I'm trying to do some basic spreadsheet tasks within OpenOffice for which the standard functions won't suffice, so I'd like to write my own macro in BASIC.
However, my function doesn't work. How do I go about debugging it? I tried inserting MsgBox("something") throughout my code but this has no effect (no message box appears when I call the function from Calc.
I also tried placing breakpoints throughout the function, which Calc simply ignores.
How can I debug my function?
The debugging guides seem to be more about subroutines than functions. I noticed that many people have commented that python would be a better option. Is that because the Basic implementation in OpenOffice is poor? Am I encoutering IDE bugs or are these features?
I can't reproduce this behavior. If I have a Function created in the IDE then I can use all what you say you can't: MsgBox and the observer. For using the observer of course the IDE has to remain open. See screenshot:
I have called the function as =MYFUNC(23) in a Calc cell, then I have fetched the IDE window to front and have run the code step by step.
Whether and how python would be a better option, you should ask the people which comment this would be so much better ;-). It is not my opinion.
Little introduction how to use python with openoffice:
https://wiki.openoffice.org/wiki/Extensions_development_python
Related
What are the best practices for developing with the Eclipse Pydev interactive console once code migrates inside a function?
I heavily use the interactive console as my projects are typically experimental, open-ended, and not spec-driven. The interactive console works well for code in the main loop, but as soon as I migrate that code to a function, I only know to use breakpoints with the debug view. However, that seems unwieldy, and I don't need the entire stack, nor error-catching mechanisms.
Is there a way to have the vanilla pydev environment interactive console inside a function, for example on the first breakpoint, or raised error?
Without this, I typically have to copy the function back into the main loop. Then, I end up using the same variable names, which leads to occasional global variable name accidents when I copy back to a function.
Thanks in advance.
I've using sublime for C coding for a while. With the help of cscope I can easily find the function/symbol calling chain (function/symbol called by which function) and jump freely. But recently I am reading perl code and cannot find a easy way to jump to the source of calling function. Don't want to search the whole project to find each time.
Is there any plugin to help display perl function calling relation?
Or is there any open source editors on mac to accomplish this?
Thanks!
I have worked shortly with Netbeans using Java and use eclipse for school projects, where I noticed there are some really cool features like it will show built-in functions on ctrl-space or it will suggest from existing variables. I recently started using matlab, I was wondering if there are any such tools available in the matlab. I know it's a scripting environment so things are probably a little different but then I wonder people who work with extensive projects with matlab, how would they manage the codes or cross-develop without such tools. I searched mathworks but could not find any useful information on that.
Anyone knows if such tools or add-ons existe for matlab? Thanks in advance guys!
As mentioned before, autocomplete works for functions, variables and fields.
If it seems to fail to work for a field (perhaps because your workspace is not properly set up) it can help to call the field without index:
For example:
entity.field instead of entity(i).field
Also if you change the workspace frequently it may be that the mfile editor does not catch up, in that case a quick workaround is to just type in the command window and copy afterwards.
As mentioned before the variables need to be in workspace.
I often approach so that I evaluate the code as I write it like this:
Write the code in cell mode
Evaluate cell with [ctrl - enter] - or button in menu bar.
Write next cell (variables of previous blocks are now available in workplace, autocomplete with tab)
There is also a matlab mode for emacs. It offers some autocomplete functionality and has a nice feature to wrap long code lines.
unfortunately on windows it lacks the ability to evaluate cells. So I rarely use it.
There are some similar design in MATLAB, one of them is called function hints.
You could find the documentation for it in here: http://www.mathworks.com/help/matlab/matlab_env/check-syntax-as-you-type.html
I'm trying to develop a small plugin that will do a sort-of auto-completion along with some other advanced features in order to create a primitive IDE to use with a custom scripting language we've developed.
So I want to know, how do the auto-complete plugins usually work? I have a basic plugin template that I'm playing around with (the C# one) and I see how the commands work, from a high level anyway, but I'm trying to figure out how I would create my auto-complete feature.
My first guess would be to make a command that spawned a new thread that retrieved the entire contents of the notepad++ text every 100 mills or so and then popped-up a little selector box or directly wrote the auto-complete possibilities when the correct pattern was matched on the newly typed text.
Any wisdom from those who have gone before me on this?
Thanks alot.
Well I figured it out. Apparently there is a "beNotified" method in my project template that I downloaded and it handles all events/notifications from the Scintilla/Notepad++ environment. I will probably be using the SCN_CHARADDED event to check the current line of text each time a character is added to the GUI.
I have a script that runs several little programs I don't have the source code to, one of which requires filling out some fields in a GUI. I've been doing this by hand, but I'd like to have the Perl script focus the GUI window, then enter some hardcoded text into each field and close the window.
In the past I've used the Win32::GuiTest module for tasks like this. Have a look at the example scripts and documentation.
Unless there's a specific reason you're using Perl, Autoit is a Basic-like scripting language pretty much specifically designed for automating GUI tasks. It's very easy to pick up.
You can have it wait for a window to become active, send any sequence of keystrokes or tell it to specifically focus on a GUI element with a given handle, and much more. You can also package the scripts as standalone executables, which can be a nice benefit when you need it.
You might want to look into Win32::OLE. There's also a book on this topic.