ShaunR

http://www.metacafe.com/watch/1154898/how_to_make_a_tinfoil_hat/ I wonder what colour the sky is on her planet

That's fantastic Just goes to show. "There are no bad programs, only bad programmers". I think I'll set that up as my wallpaper....move over Grace Park

Take a look at the speed example.

You do not need to explicitly open or create a file with the any of the high level API (the exception being "Query by ref") as it will open or create a file if one doesn't exist. Just specify the fie name. You cannot write directly to the file using standard file write functions. A SQLite file has a complex structure.

I don't think I'm that bad

In the "Dialog & User interface" palette the is a "Colorbox constant". You can wire that to it and choose a colour by clicking on it which will show the colour chooser dialogue.

Sweet

There's probably a better mathematical solution, but this works as a practical approximation for this sort of thing (certainly for repeatability at least). As Yair said. Take loads of readings. Then calculate the mean and variance of your data. Plug those value into the probably density function for a truncated normal distribution and solve of X (note that b in this case is infinity). It is flawed in that it assumes your data is normally distributed (which it isn't its 1/2 of a normal distribution). But it will give a much better approximation.

You could have just used the "Query.vi" and put in "SELECT DISTINCT Col1,Col2,Col3 FROM TableName;" then you wouldn't need to filter the results.

Yup. I had a similar experience. At my old firm IT insisted that the had to install everything. I gave them a list of Labview and all the packages I needed and went and played with some instruments. After an hour they called me back to say it was complete. Great Where are the toolkits/packages? Umm. how do you install them (they said). So I showed them. After a few hours of sitting watching the install and drinking copious amounts of coffee the IT uy said "OK I'm off home". I said "When you come in tomorrow you need to phone NI and get all the activation codes (license was in ITs name). See you tomorrow so we can activate them all. Oh and by the way, we need to do this all again forl 8.2, 7.1 and 6.0 Can we do that the day after?" Next morning I had local admin rights and they transferred the license into my name

SQLite API Version 1.2 released.

Indeed.You will still have plenty of time to learn Russian so you can read it

I hope your machine is networked to the internet. Mine isn't and after an install I have to hand type in 23 activation codes It's all got a bit ridiculous

SQLite shows a download size of 4.8 MB. But the latest upload is only 1.8 MB. It looks like its the sum of all the versions. Is that right? The download page is geared towards showing information about a particular version (file name, version number, page title etc). Shouldn't it only show the size and download speed of the latest version?

If you are using Version 1.1, then you can use the "Transaction Query.vi" as JCarmody referenced. If you are using version 1.2 (just uploaded so unlikely ) then you can use the "Insert Table.vi" or "Transaction Query.vi". Well. I cannot replicate your test because you haven't released the VIs. But I ran the speed test inserting 1,000,000 rows with version 1.2 and reset the PC. After 12 resets the DB was fine although Labview forgot all its palette settings on the 3rd reset . At that point I got bored since it takes about 1 minute for my PC to boot

I'd like to see a native "HTML" string control

SMS has been the greatest assault on the English language since, erm "American English" The number of Thai people that think "before" is actually spelt "b4" is amazing.

Yup. Still trying to get my head around all the compiler options (haven't even looked at linker stuff yet). Some compile, some don't. I've got them all installed I'm trying to compile against 10.4, 10.5, 10.6. Sometimes they won't compile at all depending on certain settings. I need to be able to figure out the minimum version since I won't have the luxury of just getting it to work on my system alone (looks like I'll have to install a few more Mac versions) Indeed. I'm in two minds about this. I would like to compile from source. But at the moment the debug is 5MB and a release is 2.4MB Xcode is not very intuitive and as soon as I figure out how and what it's doing..... will ditch it in favour of something else. In theory, we shouldn't need it at all since the Mac version of SQLite has all the scripts etc to build and install it.But I don't know what half the files are for (yet). It's a steep learning curve having never used a Mac before (just finding stuff takes me 1/2 an hour...lol). But getting there

I've used Inno Setup for a number of years now. Again its free. Very, very powerful and creates very compact intalls (typicaly 1/2 the size of msi). In it raw form its almost unusable, but with this or this GUI its very sraight forward.

That is an interesting article. But it is very Linux oriented (in particular XSF and changes between kernel versions). I don't know much about Linux file systems. But I do know NTFS does not suffer from this and the only currently supported OSs are windows. "fsync" and "write barriers" are not the same thing. The latter is a firmware feature and the former is an OS function. Admittedly, on some OSs (read Linux) you can turn it on or off. But you cannot under Windows (I don't think). That depends. If the programmers are mainly non-windows programmers,then they may not mention it since they use a "cover-all" statements. After all. If they were wanted NTFS to be "utra-robust" and were worried about ll this. then they could have used "CreateFile" with the "FILE_FLAG_NO_BUFFERING" and "FILE_FLAG_WRITE_THROUGH" options. then they would not need "FlushFileBuffers" and not have the overhead associated with it.. Your making a bit of a meal out of this. There are better ways to ensure data-integrity if the data is that important. Don't forget. I'm not "removing" the functionality. If you really feel strongly about it then you can just turn it on. Although I suspect you will never use the API since it is natural to prefer your own in-house solution.

Fantastic. Many, Many thanks. I really appreciate it. You examples didn't work straight out of the box. But I re-compiled the framework with i386 rather than x64 and labVew now accepts it and I can run your dll verson vi. I found another framework that Labview could load in the /development/frameworks. There are 3 examples located in there. 1 I can load. the others I cannot. After getting yours to work it it must be due to the bitness (although I'm not sure how to check). Thanks to your example, I've compiled a framework with sqlite 3.7.3 successfully (lots of warnings though) and your re-hashes of my low-level VIs seem fine. I have a rough idea how things fit together now and I think I should be able to compile a framework without the intermediary LV_SQlite wrapper. Looks like I'm going to be very unsociable tomorrow whilst I play with the Xcode settings I'll let you know how I get on later in the week. Muchos Grassy Arse

The order is pretty much dictated by the way the application is written and has little to do with the cache. A file is opened when you instruct the OS to open it. If you write an application as in your second example, then sure, you cannot guarantee the order because you cannot guarantee the order that Labview will execute those functions. Well. Assuming that corruption is going to happen (which I obviously don't agree with....at least on windows) then probably you are right. But it would be far worse to for the structure to be corrupted therefore not enabling retrieval of any information. Indeed. But as I said before. I believe the latter to only be relevant on non-journalled systems. Don't forget that SQLite is applicable for many OSs that do not have such a file system. I think there are very few instances (perhaps the savepoints or an in-memory database) where people will need/want to use them. Those that do would quickly see the way forward with an example or two. The "query by ref" is already in the "Query" polymorphic VI and I think fitting for that function.

well. ActiveX and dotNET are banned in my applications