Why CDT does not save the refactor history for rename of functions and variables ?
(However JDT will save all the rename refactoring done in the source code)
Is it a limitation for CDT Refactoring plugin ? How to save rename of functions and variables to Refactoring History ?
If it is not inbuilt in CDT refactoring plugin , is it possible to develop a custom plugin for saving the Refactor->Rename of variables/functions ?
Related
We are using jBASE 5.6 multivalue database. Is there a way to disable the fact that AccuTerm's WED Editor has the ability to compile programs in the underlying MultiValue database? We are running into an issue where the user is accidentally compiling records instead of programs, which creates corrupt records in the datafiles.
I believe there is also a shortcut key (Function Key) that is in use. Help!
As of AccuTerm Release 7.3a there is no explicit option to disable the compile shortcut in the AccuTerm wED (Windows Editor). There are two techniques you can apply, however, that will work:
Change the Compile command to a single space
Open the wED
Click on the Tools menu at the top
Click on Settings
Click on the Tools tab
In the Compile section, change the text in the Command input from BASIC to a single space. Note that some flavors of PICK use BASIC as the compilation command whereas others use COMPILE. Since you are running jBASE, BASIC applies in this instance but for others such as D3 the command to replace is COMPILE.
It is possible to write a wrapper program to handle compilation for you, which when cataloged could be called from this same Command input. That program could be written to decide whether or not the target record exists in a code file or a data file and then compile or skip, respectively. Due to the fact that PICK systems do not have a file type specification, this logic would be somewhat unique to your environment depending on your practices. Common practice is to denote code files as BP files, i.e. FileNameBP or FileName.BP but again it would depend on your implementation.
For the record, a request to have a more formal option for disabling the Compile command has been submitted to the AccuTerm development team for consideration in the next major release.
I'm using Visual Studio 2010 for a C/C++ project. As far as version numbers go I like to have a revision and build date; for example "Project 1.0 R2 Apr 21 2013". I display this at startup, so I can easily tell if someone is running an old version. Now I'd like to put things like this in the filename of the executable created by the build. For example, "Project10R2.exe".
I make use of the build macros as listed here. For display of the build date, I use the predefined macro as listed here. To clarify possible confusion, the build macros are usable from, say, the project properties and what they refer to as "predefined macros" are #define's. I know you can define custom custom build macros (see this) (and obviously any #define I desire).
Now what I want is to use one with the other. That is, I'd like to define a revision string in one place and have it appear both in the program output at startup (easy with a #define) and also in the build output filename (easy with a custom $ macro). I don't want to maintain two different constants.
Anyone know how to do this? It appears you cannot even put the build date in the filename (not safely, anyway, if you use the Windows %DATE% environment variable, you may end up with illegal characters.)
As you've said you can create a custom build macro and then in your project's Properties > C/C++ > Preprocessor > Preprocessor Definitions, you can add something like this: $(my_custom_build_macro);... but this would define the macro in every file.
I am currently trying to merge some changes done in VB6 code with Visual Studio 2012.
The merge tool automagically removes "Set" in front of assignments, "Variant" is changed to "Object" and calls to "Sub" methods automagically gets parenthesises "()" inserted around the arguments. This makes VB6 unable to compile the result if i check it in!
It seems that only rows that is changed or in conflict is affected.
A merge tool should not employ magical things like that! As it is now i can't use VS2012 to merge changes we make in our old VB6 code.
Please, is there any way to dissable this functionality?
Thanks!
The code reformatting is being down implicitly by the IDE as it thinks it is VB.Net code where those changes are required.
You can either turn off the auto formatting options for VB.Net code, or use another merge tool like WinMerge.
Buliding the C# solution or Build automation script ..If its for Building the solution why it has to use MSbuild for it.. And clean will also be scripted in it intially
Psake is generally a steer away from XML based build scripts.
Since Psake is a PowerShell module and Powershell is a scripting language built on top of .NET, you have access to the entire .NET framework to easily be able to write custom build scripts.
Psake calls MSBuild to build .NET projects because projects are .proj files (or language variant .proj files) and contain all of the project data and metadata in a structure that MSBuild engine understands. You could certainly manage calling the compiler and do all the leg work yourself but that would be reinventing the wheel.
Psake's true power comes in being able to easily extend the build script using a fast and powerful programming language.
So this is a question for anyone who has had to integrate the building/compilation of legacy projects/code in a Team Build/MSBuild environment - specifically, Visual Basic 6 applications/projects.
Outside of writing a custom build Task (which I am not against) does anyone have any suggestions on how best to integrate compilation and versioning of legacy VB6 projects into MSBuild builds?
I'm aware of the FreeToDev msbuild tasks at CodePlex but they've been withdrawn at the moment.
Ideally I'm looking to version and compile the code as well as capture the compilation output (especially errors) for the msbuild log.
I've seen advice on encapsulating this functionality in a custom task, but really wondered if anyone has tried another solution (aside from executing shell commands) -
In essence, does anyone have a "cleaner" solution?
Ideally, executing commands using would be a last resort..
The VB6 task will be back on Monday. With regards to versioning, there is no explicit vb versioning task in the pack, however you could make use of the TfsVersion (TaskAction="GetVersion") and the File (TaskAction="Replace") tasks. If you think there is value in creating a new task to encapsulate / provide other functions, then please let me know and I will add it to the pack for the benefit of the whole community.
Apologies for the withdrawal, but come Monday I'm sure all will understand.
I am using Nant to build VB6 projects daily. This does resort using the Nant execute command to do the builds (we build 4 projects as part of one "solution").
It also allows you to label versions in your source control repository, get latest code, check in, check out, all the normal requirements, compile the update/setup programs copy the files to required locations and send emails of the results.
The logged results are fairly minimal though as you only get the output provided by a VB6 command line compile.
For versioning, I had to write a small app to extract the version number of my compiled executable and write it to a text file that Nant could then read and use (for labels, file names etc. (A bit of a pain but VB generated version numbers don't comply anyway).
For help with other non-core tasks see NAntContrib - from the NAnt link above.