|
i think so, how would youtube or those peer-to-peer(p2p) have so many?!
i have no degree in programming but, dont you agree with me???
What i also think is that you have to make a stand alone, because you are emphasing on big big big files you know, that´s expensive to rent a server for.
Maybe create your own server, or a network of many client (p2p) to swap data, that way you wont pay anything, just power bill and telecomunications bill. 
nelsonpaixao@yahoo.com.br
trying to help & get help
|
|
|
|
|
If it is just storage then it is fairly straight forward, just add hard drives. You may want to store the different types of MM data into different tables linked to the owner (ie have an owner table with a vid attribute table anda PDF attribute table etc)
You could also store the MM in the file system and only the links in the database.
If you are doing processing on the data and it is seriously large (100gb+ without MM) I would look into Oracle.
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
Hi
I have designed a large system for storing files, any kind of files
The best way in my opinion is to store files on the hard disk and keep filenames or paths in the DB.
Another note will be useful to you: Don't put all files in one folder direcly. This would result in slow serching on a specific file. Rather, you can specify a max file count for each subfolder, say 2000 files. So, if a subfolder reached the 2000-files limit, create a new subfolder and set it the current one.
In DB, you can create a table for [SubFolder] contains the fullpath of each subfolder and the table [File] contains a colunm for filename and a column for a foregin key to the parent subfolder.
Mostafa Noor
Egypt
|
|
|
|
|
Thanx man ! I ve decided to apply the same method ! I have made up my mind about movies and mp3 files, but for PDF/TEXT/Image files it would be possible to store them in database or u suggest storing them all on hard drive ?
Mr.K
|
|
|
|
|
I want to use the system view sys.database_role_members to show what users are in what particular roles. Sys.database_role_members has exactly what I need (the role_principal_id and the member_principal_id) but I want to see them as user or group names, not ID numbers. The view sys.database_principals shows the name and the ID, so I know I can use that as a look-up, but I don't know how to properly join the two.
It seems like it must be pretty simple, but I'm new and I'm just not seeing it. Any help would be appreciated.
Thanks.
|
|
|
|
|
This should get you started: Using Joins[^]. In your case concentrate on inner join if you want only matching pairs.
The need to optimize rises from a bad design
|
|
|
|
|
I can do a basic join but here's what I am having a hard time with:
sys.database_role_members is something like this--
role_principal_id --------------- member_principal_id
5555 ------------------------------ 1
5555 ------------------------------ 2
6666 ------------------------------ 1
6666 ------------------------------ 2
But I want both columns to have names to identify them.
role_principal_id --------------- member_principal_id
data_entry ----------------------- janedoe
data_entry ----------------------- joeschmoe
data_guest ----------------------- janedoe
data_guest ----------------------- joeschmoe
Those names can all be found in one column of sys.database_principals. And I don't think I get how to connect both columns of sys.database_principals to one column of sys.database_principals.
|
|
|
|
|
Ok, I see. Could you post what kind of SQL statement have you made so far?
The basic idea is that you would have the sys.database_principals twice with different aliases (actually this is the same tecnique I used in my previous reply to you). One alias is for role and it's name and the other is for the role member and it's name.
The need to optimize rises from a bad design
|
|
|
|
|
Aha. I needed to use aliases.
I've got it. Thank you!
|
|
|
|
|
No problem 
The need to optimize rises from a bad design
|
|
|
|
|
Hi,
Does anyone know if i can change SQL server settings using a script or query?
I should set "minimum memory used" to 512MB and "SQL boost" to true using a script.
thx
kurt
|
|
|
|
|
|
Okay, thx..
just what i was looking for....
greetz
Kurt
|
|
|
|
|
No problem.
Bob
Ashfield Consultants Ltd
|
|
|
|
|
Here is what i have spent the past day banging my head against.
I have a report with page headers and in the report footer i display two subreports and i need to suppress the page header when the subreports are displayed. From what i understand i need to write a formatting formula but i have no idea what this formula is or what it looks like
I am very new to Crystal reports. Any help will be appreciated
Thanks in advance
Colwin
|
|
|
|
|
Be glad to know that in other places in the world people are banging their head to!!!
I need CR to end my applicaton, if i was more than a begginer in CR i would be glad to help you there.
Good Luck 
nelsonpaixao@yahoo.com.br
trying to help & get help
|
|
|
|
|
Create a Formula and place it in a blank and suppressed section before your sub-report, force new page after this section as well
=== Formula SuppressHeader
<br />
WhilePrintingRecords;<br />
BooleanVar SupressStuff;<br />
SupressStuff:= true<br />
Create another Formula and place it after your sub report within a blank suppressed section
=== Formula SuppressHeaderNOT
<br />
WhilePrintingRecords;<br />
BooleanVar SupressStuff;<br />
SupressStuff:= false<br />
Create a Third Formula which will give the value SupressStuff to other formula (this sounds weird but bear with me). Place a copy of this in the report header as well.
== Formula SuppressStuffDisplay
<br />
WhilePrintingRecords;<br />
BooleanVar SupressStuff;<br />
Now in your Header and Footer Sections go into Section Expert and click the formula box at Suppress (No Drill-Down) and place the third formula in the
@SuppressStuffDisplay
or if you want to be fancy use this which does the same thing
<br />
if @SuppressStuffDisplay= true <br />
then<br />
true<br />
else<br />
false<br />
How this works; simple answer is flow control. Crystal runs from Top to Botton and runs in a flow. So it always does the stuff before it does the current stuff. Event fire as well like new page and page headers etc.. So before it gets to the subreport it set the value of SupressStuff to false which then suppressed any headers and footers. Then after the subreport it sets SupressStuff to true which allow headers and footers to be printed
easy eh ?
|
|
|
|
|
In your main report, go to Section Expert and find the properties of the Report Footer. Select the "New Page Before" option for the footer section. This will force a new page to be printed right before your embedded sub-report. I've used this method and it works fine. Your sub-report will not have the page header or footer from the main report. Give it a try.
|
|
|
|
|
I have this code, how i can place the result to the MSword ?
SELECT
[Table Name] = OBJECT_NAME(c.object_id),
[Column Name] = c.name,
[Description] = ex.value
FROM
sys.columns c
LEFT OUTER JOIN
sys.extended_properties ex
ON
ex.major_id = c.object_id
AND ex.minor_id = c.column_id
AND ex.name = ''MS_Description''
WHERE
OBJECTPROPERTY(c.object_id, ''IsMsShipped'')=0
ORDER
BY OBJECT_NAME(c.object_id), c.column_id '
Thank u before... 
Irvinia
|
|
|
|
|
You may want to read up on Office Interop or PIA.
"The clue train passed his station without stopping." - John Simmons / outlaw programmer
"Real programmers just throw a bunch of 1s and 0s at the computer to see what sticks" - Pete O'Hanlon
"Not only do you continue to babble nonsense, you can't even correctly remember the nonsense you babbled just minutes ago." - Rob Graham
|
|
|
|
|
Thank u Paul, i'll try.
Irvin
|
|
|
|
|
Not a problem
Do a search on this site and there are some articles floating around here in regards to Office Interop.
"The clue train passed his station without stopping." - John Simmons / outlaw programmer
"Real programmers just throw a bunch of 1s and 0s at the computer to see what sticks" - Pete O'Hanlon
"Not only do you continue to babble nonsense, you can't even correctly remember the nonsense you babbled just minutes ago." - Rob Graham
|
|
|
|
|
I'm not sure exactly how to do this which is why I don't know exactly what to put in the post Subject. Pretty new to SQL. Help is appreciated.
I want to use the information from a system view to get information about table permissions in our database.
Right now I use:
select Grantee, Table_Name, Privilege_Type from information_schema.table_privileges
which gives me something like this:
GRANTEE---TABLE_NAME----PRIVILEGE_TYPE
User1------Table1----------Update
User1------Table1----------Insert
User1------Table1----------Select
User2------Table1----------Select
User2------Table1----------Update
But it's over 1500 records long. I'd prefer something like this:
TABLE-----USER1-----------------------USER2
Table1-----Insert/Select/Update--------Select/Update
Can anyone tell me how I could do that? Or have any better ideas of how to view the information_schema.table_privileges data in a more readable format?
Thank you!
|
|
|
|
|
The readability would be a matter of an opinion
If you set up your results as in your post, you could have one very long row (depending on how many users you have).
Perhaps, you could get a starting point if you take tables as rows and then users for each table with fixed columns as privileges. Something like the following. Then if you like you can start pivoting on tables or users using sub selects in FROM section and modify the structure as you like.
Also notice that you can place FOR XML AUTO at the end of the query, if it makes the results more readable.
select userinfo.Table_Name,
userinfo.Grantee ,
case when selectinfo.GRANTOR is not null then 'Yes' else '' end SelectPermission,
case when insertinfo.GRANTOR is not null then 'Yes' else '' end InsertPermission,
case when updateinfo.GRANTOR is not null then 'Yes' else '' end UpdatePermission,
case when updateinfo.GRANTOR is not null then 'Yes' else '' end DeletePermission
from information_schema.table_privileges as userinfo
left outer join
information_schema.table_privileges as selectinfo
on userinfo.grantee = selectinfo.grantee
and userinfo.table_catalog = selectinfo.table_catalog
and userinfo.table_schema = selectinfo.table_schema
and userinfo.table_name = selectinfo.table_name
and selectinfo.PRIVILEGE_TYPE = 'SELECT'
left outer join
information_schema.table_privileges as insertinfo
on userinfo.grantee = insertinfo.grantee
and userinfo.table_catalog = insertinfo.table_catalog
and userinfo.table_schema = insertinfo.table_schema
and userinfo.table_name = insertinfo.table_name
and insertinfo.PRIVILEGE_TYPE = 'INSERT'
left outer join
information_schema.table_privileges as updateinfo
on userinfo.grantee = updateinfo.grantee
and userinfo.table_catalog = updateinfo.table_catalog
and userinfo.table_schema = updateinfo.table_schema
and userinfo.table_name = updateinfo.table_name
and updateinfo.PRIVILEGE_TYPE = 'UPDATE'
left outer join
information_schema.table_privileges as deleteinfo
on userinfo.grantee = deleteinfo.grantee
and userinfo.table_catalog = deleteinfo.table_catalog
and userinfo.table_schema = deleteinfo.table_schema
and userinfo.table_name = deleteinfo.table_name
and deleteinfo.PRIVILEGE_TYPE = 'DELETE'
order by Table_Name, Grantee
The need to optimize rises from a bad design
|
|
|
|
|
I believe the answer to your problem is to use a Pivot query in SQL 2005. I'm not sure how to due this, but it's worth investigating.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
