Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.lang.clipper.visual-objects > #737 > unrolled thread
| Started by | JoseMaria <jmespinosabaviera@yahoo.com> |
|---|---|
| First post | 2011-04-07 02:48 -0700 |
| Last post | 2011-04-07 22:25 +0000 |
| Articles | 10 — 6 participants |
Back to article view | Back to comp.lang.clipper.visual-objects
Problem with VO- duplicate registers in sql server JoseMaria <jmespinosabaviera@yahoo.com> - 2011-04-07 02:48 -0700
Re: Problem with VO- duplicate registers in sql server Meinhard Schnoor-Matriciani <msm@appfact.de> - 2011-04-07 13:46 +0200
Re: Problem with VO- duplicate registers in sql server E®!k \\/!$$E® <nospam@nospam.com> - 2011-04-07 21:27 +0200
Re: Problem with VO- duplicate registers in sql server JoseMaria <jmespinosabaviera@yahoo.com> - 2011-04-07 14:24 -0700
Re: Problem with VO- duplicate registers in sql server Karl Faller <k.faller_withoutthat_@onlinehome.de> - 2011-04-07 23:45 +0200
Re: Problem with VO- duplicate registers in sql server JoseMaria <jmespinosabaviera@yahoo.com> - 2011-04-07 14:54 -0700
Re: Problem with VO- duplicate registers in sql server "Willie Moore" <williem@wmconsulting.com> - 2011-04-08 08:22 -0500
Re: Problem with VO- duplicate registers in sql server JoseMaria <jmespinosabaviera@yahoo.com> - 2011-04-08 11:46 -0700
Re: Problem with VO- duplicate registers in sql server "Willie Moore" <williem@wmconsulting.com> - 2011-04-08 14:46 -0500
Re: Problem with VO- duplicate registers in sql server "Geoff Schaller" <geoffx@softwarexobjectives.com.au> - 2011-04-07 22:25 +0000
| From | JoseMaria <jmespinosabaviera@yahoo.com> |
|---|---|
| Date | 2011-04-07 02:48 -0700 |
| Subject | Problem with VO- duplicate registers in sql server |
| Message-ID | <bdfa758b-f750-4484-b417-a7002c6a98c9@a26g2000vbo.googlegroups.com> |
using VO 2.8 sp3 and RDD version of VO2ADO to access SQL server 2008 R2 express: I have noticed that whenever there are in one table duplicate records (2 or 3 records with exactly the same information) and I try to delete/ update/insert records to this table I have sometimes problems with strange crashes in my application. Is there any easy solution to this problem ? Is it possible to add one column to each table with a numeric value that is automatically informed with a different value each time I insert a record ? I have tried adding column uniqueidentifier to a table but it is populated with nulls, not values. Thanks for your help.
[toc] | [next] | [standalone]
| From | Meinhard Schnoor-Matriciani <msm@appfact.de> |
|---|---|
| Date | 2011-04-07 13:46 +0200 |
| Message-ID | <4d9da73a$0$7663$fca84a42@newsspool1.arcor-ip.de> |
| In reply to | #737 |
JoseMaria, you apparently haven't worked much with SQL servers, have you ? In SQL a table should ALWAYS have a column with values which are unique across all columns, so that you can adress every single row uniquely (this is referred to as Primary Key). Your approach with a uniqueidentifier is one way to do so, but if you don't put the guid's in there when inserting, you have to tell the db to do so automatically, which is usually done by a default value in the column definition. I strongly recommend to get yourself a book about SQL and the way it works. Regards Meinhard Am 07.04.2011 11:48, schrieb JoseMaria: > using VO 2.8 sp3 and RDD version of VO2ADO to access SQL server 2008 > R2 express: > > I have noticed that whenever there are in one table duplicate records > (2 or 3 records with exactly the same information) and I try to delete/ > update/insert records to this table I have sometimes problems with > strange crashes in my application. > > Is there any easy solution to this problem ? > > Is it possible to add one column to each table with a numeric value > that is automatically informed with a different value each time I > insert a record ? I have tried adding column uniqueidentifier to a > table but it is populated with nulls, not values. > > Thanks for your help.
[toc] | [prev] | [next] | [standalone]
| From | E®!k \\/!$$E® <nospam@nospam.com> |
|---|---|
| Date | 2011-04-07 21:27 +0200 |
| Message-ID | <906hghFpceU1@mid.individual.net> |
| In reply to | #737 |
JoseMaria, In addition of what Meinhard wrote: The Vo2Ado RDD is not showing you the new added record after an append if you have a ClientSide cursor. This can be solved with a serverside cursor, but that will slow down the whole thing dramaticly. My suggestion will be not using the VO2ADO.RDD at all. You do not take full advantage of SqlServer if you use it like a DBF table. You best use the AdoServer class for data attached to a form, or use your own scatter/gather. I always use a AdoCommand with an Insert command to add records to a table and retrieve the PrimaryKey value of the new row with SELECT @@IDENTITY hth Erik "JoseMaria" wrote in message news:bdfa758b-f750-4484-b417-a7002c6a98c9@a26g2000vbo.googlegroups.com... using VO 2.8 sp3 and RDD version of VO2ADO to access SQL server 2008 R2 express: I have noticed that whenever there are in one table duplicate records (2 or 3 records with exactly the same information) and I try to delete/ update/insert records to this table I have sometimes problems with strange crashes in my application. Is there any easy solution to this problem ? Is it possible to add one column to each table with a numeric value that is automatically informed with a different value each time I insert a record ? I have tried adding column uniqueidentifier to a table but it is populated with nulls, not values. Thanks for your help.
[toc] | [prev] | [next] | [standalone]
| From | JoseMaria <jmespinosabaviera@yahoo.com> |
|---|---|
| Date | 2011-04-07 14:24 -0700 |
| Message-ID | <f3515e10-f5b1-421f-bc5e-d3e3bc73507c@p3g2000vbv.googlegroups.com> |
| In reply to | #744 |
Thanks to bothe of you for the help. Meinhard is right I have very little experience using SQL server. I have solved the problem of application crashes (for now) by adding a bigint identity column to each table, so whenever I append a record, sql server informs this column with a sequential number. What worries me now is what erik has said, which I don't understand. Does it mean that if one user appends a record to a table, the other users in the network are not going to see that record until they close the database and use it again ? Regarding the rdd, it is the only solution I know to avoid recoding hundreds or thousands of code lines, of an application that so far has used DBFs. Thanks Jose Maria On Apr 7, 9:27 pm, E®!k \\/!$$E® <nos...@nospam.com> wrote: > JoseMaria, > > In addition of what Meinhard wrote: > The Vo2Ado RDD is not showing you the new added record after an append if > you have a ClientSide cursor. > This can be solved with a serverside cursor, but that will slow down the > whole thing dramaticly. > My suggestion will be not using the VO2ADO.RDD at all. You do not take full > advantage of SqlServer if you use it like a DBF table. > You best use the AdoServer class for data attached to a form, or use your > own scatter/gather. > > I always use a AdoCommand with an Insert command to add records to a table > and retrieve the PrimaryKey value of the new row with SELECT @@IDENTITY > > hth > > Erik > > "JoseMaria" wrote in message > > news:bdfa758b-f750-4484-b417-a7002c6a98c9@a26g2000vbo.googlegroups.com... > > using VO 2.8 sp3 and RDD version of VO2ADO to access SQL server 2008 > R2 express: > > I have noticed that whenever there are in one table duplicate records > (2 or 3 records with exactly the same information) and I try to delete/ > update/insert records to this table I have sometimes problems with > strange crashes in my application. > > Is there any easy solution to this problem ? > > Is it possible to add one column to each table with a numeric value > that is automatically informed with a different value each time I > insert a record ? I have tried adding column uniqueidentifier to a > table but it is populated with nulls, not values. > > Thanks for your help.
[toc] | [prev] | [next] | [standalone]
| From | Karl Faller <k.faller_withoutthat_@onlinehome.de> |
|---|---|
| Date | 2011-04-07 23:45 +0200 |
| Message-ID | <9qbsp6tfvefmn08rjfeu9ijbb89tv9kek0@4ax.com> |
| In reply to | #745 |
JoseMaria >Does it mean that if one user appends a record to a table, the other >users in the network are not going to see that record until they close >the database and use it again ? No - AFAIK the others will, as fast as your append/insert is written through. But with a clientside cursor YOU are looking at a "snapshot" of the db, taken at the moment you issued your select. So, this cursor will not reflect any operations you perform later on that table - to see them, you'll have to repeat the select statement. Beware, that others might correct me, as my knowlegde of these things is rather limited ;-) Karl
[toc] | [prev] | [next] | [standalone]
| From | JoseMaria <jmespinosabaviera@yahoo.com> |
|---|---|
| Date | 2011-04-07 14:54 -0700 |
| Message-ID | <623f0ac9-ab83-480d-85d5-8f83b8822f94@p3g2000vbv.googlegroups.com> |
| In reply to | #746 |
Karl, And when do I issue a "select" ? I don't issue selects as I use the rdd version, I use dbseek( cKey ) to search for records and after that a "while .not. eof() + dbskip" loop. You mean every time I do a dbseek I refresh the information and can see all new data added ? Thanks Jose Maria On Apr 7, 11:45 pm, Karl Faller <k.faller_withoutth...@onlinehome.de> wrote: > JoseMaria>Does it mean that if one user appends a record to a table, the other > >users in the network are not going to see that record until they close > >the database and use it again ? > > No - AFAIK the others will, as fast as your append/insert is written > through. But with a clientside cursor YOU are looking at a "snapshot" > of the db, taken at the moment you issued your select. So, this cursor > will not reflect any operations you perform later on that table - to > see them, you'll have to repeat the select statement. > > Beware, that others might correct me, as my knowlegde of these things > is rather limited ;-) > > Karl
[toc] | [prev] | [next] | [standalone]
| From | "Willie Moore" <williem@wmconsulting.com> |
|---|---|
| Date | 2011-04-08 08:22 -0500 |
| Message-ID | <inn25l$ui4$1@speranza.aioe.org> |
| In reply to | #747 |
Jose, While this is not a free solution, the quickest way to get a DBF application up and running in SQL is to buy Geoff's SQL classes. They do a really good job of shielding you from SQL but do not limit you to just DBF style coding. As you get more comfortable with SQL, you can start using proper selects which will speed up your program. I have used Geoff's classes and they work. Regards, Willie "JoseMaria" wrote in message news:623f0ac9-ab83-480d-85d5-8f83b8822f94@p3g2000vbv.googlegroups.com... Karl, And when do I issue a "select" ? I don't issue selects as I use the rdd version, I use dbseek( cKey ) to search for records and after that a "while .not. eof() + dbskip" loop. You mean every time I do a dbseek I refresh the information and can see all new data added ? Thanks Jose Maria On Apr 7, 11:45 pm, Karl Faller <k.faller_withoutth...@onlinehome.de> wrote: > JoseMaria>Does it mean that if one user appends a record to a table, the > other > >users in the network are not going to see that record until they close > >the database and use it again ? > > No - AFAIK the others will, as fast as your append/insert is written > through. But with a clientside cursor YOU are looking at a "snapshot" > of the db, taken at the moment you issued your select. So, this cursor > will not reflect any operations you perform later on that table - to > see them, you'll have to repeat the select statement. > > Beware, that others might correct me, as my knowlegde of these things > is rather limited ;-) > > Karl
[toc] | [prev] | [next] | [standalone]
| From | JoseMaria <jmespinosabaviera@yahoo.com> |
|---|---|
| Date | 2011-04-08 11:46 -0700 |
| Message-ID | <dffada46-6485-4a96-8fa3-8e970d2a2dc7@u12g2000vbf.googlegroups.com> |
| In reply to | #767 |
Thanks to all for your sugestions and your help. Jose Maria
[toc] | [prev] | [next] | [standalone]
| From | "Willie Moore" <williem@wmconsulting.com> |
|---|---|
| Date | 2011-04-08 14:46 -0500 |
| Message-ID | <innomp$mo0$1@speranza.aioe.org> |
| In reply to | #770 |
Jose, As you move into SQL, feel free to ask questions. Most of us who have made the jump are willing to help. SQL can be frustrating at times but it is well worth the move (in my opinion). Regards, Willie
[toc] | [prev] | [next] | [standalone]
| From | "Geoff Schaller" <geoffx@softwarexobjectives.com.au> |
|---|---|
| Date | 2011-04-07 22:25 +0000 |
| Message-ID | <4d9e39c6$0$11192$c30e37c6@exi-reader.telstra.net> |
| In reply to | #745 |
Jose, > I have solved the problem of application crashes (for now) by adding a > bigint identity column to each table, so whenever I append a record, > sql server informs this column with a sequential number. Do not waste a big int - just use an int. Are you going to add more than 2,000 billion rows? <g>. > Does it mean that if one user appends a record to a table, the other > users in the network are not going to see that record until they close > the database and use it again ? No. Simply call Refresh() on the record set. The server class also has a refresh. If you are using bBrowser, set up and auto refresh for every 30 seconds or something and do oa deliberate one prior to an action like editing. This can be automated in your base classes. > Regarding the rdd, it is the only solution I know to avoid recoding > hundreds or thousands of code lines, of an application that so far has > used DBFs. DO NOT use the RDD. It was really only written as an example of the code (because you have the source) and cannot be used in a live situation. I have a series of classes (which I sell - they are not free) which allows you to substantially use your existing code without modification. But you can then use pure DFB syntax without having to know SQL. But I encourage you to learn SQL. Ultimately it is the only way. Geoff
[toc] | [prev] | [standalone]
Back to top | Article view | comp.lang.clipper.visual-objects
csiph-web