i2dx

Finally, I have an Update for you: ADO-Toolkit V 1.6.3 I would be glad, If you give it a chance

QUOTE(Jim Kring @ Feb 25 2007, 12:34 AM) ACK and I'd like to have "automatic Member storage" - like in LV2 Globals ... and, and and ... or - to cut a long story short - I'd like to have C++/C# like LV-Objects, with a "NEW" operator (which implies by reference objects ...)

QUOTE(Tomi Maila @ Feb 20 2007, 11:52 PM) completely wrong - if it was so we would not have a mouth and a tounge, but cable reels and wire cutters (instead of hands)

QUOTE(James P. Martin @ Feb 17 2007, 09:33 AM) I don't think you need an "experienced pogrammer", you just have to have a look at the example finder, and use one of the "pulse train generation" examples http://lavag.org/old_files/monthly_02_2007/post-885-1171702212.png' target="_blank"> just if you have never seen the example finder: click "Help" --> "find examples"

QUOTE(yannis @ Feb 13 2007, 10:57 AM) the bad news first: you need some German to understand the description the good news: http://forum.ib-berger.com/index.php?showtopic=64' target="_blank">check this examle

I just wanted to bring that thread up again ... the problem still exists ...

the project will crash, independent of the number of VIs. I reproduced this bug with 4 VIs. IMHO and as far as I tracked that down, there is a bug in the updating mechanism, which updates all VIs after the edit, when you confirm the "replace ?" - dialog. This bug seems only to concern CONSTANTS, and does not occur if you use only one instance of that typedef in one VI The service Request Nr. is 396436 (Nov 23. 2006) - as I mentioned above, I forgot to care about that and could not find any responses about this topic. Maybe this one needs a warm up and a CAR# ?

I mentioned this behaviour some time ago here ... which reminds me on asking, what happend to my input to NI ...

hmm ... then let the other forum answer the question

thx are you shure? I have my "knowledge", because in one of my last RT projects with LV 7.1, I had to learn, that using the exit LabVIEW primitive in the RT main VI can lead to that "weird behavoiur" (the startup.exe was restarted in parts when connecting with MAX). I did not experience, that the exit LabVIEW primitive forced a controler reboot

yes, I am sure, I have done this several times, and it works. Maybe you should try to use fixed IP adresses e.g. 192.168.1.1 and 192.168.1.2 and no host names, which would require a DNS service (and that would be quite unusual on a PtP network ...)

I politely disagree: The Exit LabVIEW is NOT intended to be used on RT machines, therefore it is not included in the RT-View of the Appcontrol palette. Using the exit LabVIEW primitive on an RT machine is not recommended, because it can lead to weird behaviour. I have experienced that such VIs can restart occasionally, e.g. if you connect to the RT machine via MAX. Maybe it could reboot the machine, but thus is not intended. The LabVIEW Runtime Engine closes automatically when all VIs have stopped, so you don't need any mechanism to close the Runtime Engine explicitly. So the answer to the question could be: stop all your VIs (e.g. with occurences) if you detect the power down and exit your application "the normal way". If you don't have a normal way, it would be wise to create one

I have general concerns agains VISTA. I don't like this DRM b******t, threrefore I am planing to use XP als long as possible.

Hi Folks, I am currenty trying to improve my ADO Toolkit, but I have reached a point, where I can't find a solution for this specific problem: The variant input "Restrictions" of the Invoke Node "OpenSchema" needs a special input which is defined as follows: _bstr_t bstrTableType("Table"); varCriteria[0].vt = VT_EMPTY; varCriteria[1].vt = VT_EMPTY; varCriteria[2].vt = VT_EMPTY; varCriteria[3] = bstrTableType; VARIANT varData; SAFEARRAY FAR* psa; SAFEARRAYBOUND rgsabound[1]; rgsabound[0].lLbound = 0; rgsabound[0].cElements = 4; psa = SafeArrayCreate(VT_VARIANT, 1, rgsabound); // Fill the safe array. for( lIndex = 0; lIndex < 4;lIndex++) { hr = SafeArrayPutElement(psa, &lIndex,&varCriteria[lIndex]); } // Initialize variant with safearray. varData.vt = VT_VARIANT | VT_ARRAY; V_ARRAY(&varData) = psa; Con.CreateInstance(__uuidof(ADODB::Connection)); Con->Open(L"DSN=pubs", L"", L"", -1L); Rs1 = Con->OpenSchema(ADODB::adSchemaTables,varData); does anyone know, how to build that specific variant, which is passed to the function OpenSchema(ADODB::adSchemaTables,varData) ? Furthermore this problem does not only concern this example, but there are other cases, where I'd like to pass a *variant structure* (??) to an ActiveX Invoke node, e.g. if you use the Internet Explorer Object and invoke the print method, you need to pass a *special* varaiant to the method, too, in order to remove the default print headers (wich look pretty uggly ...) Maybe such variants can be created by calling *other* ActiveX Invoke or Property nodes - but which class should I use ? Any help would be realy appreciated!

so what? With CC you can clearly define, how your work can be used and which rights are reserved. Why should I care about Software License Compatiblity or that the license I use is certified by the Open Source Initiative?

sorry, this somewhat of a cross post, but what is wrong with the Creative Commons license? I spread all my *open source work* with this license ...

I saw it the first time, so I thought ...

ah, thx I considered allready that the reason for this bug is, that the graph tries to adapt to the properties of the waveform, which is *currently* empty. I stumbled on this bug, when I connected a WF Graph to the DAQmx VI "AI Read 1D Wfm" and tried to rename the channels ... whatever, It's not my job to think about the reasons of a bug, my job is to get the task finished anyway

whatever ... it's nice to see that NI has a "competitor". It's like linux it will never become more popular than windows until *we* change our way how and why we use software, but it had a nice impact on the development of windows and the guys from MS$ had to be a little bit more carefull with what they do. It's overconfident to think that a 3 man team can measure up with the NI R&D, just due to the sheer manpower they have and not to talk about 20 or more years experience in developing LabVIEW. So if there is a *product* in measurable time, it will predictably not compete with LabVIEW (V 12.x ??) nor will any LV professional consider seriously to use it in one of his/hers projects.

I am aware that I could drop in a brick with this advice. I don't know anything about Lookout, but I had a longer phone call with a colleague yesterday and he was complaining a lot about the quality of the DSC Module. It seems to be a Module which is not used very often, and therefore full of bugs, etc ... so my adavice would be: get a trial version of the DSC Module and try to implement the mission critical parts with DSC to see if you can reach your goal with this module.

confirmed: CAR# 45M2HEA4

Of course I do! I just wanted a confirmation from someone else that this behaviour is reproducable before I bother the NI support with this. I'll report it to NI tomorrow (it's 8 p.m. here and I'm the only donkey still working [except the other donkeys ]) and post the CAR# here ...

lol got da HH blues?

Hi Folks, I found a bug today in LV 8.20, when I was trying to rename a channel of a Waveform chart: please take the attatched VI, click into the plot legend and edit e.g. the name of the first channel. Click somewhere else to finish --> LV crashes ... Download File:post-885-1169473346.zip If you can reproduce, I will bother NI with that - if not Stephen cares about that