Rocket MultiValue

Has anyone successfully been able to transfer binary files from server to client through SBClient (UniData)?It appeared to be possible to do a PD.LO using B and/or X options, and I tried also via UniBasic with OSREAD, but it does not work.  I opened a recent Rocket support case on this and the official answer was, despite some initial success from Rocket's side, ultimately it isn't supported (SBXA-18778).  Despite that news, I'm interested if anyone here has managed to get it working with (any version of) SBClient with some combination of luck and skill?  I have heard rumors it is possible if the file is uploaded through the client to the server in the first place, but that is not the case here.  The use case is, small Excel files are generated for reporting on the server.  The users would like to download these to their desktops.  I could arrange alternate delivery, but they prefer saving on their PC.  They are used to downloading .CSV files, which are text and transfer through SBClie

I have been testing D3 V10.3.2 on a test server before moving on from D3 V9.2.2.All programs must be recompiled when moving to the new version. I have found an issue with the BASIC EXECUTE statement whereby programs using the old style of ,//IN.< would not compile, bombing out with an error instead. Changing the code to the new(ish) style fixed the problem. I have notified Rocket of the issue via our support provider and hopefully, there'll be a fix in D3 V10.3.4.------------------------------Walter KiessManager ITPolice Superannuation BoardAdelaide SA Australia------------------------------

Hi ,Using in-basic python I interface function like var=pyCallFuntion(module,function)  how can I define if the var is a uniBASIC Var or a PYOBJECT ? Then, is there a methods from UNI BASIC to perform python str(var). Any help is welcome.------------------------------Manu Fernandes------------------------------

what is the best strategy to backup universe database?we plan to use uvbackup tool. Will this impact performance?------------------------------Abdus ThommilTech ArchitectFlextronics International Management Services LtdSan Jose CA United States------------------------------

I'll claim a senior moment here... An associate is trying to write an I-type dictionary that retrieves the first subvalue in a set of  fields. I can certainly do it with a subroutine and the FIELD function, but the FIELDS function specifically excludes use of the value mark or sub-value mark as delimiters. Suggestons? I'm sure it will be a forehead slapper when presented...The existing data is:ABC@SVMDEF@VMGHI@SVMZBA@VMQARHe wants it to return:ABC@VMGHI@VMQARThanks, in advance!------------------------------Jeff TeterWoodforest National BankThe Woodlands, TX------------------------------

Just wondering whether there should be an 'SB/XA' tag for this forum?#SB/XA​​

Does anyone use SSRS (Sql Server Reporting Services) to report D3 data.We've tried using the D3odbc and SQL mappings on D3 files, and it works very easily using Excel, but SSRS cannot see the D3 data source.This needs to be usable at an end-user level, so no program writing involved. MVC Toolkit and OpenDB seem to involve programming and some expense.We would like to know how anyone else is doing this.------------------------------Graham FallwellOwnerSoftware Applications Ltd------------------------------

When viewing an U2 BASIC program within VS code, Universe Basic @ variables such as @WHO are underlined with '~' characters.  According to VS code, these variables are undefined​​​.  I would have expected these not to show as undefined.  Can this be corrected?------------------------------Steve GleasonSenior Programmer AnalystGuitar Center------------------------------

We have recently developed the code lens feature in vs code extension and would like to share it to you.This feature works for all basic definition statements, including: variable declarations, function declarations, equate declarations and label declarations. It also displays the number of references on these statements. Clicking on the value of a reference will display the specific reference information. See the following example for details. Also, if you want the feature to prompt other information, or if you have other ideas or comments, please share with us.Thanks!------------------------------Jingwen LiRocket------------------------------

Hi All, Recently we are investigating that how to implement BASIC compilation in VS Code extension. I want to share our ideas, and we are very appreciate if you can provide suggestions and ideas. (The following content is just our initial ideas and design prototypes.) The main use of compilation is to find out the errors or warnings after code change. User can choose to sync the source code to U2 server or not. If yes, source code and objective code on sever would be overwritten after compilation. Before using compilation, user needs connect to a U2 server. And we will provide 2 methods to compile BASIC programs. 1. Compile BASIC programs through VS Code task. User can run VS Code default build task to start compilation. We can provide a target files list in the task configuration file, so that multiple BASIC program files in current account folder could be built together. 2. Compile modified BASIC programs.User can right click a file in the editor to display context menu and select "

Hi All, Recently we are investigating that how to implement BASIC compilation in VS Code extension. I want to share our ideas, and we are very appreciate if you can provide suggestions and ideas. (The following content is just our initial ideas and design prototypes.) The main use of compilation is to find out the errors or warnings after code change. User can choose to sync the source code to U2 server or not. If yes, source code and objective code on sever would be overwritten after compilation. Before using compilation, user needs connect to a U2 server. And we will provide 2 methods to compile BASIC programs. 1. Compile BASIC programs through VS Code task. User can run VS Code default build task to start compilation. We can provide a target files list in the task configuration file, so that multiple BASIC program files in current account folder could be built together. 2. Compile modified BASIC programs.User can right click a file in the editor to display context menu and select "

Hi,I'm fairly new to MVIS and I'm planning to enable Okta based authentication for all incoming request to my UV (v11.3.1) . This authentication method will be a token based authentication where a user will submit "client id" and "client secret" to generate a token (valid for a day) , which will be used to get the details from my UV files.I can see an option as "HTTP" authentication in MVIS with option as "OAuth"  but it only allows a filepath to be entered, which may not work with dynamic tokens.Has someone implemented a similar Okta based Authentication with MVIS and can share some thoughts on how to achieve token based authentication.Thanks,Amarendra

Hi All Assuming I have appropriate permissions on the file system can anyone tell me if it is possible to copy/move from a UNIBASIC program?I have explored READBLK and WRITEBLK but the file always seems to be corrupt after writing.Example scenario; move/copy a word doc or PDF file from one directory to another.Alternatively is there a way to shell execute a command to the local server or execute a powershell command etc.Thanks In AdvanceAaron

Hi Can you backup only the data portion of a file with uvbackup?Regards,Thando

hi vscode users,I'll share a strange behavior with rocket-mvbasic 1.1.0, where the java runtime goes cpu high and really slow.The context of runtime is special - totaly isolated.On a client site, we have a  server dedicated to development and running windowsTerminal sessions where we develop with vscode / rocket-mvbasic extension. The server and our RDP sessions are not authorized to internetWe are not authorized to extract the code outside the client's server. When coding I note a high usage of cpu (17% ) by java running rocket-mvbasic. I note the extension respond really slowly - not at all, colorization is 'late' , code formatting is delayed 30sec to repond ... I'll join a  dump of the java process. windows dump java process (shared by onedrive)here the vscode/extension log vscode log (shared onedrive)Is anyone who share these bad experience ?

I've been struggling with this for some time now (years, that is): The precision of calculations performed in D3.As everyone knows, D3 performs calculations to a default of 4 decimal places. It then truncates the result so everything past the 4th decimal place is removed. This in itself is a cause of concern as it introduces the 1st of several calculation inaccuracies. From my understanding, this occurs for every calculation performed, whether it's in a single expression or in separate statements. Thus there may be several inaccuracies in a single expression, particularly when fractions or division is involved.To get around this, I have arranged my expressions such that division is always performed last (where possible). But this doesn't solve the problem completely though. There is always an intrinsic error in the calculation/expression due to the truncation as the expression is not calculated as a floating-point expression. When you're dealing with hundreds of thousands or even milli

While PickBASIC object code ( pseudocode, not flash object ) is forward-compatible ( later D3 versions ) without recompilation, it is not completely backward-compatible ( earlier D3 versions ) because of op codes that have been added in later versions of D3. Example: Code compiled on 10.2 will run on 10.3, but code compiled on 10.3 may not run on 10.2.I didn't know this until just a couple of days ago. For years I've helped people upgrade D3 versions knowing that they don't need to recompile their source, especially when the source has been stripped and only object code deployed. Turns out I never helped anyone go the other way with stripped source. This came to my attention when a friend of mine in the UK upgraded his dev box to 10.3.3 and tried his normal deployment mechanism to a 10.2 box. Blew sky high.  This had never happened to him before.Turns out that 10.3 contains a LOT of op code changes to facilitate the mvBase to D3 upgrade plan as well as some other optimization for perfo

UniData 8.2.2Hi All,Does anyone know if Rocket have recently changed the key length for the SAVEDLISTS file? On older versions I had no issues saving keys like -TRAWL*LAZ.PROGS*197609I know about the potential 000-999 issue when the OS tags the chunk numbers onto the end of the key, but I could at least save the list. Now, the response I get from the system is - can not open file: 'SAVEDLISTSTRAWL*LAZ.PROGS*197609I tried tagging some alphas onto the end of the key to get around the numeric chunking clash, but it didn't make any difference. If I hardcode the list key to 'BOB', it saves the list as BOB000 just fine.I don't really want to reduce the length of the key because I want to keep it unique and also visibly descriptive. ThanksLeon

I have several servers with UV 11.3.1.6022 installed and, on one, I'm getting a very odd behavior on selects. This is only happening on our development server, but it's in multiple accounts and the configuration is the same as other servers.If I create a list with only one item in it and then do a "GET-LIST list_name". The system shows zero records selected and will not list the item. It behaves the same if I'm using &SAVEDLISTS& or another file for storing the lists. >CT MLISTS WNB710_20180101.WBP_TEST       WNB710_20180101.WBP_TEST 0001 TEST.FNCODESGET >GET-LIST MLISTS WNB710_20180101.WBP_TEST Zero records SELECTed.  SELECT list number 0 cleared. If I create a list with multiple items and then do a "GET-LIST list_name", the selected count will consistently show one less item than is in the list. However, if I do the "GET-LIST" and then list the file for the items, all are shown... >CT MLISTS WNB710_20180101.BLBP       WNB710_20180101.BLBP 0001 FNCODESGET 0002 FNCODES

@Leon Pearce and @Kathleene HunterJust to clear something up here on the question on the SAVEDLIST length question thread. The SAVEDLISTS error with a '*' will only happen on UniData on Windows, in fact there are several characters that Windows forbids in file names so it's not just a '*'. On UniVerse on Windows the disallowed characters are encoded so the list can be saved, Kathleene if you check in your SAVEDLISTS directory on Windows you will find where you had an '*' in your name it's encoded as %A, so this is why you don't see the error on UniVerse. UniData does not encode the disallowed characters.​​​------------------------------Jonathan SmithUniData ATS and Principal EngineerRocket Software------------------------------

Hi Everyone,I'm wondering if there is a built-in way to determine when a BASIC program was last executed on a UniVerse system.  (I'm running v11.1.9 on Linux)  I worked on a different platform years ago that had a DICT item in the BP file that would return that info, but I haven't found anything like it on UniVerse.Thx

Continuing on my search for more information about some of the uvconfig settings and their possible real-world tuning outcomes...The OPTMEM setting reads as an important option that could contribute to the performance of all RetrieVe commands.The uvconfig section reads as:# OPTMEM - This is the amount of memory allocated# for the query optimizer's workspace. It is# specified in 1K units (1024 bytes).OPTMEM 64Has anyone ever changed this setting?If so, what setting was used and what impact does/did it have?Thanks------------------------------Gregor ScottSoftware ArchitectPentana SolutionsMount Waverley VIC Australia------------------------------

I have occasionally wondered what the SCRMIN, SCRMAX and SCRSIZE settings in the uvconfig file actually influence.I know I have never adjusted these values since we started with UV 9 (currently working on our UV 11.3.3 in-place upgrades) so I was wondering if anyone has changed these settings, and if so then to what, and why?From the uvconfig file:# SCRMIN - specifies the minimum size of# UniVerse scratch buffer pool. This# value must be at least 1 and must be# specified before SCRMAX.SCRMIN 3# SCRMAX - specifies the maximum size of# UniVerse scratch buffer pool. This# value must be no smaller than SCRMIN# and must be specified after SCRMIN.SCRMAX 5# SCRSIZE - specifies the size of newly# allocated internal scratch buffers. This# value must not be smaller than 512 and# values greater than 2048 are not advised.SCRSIZE 512------------------------------Gregor ScottSoftware ArchitectPentana SolutionsMount Waverley VIC Australia------------------------------

HelloI'll check performance and CPU charge of program.Usually I use this$OOTIONS TIME.MILLISECONDST=SYSTEM(12)DO...PRINT SYSTEM(12)-TIts about e execution time but I'll identify the CPU charge... To use efficient instructions because the program run for large qty if users. May I use SYSTEM(9) instead system 12?Thanks for any suggestions.------------------------------Manu Fernandes------------------------------

I am in the process of debugging several hundred calls from UniObjects to host routines. I wrote a debug program that gives me SYSTEM(11), the call stack and other items that are written to a file. The problem is that when I sort the file by the CPU time (SYSTEM(11)) there are several (5) calls all at the same time. Is there a way to get a time that has a higher resolution than just  milliseconds?TIA,Steve------------------------------Steve MayoSoftware Engineerbasys------------------------------