Search the Community
Showing results for tags 'best practices'.
Found 2 results
Edit: I'm asking primarily if there is a reason why/when I should/shouldn't use .net functions in my LabVIEW development. The example below is just to demonstrate a case of my question. _________________________________________________________________________________________________________________________________________________________ I recently found a custom library that calculates SHA256 hash algorithms However, in this post I see that the same thing can be done with .net My main question is: Is there any reason to build or use a custom library for something where .net functionality already exists? Are there any disadvantages to offloading work to .net ?
I've been using Labview for 6 years now for research and one of the most confusing things for me is how code is supposed to be managed. I have been using Tortoise SVN with folder organization but with things getting more complicated I am starting to wonder what the "right" way of doings things is. I've looked through a variety of posts and official documentation and haven't found what I am looking for (or haven't realized it was what I needed). I'm using LV2009 32 bit, JKI TotoiseSVN Tool 126.96.36.199 (demo), Tortoise SVN 1.7.6, Win 7 x64, and have little interest in actual code deployment. Features I am looking for: - avoiding cross linking problems - prompts to save a vi only when it has been changed - svn integration - code that works with multiple developers instead of corrupted files (mainly thinking about project files) My code base consists of project or modules (not lvproj) just a set of code in a directory & subdirectories. When I am running a program, many of these modules will run together and collect data, with a messaging system in between where necessary. In addition I have generic code libraries (not lvlib), with generic functions (math, string, table, server, etc). It seems like my code libraries should be lvlib files. After some reading it seems like Project Libraries can be used for preventing namespace collisions. On a side note the term Labview Project Libraries is very confusing because I think of Labview Project Explorer which I have no gathered does nothing for changing the namespace. I did happen to find this link, which helped get me started but I'm still not understanding some things. As a general rule I pretty much want library membership to always apply to files that are in my libraries directory, which doesn't seem possible. Question 1: How do I delete a library file via SVN and remove it from the project. I am using the JKI TSVN toolbox and removal of the file doesn't seem to effect removal from the library. Question 2: If I create a vi and later decide to move it into the library, how do I accomplish this in Labview with SVN tracking. For example, if I create a vi in one of my modules and realize that it is fairly generic and would be better in a library, how do I move it to the library (ideally on disk and into the library file) so that both Labview and SVN are happy. Another situation might be moving a file from one module to another module, ideally I could move both library association and disk location. Question 3: If I am running multiple modules, how do I ensure that they have no namespace collisions? Should these be libraries as well? Do I only need a project if I want to deploy my code? Question 4: When someone creates a vi outside of the library but in the library directory how can this be detected and fixed? Question 5: Is there any way to get better svn integration into the lvlib and lbproj right click menu? My current approach would be to do this through the tools menu (tools -> TortoiseSVN -> rename). Side note, this doesn't doesn't currently work on my computer although the actual TortoiseSVN through Windows Explorer works just fine. Question 6: My files are constantly prompting for saving. It seems like most of these are changes from other files and not the file I changed. Since I don't know really know how this works I am not sure what exactly is going on. The end result though is that I want to minimize version changes (as tracked through SVN). I seem to remember reading somewhere that in LV2010 you can separate source code and compiled code. Thoughts? Would this be fixed by better lvlib and lvproj usage or is something else going on? Thanks, Jim