![]() | |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
To: comp.databases.pick Just FYI for those using the MVSP .Net library, the Insert function has a very ugly bug. When doing an insert to a blank string, it generates bad data. value = TIGR.MVSP.Functions.Insert(value, -1, "new value") Instead of the contents of value being "new value", it's actually blank> @am "new value". This happens because the Insert() routine isn't checking to see if the destination string is blank - it just blindly tacks on an @am followed by the value specified for the insert. I've posted this to the MVSP forum at TigerLogic. I hope they get this one fixed soon because it's a show stopper in my process of moving my old ODBC class library code to MVSP. If anyone is using the Java version of the MVSP library, you might want to check to see if this bug is present there as well. |
#3
| |||
| |||
|
|
From Newsgroup: comp.databases.pick On 9/23/11 9:48 AM, Gene Buckle wrote: To: comp.databases.pick Just FYI for those using the MVSP .Net library, the Insert function has a very ugly bug. When doing an insert to a blank string, it generates bad data. value = TIGR.MVSP.Functions.Insert(value, -1, "new value") Instead of the contents of value being "new value", it's actually blank> @am "new value". This happens because the Insert() routine isn't checking to see if the destination string is blank - it just blindly tacks on an @am followed by the value specified for the insert. I've posted this to the MVSP forum at TigerLogic. I hope they get this one fixed soon because it's a show stopper in my process of moving my old ODBC class library code to MVSP. If anyone is using the Java version of the MVSP library, you might want to check to see if this bug is present there as well. By "blank string" do you mean "a string containing one character, the space character" or do you mean "a null (empty) string"? If you mean the former, then the insert is performing correctly, giving you an end result with two attributes, the first being the one space character and the second being your newly inserted value. By blank I mean a totally empty, null string. I would agree that if the |
#4
| |||
| |||
|
|
TL has opened a new bug for this. g. |

#5
| |||
| |||
|
|
From Newsgroup: comp.databases.pick TL has opened a new bug for this. g. A new action item? Heh.. I know whacha mean but it looks like a subconscious slip-up suggesting this problem will now be replaced by a different and new problem. Yeah. The MVSP library is pretty minimal and is pretty lackluster in |


|
PS: I'll try to pop in IRC soon during the day.. lots of construction stuff going on and trying to push out the new shopping cart/checkout before spooky day. |

#6
| |||||||
| |||||||
|
|
Glen wrote: A new action item? Heh.. I know whacha mean but it looks like a subconscious slip-up suggesting this problem will now be replaced by a different and new problem. |
|
Yeah. The MVSP library is pretty minimal and is pretty lackluster in comparison with the feature set in the D3 Class Library. |
|
After I'm done with my current porting project I'm going to develop something that'll encapsulate the MVSP stuff in something that will be transparently compatible with the old style class library. |
|
If it doesn't turn out to be a total dogs breakfast I'll post it online for others to use. |
|
The MVSP docs are pretty bad too - whoever was responsible for that particular part of the project really should go stand in the corner. ![]() |
|
TL tells me that the fix is in the testing process now. It surprises me that they're on it this fast considering my prior experiences. ![]() |

|
PS: I'll try to pop in IRC soon during the day.. lots of construction stuff going on and trying to push out the new shopping cart/checkout before spooky day. Email me first - My IRC "schedule" has become extremely erratic. ![]() g. |
#7
| |||
| |||
|
|
"Gene Buckle" wrote: Glen wrote: A new action item? Heh.. I know whacha mean but it looks like a subconscious slip-up suggesting this problem will now be replaced by a different and new problem. Heck man, a problem was reported and an action item filed within a day. *What else should they be doing? *Glen, your cause/effect relationship there is typical for CDP but very untypical for you. |
|
Been a long time since IRC (picksource days). *What channel? *Any schedule? Best, T |
#8
| |||
| |||
|
|
TL tells me that the fix is in the testing process now. *It surprises me that they're on it this fast considering my prior experiences. ![]() |
#9
| |||||
| |||||
|
|
From Newsgroup: comp.databases.pick "Gene Buckle" wrote: Yeah. The MVSP library is pretty minimal and is pretty lackluster in comparison with the feature set in the D3 Class Library. Sure but it's also extensible. I have a small library of extensions now to give it more "oompf" (and to work around little issues like this -1 thing, of which I've found a few). It just took a couple days |
|
After I'm done with my current porting project I'm going to develop something that'll encapsulate the MVSP stuff in something that will be transparently compatible with the old style class library. I was tempted to do something similar. But MVSP is a completely new library and I think it would be better to move forward with it than back. Extend what's there rather than encapsulating in something new. The number of people who would benefit from something that looks like D3CL is really tiny. Sure, add features, but I'm just saying it's not necessary to try to shoehorn those features into the same class structures or signatures as D3CL. |
|
If it doesn't turn out to be a total dogs breakfast I'll post it online for others to use. YMMV but I've been so turned off from contributing FOSS to this community that it's really depressing. It's the whole per-seat pricing mentality that makes it tough Tony. That and |
|
The MVSP docs are pretty bad too - whoever was responsible for that particular part of the project really should go stand in the corner. ![]() You may recall that I was writing a small book on MVSP when it first came out. After a couple chapters I decided the audience would be so small that it really wasn't worth the time anymore. Maybe that's what they were thinking at TL too. :b A good treatise on MVSP might attract more people to it - you never know. |
|
Been a long time since IRC (picksource days). What channel? Any schedule? irc.freenode.net, channel is ##pick. |
#10
| ||||
| ||||
|
|
The -1 issue is embarassing. It NEVER should have made it out of beta with such a fundamental flaw. I suspect the guy that wrote the code has either never written code in Pick BASIC or was very, very new to it. |
|
It's not a matter of shoe-horning features in. It's about making my life easier. I've got a 200k+ LOC project to port to MVSP and the last thing I want to do is do a whole ton of hand editing to change thousands of brExtractStr() and brInsertStr() to their MVSP equivalents. If it doesn't turn out to be a total dogs breakfast I'll post it online for others to use. |
|
YMMV but I've been so turned off from contributing FOSS to this community that it's really depressing. It's the whole per-seat pricing mentality that makes it tough Tony. That and a whole lot of "what's in it for ME?!" attitude. Frankly, I don't care either way. If I deem it a worthy example of my work (and I get permission from my employer), I'll post it. If someone benefits from it, great. If not, no big deal. |

|
The MVSP docs are pretty bad too - whoever was responsible for that particular part of the project really should go stand in the corner. ![]() You may recall that I was writing a small book on MVSP when it first came out. After a couple chapters I decided the audience would be so small that it really wasn't worth the time anymore. Maybe that's what they were thinking at TL too. :b A good treatise on MVSP might attract more people to it - you never know. Frankly, I'd love to see a huge uptake in the use of MVSP, especially over per-seat priced solutions. |

![]() |
| Thread Tools | |
| Display Modes | |
| |