rdelaage
/
postgresql
mirror of https://git.postgresql.org/git/postgresql.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Changed getCrossReference to return information about each composite key seperately 

It used to return them as a,b in one row, and now returns
a in one row, and b in one row
This commit is contained in:
Dave Cramer 2002-07-12 13:07:48 +00:00
parent b82b2db7c1
commit 791a40f943
1 changed files with 20 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

55
src/interfaces/jdbc/org/postgresql/jdbc2/DatabaseMetaData.java
View File

@ -15,7 +15,7 @@ import org.postgresql.util.PSQLException;
/*
* This class provides information about the database as a whole.
*
* $Id: DatabaseMetaData.java,v 1.57 2002/06/25 16:30:49 davec Exp $
* $Id: DatabaseMetaData.java,v 1.58 2002/07/12 13:07:48 davec Exp $
*
* <p>Many of the methods here return lists of information in ResultSets. You
* can use the normal ResultSet methods such as getString and getInt to
@ -2625,8 +2625,8 @@ WHERE
// Parse the tgargs data
StringBuffer fkeyColumns = new StringBuffer();
StringBuffer pkeyColumns = new StringBuffer();
String fkeyColumn="";
String pkeyColumn="";
// Note, I am guessing at most of this, but it should be close
@ -2634,46 +2634,31 @@ WHERE
// the keys are in pairs and start after the first four arguments
// the arguments are seperated by \000
int numColumns = (rs.getInt(8) >> 1) - 2;
int keySequence = rs.getInt(4); //KEY_SEQ
// get the args
String targs = rs.getString(9);
// start parsing from the end
int pos = targs.lastIndexOf("\\000");
// args look like this
//<unnamed>\000ww\000vv\000UNSPECIFIED\000m\000a\000n\000b\000
// we are primarily interested in the column names which are the last items in the string
for (int c = 0;c < numColumns;c++)
StringTokenizer st = new StringTokenizer(targs, "\\000");
int advance = 4 + (keySequence-1) * 2;
for( int i=0; st.hasMoreTokens() && i < advance ; i++ ) st.nextToken(); // advance to the key column of interest
if ( st.hasMoreTokens() )
{
// this should never be, since we should never get to the beginning of the string
// as the number of columns should override this, but it is a safe test
if (pos > -1)
{
int pos2 = targs.lastIndexOf("\\000", pos - 1);
if (pos2 > -1)
{
// seperate the pkColumns by ',' s
if (pkeyColumns.length() > 0)
pkeyColumns.insert(0, ',');
// extract the column name out 4 characters ahead essentially removing the /000
pkeyColumns.insert(0, targs.substring(pos2 + 4, pos)); //PKCOLUMN_NAME
// now find the associated fkColumn
pos = targs.lastIndexOf("\\000", pos2 - 1);
if (pos > -1)
{
if (fkeyColumns.length() > 0)
fkeyColumns.insert(0, ',');
fkeyColumns.insert(0, targs.substring(pos + 4, pos2)); //FKCOLUMN_NAME
}
}
}
fkeyColumn = st.nextToken();
}
if ( st.hasMoreTokens() )
{
pkeyColumn = st.nextToken();
}
tuple[3] = pkeyColumns.toString().getBytes(); //PKCOLUMN_NAME
tuple[7] = fkeyColumns.toString().getBytes(); //FKCOLUMN_NAME
tuple[3] = pkeyColumn.getBytes(); //PKCOLUMN_NAME
tuple[7] = fkeyColumn.getBytes(); //FKCOLUMN_NAME
tuple[8] = rs.getBytes(4); //KEY_SEQ
tuple[11] = rs.getBytes(3); //FK_NAME