2010-05-06T04:29:55 *** goffena has quit IRC 2010-05-06T04:29:59 *** goffena has joined #opticks 2010-05-06T05:34:02 *** dsulgrov has joined #opticks 2010-05-06T05:48:52 *** kstreith has joined #opticks 2010-05-06T05:48:56 *** ChanServ sets mode: +o kstreith 2010-05-06T05:55:17 *** tclarke has joined #opticks 2010-05-06T05:55:17 *** ChanServ sets mode: +o tclarke 2010-05-06T08:28:12 *** nate__ has joined #opticks 2010-05-06T08:29:57 hello nate 2010-05-06T08:30:07 good morning 2010-05-06T08:30:14 when I join, do you automatically see me or do I need to select you to talk to? 2010-05-06T08:30:36 everyone sees when someone joins a channel they are in 2010-05-06T08:30:44 if you select me, we would have a private conversation 2010-05-06T08:30:48 just you and I 2010-05-06T08:30:59 ok. so where should we begin 2010-05-06T08:31:06 ok, so it looked like your Windows SDK might not be configured properly 2010-05-06T08:31:09 so I wanted to compare with mine here 2010-05-06T08:31:20 alright 2010-05-06T08:31:31 Do you have "C:\Program Files\Microsoft SDKs\Windows\v6.0\VC\Bin", specifically cl.exe and link.exe in that directory? 2010-05-06T08:34:13 I have neither. I do have a lc.exe, but not cl.exe. link.exe is missing 2010-05-06T08:34:34 do you have 100+ entries in "C:\Program Files\Microsoft SDKs\Windows\v6.0\VC\INCLUDE"? 2010-05-06T08:35:46 and 50+ entries in "C:\Program Files\Microsoft SDKs\Windows\v6.0\VC\LIB"? 2010-05-06T08:36:02 I do have a cl.exe and link.exe in the same path, but extend it out to VC\Bin. These two files are located here 2010-05-06T08:36:22 so you have cl.exe and link.exe in VC\Bin? 2010-05-06T08:36:23 I have 159 files/folders in v6.0\Include. The folders count is 3 2010-05-06T08:38:06 whoops, yes I do have the cl.exe and link.exe above in the directory you mentioned. I was looking at v6.0\bin and not v6.0\vc\bin 2010-05-06T08:38:14 ok, good 2010-05-06T08:38:17 so far this is matching what I have 2010-05-06T08:38:22 ok, go to a command prompt 2010-05-06T08:38:26 lib fiolder has 55 files 2010-05-06T08:38:32 command prompt up 2010-05-06T08:38:46 and run the setenv.cmd with /xp /debug /x86 2010-05-06T08:42:09 and then type "echo %INCLUDE%" and paste that to chat 2010-05-06T08:42:12 ok. I ran it as described in your document. message came back with the VCDetect is unable to set the environment, then reported that it was setting SDK environment relative... 2010-05-06T08:42:23 then it said targeting windows xp x86 debug, then command prompt is back. Text turned yellow 2010-05-06T08:42:37 did you see my earlier message about "echo %INCLUDE%"? 2010-05-06T08:44:18 I would do the paste, however, my laptop (which I am running this from) is only able to hit a week wireless internet connection which tends to go down often. I am typing this from my work machine which is on a high speed internet 2010-05-06T08:44:35 ah, i see 2010-05-06T08:44:47 here is mine: 2010-05-06T08:44:56 C:\Program Files\Microsoft SDKs\Windows\v6.0\VC\Include;C:\Program Files\Microso ft SDKs\Windows\v6.0\VC\Include\Sys;C:\Program Files\Microsoft SDKs\Windows\v6.0 \Include;C:\Program Files\Microsoft SDKs\Windows\v6.0\Include\gl;!Include! 2010-05-06T08:45:33 yes, I have this 2010-05-06T08:45:38 ok, cool 2010-05-06T08:45:46 the Windows SDK appears to check out then 2010-05-06T08:45:52 so from the same command prompt 2010-05-06T08:45:55 set OPTICKS_CODE_DIR 2010-05-06T08:45:58 and set OPTICKSDEPENDENCIES 2010-05-06T08:48:26 change to directory where your plug-ins are and run scons 2010-05-06T08:50:40 ok I am in c:\temp\optickstest\nateradar. This is where my cpp, h, and modulemanager.cpp file present. Sconstruct is also present. I do have a build folder from yesterday and a .sconsign.dblite file. Should I delete these before running scons? 2010-05-06T08:50:49 leave for now 2010-05-06T08:50:54 just want to see what happens at this point 2010-05-06T08:51:12 came back with the message 'scons' is not recognized as an internal or external command... 2010-05-06T08:52:42 ok, so run this instead "C:\Python25\Scripts\scons" 2010-05-06T08:55:00 ok. That seemed to move this along. The prompt did come back that I have some errors. One refers to a fatal error in my Skewness1.cpp, line 27 about a Tutorial5.h file: no such file or directory. Probably a mishap on my part. 2010-05-06T08:55:25 Also I get scons: *** [Build\win32\release\plugins\skewness1.obj] Error 2 2010-05-06T08:55:29 and scons: building terminated because of errors. 2010-05-06T08:56:58 ah, good 2010-05-06T08:57:00 now we are compiling 2010-05-06T08:57:09 so now you just have to fix the compile errors 2010-05-06T08:57:21 ok. Sounds good. 2010-05-06T08:57:28 don't know if you time to fix those now 2010-05-06T08:58:39 if you do, we should be able to get it to compile 2010-05-06T08:58:44 and then I can show you how to run it with Opticks 2010-05-06T08:59:34 let me try here a few minutes..... 2010-05-06T09:10:40 ok. I took out the header file I had for Tutorial5 and now I am getting an error at the switchOnEncoding line. The response indicates it takes 6 arguments, but it looks like I have 8. 2010-05-06T09:10:52 I think I need to spend some more time on this for trouble shooting. I think it will take me a while to look over. 2010-05-06T09:13:33 A couple of questions: each time I wan to compile, do I need to go through the steps we went through today? can these be put in a .BAT file or a file to run so that one doesn't have to type this in all the time? 2010-05-06T09:13:37 funny you mention that 2010-05-06T09:13:42 i just sent you an e-mail with two such batch files 2010-05-06T09:13:46 one for a debug compile of your plug-ins 2010-05-06T09:13:48 one for a release compile 2010-05-06T09:13:59 I did put the c:\python25\script in my path, so should scons run as you initially indicated or use the full path to where scons is located? 2010-05-06T09:14:01 and I designed them so you can just double-click on them, no need to drop into a command prompt 2010-05-06T09:14:07 Thank you. 2010-05-06T09:15:16 you need to put "scripts" in your path not "script" 2010-05-06T09:15:24 if you do that you can use just "scons" without specifying the whole path 2010-05-06T09:15:52 I took he edge detection sample algorithm and modified it for my "skewness1 algorithm. I need to go back and check a few things I did and I will see if I can figure this part out. I will let you know where I get. 2010-05-06T09:15:59 sounds good 2010-05-06T09:16:00 Thank you for all of your help and time. I really appreciate it. 2010-05-06T09:16:09 no problem. thanks for staying the course 2010-05-06T09:19:22 Going through this on my own will be similar to my students if they decide to go down this road. I don't think I will be providing this too them just yet, but I do want to put together a document that they can refer to on their own for becoming familar with writing image procesing algorithms. My experience has been that if they have a series of steps to follow in a logical manner that it cuts down on problems with "building" a p 2010-05-06T09:19:31 problems related to a specific piece of code, method, algorithm, etc. 2010-05-06T09:21:11 yeah, i agree 2010-05-06T09:21:20 i think the "Scons" set-up as this point is meeting my goal 2010-05-06T09:21:23 For myself, I would like to build some of the algorithms I currently have so I can have them available in Opticks where my students would be users of the Opticks software and not developers (for the most part). I will likely have a few that are interested in programming. I would also like to offer a seminar type class or short course in remote sensing algorithm development. Opticks would be a good platform for this 2010-05-06T09:21:23 of reducing the total number of steps 2010-05-06T09:21:33 compared to Visual Studio 2010-05-06T09:21:47 that all sounds great and we can help with that 2010-05-06T09:22:02 I agree with you too for the command line version, especially if people aren't real familar with Visual studio and don't necessarily want to fork out the cost for the Pro version of VS 2010-05-06T09:23:16 ok. Thank you again. I will keep working on my end and let you know how it goes. 2010-05-06T09:25:03 *** nate__ has left #opticks 2010-05-06T09:33:56 *** nate__ has joined #opticks 2010-05-06T09:34:18 kip, here is a question. does the switchOnEncoding number of parameters need to match the algorithm parameters 2010-05-06T09:34:33 hold on, refreshing memory of switchOnEncoding 2010-05-06T09:34:43 such as void edgedetection(T*, DataAccessor pSrcAcc, int row, int col, int rowSize, int colSize) ? 2010-05-06T09:34:58 the swithOnEncoding line has 8 parameters where as the void statement above has 6. 2010-05-06T09:35:13 the switchOnEncoding call will have 1 more paramter than the actual function call 2010-05-06T09:35:20 so in your case, when you call switchOnEncoding, you should have 7 parameters 2010-05-06T09:35:25 the first is consumed by switchOnEncoding 2010-05-06T09:35:31 sorry, sorry 2010-05-06T09:35:32 an example: 2010-05-06T09:37:52 void foo(T*, bool arg1) 2010-05-06T09:38:01 switchOnEncoding(encoding, foo, True) 2010-05-06T09:40:13 can you send along your call to switchOnEncoding? 2010-05-06T09:40:19 ok, can you take a look at the edge detection example? I see two more arguments in the switchOnEncoding than in the function call for edge detection. 2010-05-06T09:40:34 Yes, I can... 2010-05-06T09:40:52 yeah, my earlier example was wrong 2010-05-06T09:42:03 void foo(T*, bool arg1); 2010-05-06T09:42:14 EncodingType type = INT1UBYTE 2010-05-06T09:42:18 template void Skewness1(T* pData, DataAccessor pSrcAcc, int row, int col, int rowSize, int colSize) 2010-05-06T09:42:20 switchOnEncoding(type, foo, NULL, true) 2010-05-06T09:42:30 switchOnEncoding(pDesc->getDataType(), edgeDetection, pDestAcc->getColumn(), pSrcAcc, row, col, pDesc->getRowCount(), pDesc->getColumnCount()); 2010-05-06T09:42:37 ah, i see the problem 2010-05-06T09:42:44 the second arg to switchOnEncoding is the name of the function to call 2010-05-06T09:42:50 you need to change that from "edgeDetection" to "Skewness1" 2010-05-06T09:44:40 yes. I did this. sorry, I copied my code from a different cpp file. I made this change. Basicaly, this call is exactly the same as the edgedetection algorithm. The only thing I do is calculate some other stats based on the window size (similar to the edge detection). 2010-05-06T09:45:04 the rest looks ok 2010-05-06T09:45:17 I get a response back when I compile that the number of arguments needs to be 6. specifically it states skewness1::skewness1: no overload function takes 6 arguments 2010-05-06T09:45:45 I thought my problem might be an incorrect number of arguments problem. I guess there could be a problem within another part of my code. 2010-05-06T09:45:48 thinking... 2010-05-06T09:50:13 ah, i see the problem 2010-05-06T09:50:23 went back and looked at the .zip you sent me of this some time back 2010-05-06T09:50:29 your class is also called Skewness1 2010-05-06T09:50:34 so you have a class and function with the same name 2010-05-06T09:50:38 I just sent you the code again 2010-05-06T09:50:50 and compiler is choosing class, but you don't have constructor that takes 6 args 2010-05-06T09:51:00 so I guess I need to not name my class and function the same.... 2010-05-06T09:51:05 yep 2010-05-06T09:51:11 ok. I will make these changes and see what happens. Thanks, 2010-05-06T09:52:22 np 2010-05-06T09:52:31 C++ provides such great error messages ;-) 2010-05-06T09:59:03 ok..I just made these changes and a mistyped variable (I thought Visual Studio made case changes automatically).... 2010-05-06T10:01:05 I re-ran the scons and it brought back a prompt without errors. It did go through a create library process. I do have a number of files like ..dll, ..exp, ..lib, ..pdb files, but there is no "file name" before the ".." 2010-05-06T10:01:13 did I compile a script? 2010-05-06T10:01:28 do you have a "Build" folder? 2010-05-06T10:01:40 same directory as SConstruct is in 2010-05-06T10:01:44 build folder has two other folders Binaries-win32-release and win32 2010-05-06T10:01:50 *** gmartin_cn has joined #opticks 2010-05-06T10:01:51 what's in Binaries folder? 2010-05-06T10:01:55 binaries has pdbs and plugins folders 2010-05-06T10:02:18 in either of these are the ..pdb, ..dll, etc, but with no root name before the .. 2010-05-06T10:05:08 *** rforehan has joined #opticks 2010-05-06T10:05:08 ok, is your Skewness.cpp, .h and ModuleManager.cpp in same directory as SConstruct? 2010-05-06T10:05:20 binaries folder has pdbs folder with one file ..pdb (~1.5 Mb); the plugins folder contains ..dll, llexp, ..lib, ..pdb 2010-05-06T10:05:31 sconstruct and the skewness files are in the same folder 2010-05-06T10:05:37 ah, that is why they are called "..dll" 2010-05-06T10:05:46 create a subdirectory in that directory, whatever you want your dll to be named 2010-05-06T10:05:56 move Skewness.cpp, .h and ModuleManager.cpp into the newly created subdirectory, note move and not copy 2010-05-06T10:06:00 delete your build directory 2010-05-06T10:06:03 re-run SCons 2010-05-06T10:09:13 ok. I called a subdirectory skew, re-ran, I now have within the Build folder\binaries...\Plugins, skew.dll, skew.exp, skew.lib, skew.pdb 2010-05-06T10:09:19 pdbs folder contains skew.pdb 2010-05-06T10:09:21 ok, skew.dll is an Opticks plug-in 2010-05-06T10:09:33 find your 4.3.3 Opticks install 2010-05-06T10:09:37 my guess C:\Program Files\Opticks\4.3.3 2010-05-06T10:09:42 go the PlugIns folder 2010-05-06T10:09:48 win32 contains the modulemanager, the skew dll, manifest, exp, lib, map, and obj files 2010-05-06T10:09:49 *** gmartin_cn has quit IRC 2010-05-06T10:10:05 and copy skew.dll into that directory (note this works because you built a release version of your plug-in and and Opticks install is a release build, if you had built a debug version of your plug-in and put it here it still wouldn't work) 2010-05-06T10:10:12 start up Opticks 2010-05-06T10:10:16 go to "PlugIns" tab of Session Explorer Window 2010-05-06T10:10:20 scroll down until you find "Skew" 2010-05-06T10:10:33 if that works, your plug-in is loading 2010-05-06T10:13:42 yes. Looks like this is good. skew.dll shows up in the plug-ins area and I have a RADAR menu option with the skewness 1 option 2010-05-06T10:13:49 load an image 2010-05-06T10:13:51 I wil try and test this out on an image to see if it actually does anything 2010-05-06T10:13:56 just what I was going to say 2010-05-06T10:20:29 ok. well it did something. It looks like it processed the image (in my case a small 1 band radar image). Since I used a small window size, the result is sort of what I expect. I think I also need to look at my bit types (the input may be a signed 16-bit image) and my output pixel values are likely pretty small, so the scaling is kind of off, but the good thing is it processed an image and produced a result that seems to be rig 2010-05-06T10:22:03 good to hear 2010-05-06T10:23:55 thank you all for your help. this has been a long time coming for me. I will document some of the steps I have gone through and forward them back to you so you can use them if you want. 2010-05-06T10:23:57 Thank you agin. 2010-05-06T10:24:07 much appreciated 2010-05-06T10:28:05 *** nate__ has left #opticks 2010-05-06T10:41:59 New news from nightlybuilds: Opticks4.4Dev-Nightly-Solaris #16 (SUCCESS) 2010-05-06T11:26:50 *** gmartin_cn has joined #opticks 2010-05-06T11:43:23 *** gmartin_cn has quit IRC 2010-05-06T11:43:31 *** gmartin_cn has joined #opticks 2010-05-06T12:14:33 New news from nightlybuilds: Opticks4.4Dev-Nightly-Windows #15 (SUCCESS) 2010-05-06T13:00:52 *** epifanio has joined #opticks 2010-05-06T14:18:15 *** dsulgrov has quit IRC 2010-05-06T15:05:01 *** gmartin_cn_ has joined #opticks 2010-05-06T15:08:43 *** gmartin_cn has quit IRC 2010-05-06T15:08:43 *** gmartin_cn_ is now known as gmartin_cn 2010-05-06T15:45:15 *** epifanio has quit IRC 2010-05-06T16:09:24 *** tclarke has quit IRC 2010-05-06T16:48:54 New news from opticksannounce: Spectral Processing Extension 1.0.0rc1 Released || Opticks 4.3.3rc1 Released || Python Scripting Extension 1.0.0rc1 Released || IDL Scripting Extension 1.0.2rc1 Released || Win a PS3 or Wii in the Opticks Programming Contest! || Spectral Processing Extension 1.0.0 Released || Python Scripting Extension 1.0.0 Released || Opticks 4.3.3 Released || IDL Scripting Extension 1.0.2 Released || Spectral Processing Extension 1.0.0rc1 Released || Python Scripting Extension 1.0.0rc1 Released || IDL Scripting Extension 1.0.2rc1 Released || Opticks 4.3.3rc1 Released || New Production Extras Release - Extras 1.0.1 || New Production Release - Opticks 4.3.2 || New Release - Opticks 4.3.2rc2 - second release candidate for 4.3.2 || Opticks Programming Contest - Help Get the Word Out || New Extras Release - Extras 1.0.1rc1 - first release candidate for 1.0.1 || New Release - Opticks 4.3.2rc1 - first release candidate for 4.3.2 || Win a Wii in the Opticks Programming Contest! || Video Tour - 4.3.1 New Features || New Production Release - Opticks 4.3.1 || New Release - Opticks 4.3.1rc2 - second release candidate for 4.3.1 || New Release - Opticks 4.3.1rc1 - first release candidate for 4.3.1 2010-05-06T17:19:22 New news from opticksannounce: Spectral Processing Extension 1.0.0rc1 Released || Opticks 4.3.3rc1 Released || Python Scripting Extension 1.0.0rc1 Released || IDL Scripting Extension 1.0.2rc1 Released || Win a PS3 or Wii in the Opticks Programming Contest! || Spectral Processing Extension 1.0.0 Released || Python Scripting Extension 1.0.0 Released || Opticks 4.3.3 Released || IDL Scripting Extension 1.0.2 Released || Spectral Processing Extension 1.0.0rc1 Released || Python Scripting Extension 1.0.0rc1 Released || IDL Scripting Extension 1.0.2rc1 Released || Opticks 4.3.3rc1 Released || New Production Extras Release - Extras 1.0.1 || New Production Release - Opticks 4.3.2 || New Release - Opticks 4.3.2rc2 - second release candidate for 4.3.2 || Opticks Programming Contest - Help Get the Word Out || New Extras Release - Extras 1.0.1rc1 - first release candidate for 1.0.1 || New Release - Opticks 4.3.2rc1 - first release candidate for 4.3.2 || Win a Wii in the Opticks Programming Contest! || Video Tour - 4.3.1 New Features || New Production Release - Opticks 4.3.1 || New Release - Opticks 4.3.1rc2 - second release candidate for 4.3.1 || New Release - Opticks 4.3.1rc1 - first release candidate for 4.3.1 || New Release - Opticks 4.3.1Unofficial for Linux - first alpha release for Linux || New Production Extras Release - Extras 1.0.0 || New Production Release - Opticks 4.3.0 || New Extras Release - Extras 1.0.0rc1 - first release candidate for 1.0.0 || Announcing Opticks Extras || New Release - Opticks 4.3.0rc1 - first release candidate for 4.3.0 || New Production Release - Opticks 4.2.5 || New Release - Opticks 4.2.5rc1 - first release candidate for 4.2.5 || New Production Release - Opticks 4.2.4 || New Release - Opticks 4.2.4rc1 - first release candidate for 4.2.4 2010-05-06T18:10:54 *** rforehan has left #opticks 2010-05-06T19:24:18 *** kstreith has quit IRC 2010-05-06T23:33:02 *** gmartin_cn has quit IRC