Thus anyone that has not developed without the IDE will be happy with it. BTW the main reason I do not use the IDE is not any of the above, it is because it is incable of developing multiple flavors without separate directives. Also, I would never use an IDE where a non-replacable editor was less capable than CodeWright. Erik
"Thus anyone that has not developed without the IDE will be happy with it." Not necessarily. I have developed without IDEs, and with IDEs. Currently, I use the Keil IDE. Yes, it has limitations - and I frequently use other editors (particularly CodeWright) in favour of the built-in one (I have created an item on the uVIsion 'Tools' menu to do it). But I find that the advantages outweigh the limitations. "BTW the main reason I do not use the IDE is not any of the above, it is because it is incable of developing multiple flavors..." I agree that this is a particular weak point in uVision (i've commented on it here before). Currently, it's not a big deal for me - so I stick with uVision. But if I needed to do loadsa variants as you do, I think that probably would preclude uVision... :-( PS Have you looked at Eclipse: http://eclipse.org/ Perhaps if Keil et al just concentrated on making their tools Eclipse-compatible, they could all stop re-inventing their own IDE wheel...?
Perhaps if Keil et al just concentrated on making their tools Eclipse-compatible, they could all stop re-inventing their own IDE wheel...? Not only their own IDE wheel, but how much effort have they wasted on making an inferior editor. Oh, if the toolmakers could agree on a common interface * Keil et al could free up resources to make the compilers even better * CodeWrights succesor et al could free up resources to make the editors even better * Eclispe et al could free up resources to make the IDE even better * SILabs et al could free up resources to make the debugger even better * the developers could have the same look and feel when cnanging platform Erik PS any rumors as to where the CodeWright developers went after Borland (those bastards) killed CodeWright.
"PS any rumors as to where the CodeWright developers went after Borland (those bastards) killed CodeWright." Presumably, Borland intended that they'd stay with Borland and put their skills to work on other Borland products - just like a certain well-known FPGA manufacturer did recently when they bought another company and immediately canned their product entire line...
Presumably, Borland intended Just like Borland intended when they killed "Brief", those guys escaped and made CodeWright and now got "caught" again. My guess is that once you have escaped, you will do it again. Erik