Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>Generally speaking, an auto-incrementing key will not collide; in fact, I'm not sure how to make it happen (since most definitions in SQL include a Unique constraint). You might get <code>duplicate key</code> errors, though. Why are you letting outside vendors dictate what your internal keys are? Accept a completed record from them, and generate the keys yourself. If you're running out of keys, increase the size of the column (to a long). You could generate a new GUID column, yes, but you'd likely still be generating the GUIDs anyways, so what would be the difference? </p> <p>Often, most tables in a database will have more than one 'unique' key. These are the 'natural' keys - the actual columns that make up a unique set of data; unfortunately, this may end up being the entire width of the table (which is why <code>id</code> columns are used). Also, unless there is some bizarre use case, don't make the server know about the client ids - make the clients know about the server ids, and keep track of it in a separate column (if necessary) on the client. Then can then have their own internal ids that they use before uploading, that have no relation to the keys you give back - and which they never share with the server.</p> <p>It doesn't really matter what the server id is (<code>int</code> based or a GUID), although I'd recommend sticking with whatever you have. Whenever the client gives you a row to put on the server, give them back the generated id from the server. This prevents clients from attempting to insert already-in-use ids, and related issues (how do you know the id they gave you shouldn't have been generated by some other device?). I would say that clients do not get to dictate internal ids on the server </p> <p><HR/> EDIT:</p> <p>In light of some quick research into replication (in response to HLGEM), and re-reading the original question; I initially took the question as asking about letting clients dictate the internal primary keys (which is still bad), and/or replacing the <code>int</code> primary keys with the GUID. Although your exact needs may not require it, adding an extra GUID column that the client can populate would work. Some recommendations:<br> 1) Treat unique key violations (accidental or malicious) on the GUID column as a <em>business</em> exception, not a <em>system</em> exception. I'd probably recommend telling the client to re-generate the GUID and re-submit (silently). 2) The client <em>never</em> gets to dictate the acutal primary key column values in the database. In fact, if you use a GUID column, the client may never even need to know about them.</p>
    singulars
    1. This table or related slice is empty.
    1. POInt Autoincrementing Primary Key and GUID column
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USClockwork-Muse
    1. USClockwork-Muse
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. POInt Autoincrementing Primary Key and GUID column
      singulars
      1. PTQuestion
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTDownMod
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTAcceptedByOriginator
    1. COI believe they would collide in the scenario given by the OP.
      singulars
      1. PO
      1. USHLGEM
    2. CO... I specifically stated that clients shouldn't get to dictate server ids specifically to *avoid* the colissions. If they are attempting to re-use unique natural keys, that is a *business* problem, not a *system* one, and should be accounted for as such.
      singulars
      1. PO
      1. USClockwork-Muse
    3. COI entered up working with the Microsoft Sync Framework and adding a column guid to the sync tables along with the existing auto int primary keys' - during sync for the client on the guid's are synced down . this is working really well - the sync framework does the object matching based on the GUID's - whilst existing db code still functions as usual as we still have the auto incre keys
      singulars
      1. PO
      1. USTheTiger
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload