Re: JDBC URL
Thufir wrote:
On Jan 26, 5:27 am, Lew <no...@lewscanon.com> wrote:
Thufir wrote:
What's the purpose of including the scheme twice? (different driver)
Class.forName("com.microsoft.jdbc.sqlserver.SQLServerDriver");
con = java.sql.DriverManager.getConnection(getConnectionUrl
(),userName,password);
http://support.microsoft.com/kb/313100
the first line indicates a scheme of jdbc in the driver type, why repeat
the scheme information again in the URL? Seems redundant and an
What's redundant? I see no repetition here.
The database connection is already jdbc, as established on the first
line. For the connection URL, why not take that information from the
driver? That is, what else could possible be there? There's no
Because URLs must have a scheme. That's required. Otherwise it isn't a URI/URL.
rationale for specifying the connection scheme as anything other than
jdbc, as that's what the driver expects.
The URL cannot know what the driver expects. It's a namespace and must be
completely specified.
Since the driver expects a specific scheme, why give an opportunity to
specify an incorrect scheme?
You cannot omit the scheme from a URI. If you do, it's not a URI. If you
specify an "incorrect" scheme, the system won't even ask the JDBC drivers if
they can handle it. It'll ask the drivers, if any, for the scheme you do specify.
You have to specify the "jdbc:" scheme, otherwise, how would the system know
to search the "jdbc:" drivers and not the "http:" drivers or the "foo:"
drivers? The only way the system knows which drivers to search is by the
scheme specification in the URL. Nothing at all even the least bit slightly
redundant about it at all even the least bit.
It is the scheme that tells the system which drivers to search.
--
Lew