Re: Utl_file.

From: bill turner <510048981350-0001NoNoNo_at_t-online.de>
Date: Tue, 11 Feb 2003 07:31:58 +0100
Message-ID: <126h4vsrd4t08q5je7kgd1ei112p5epc10_at_4ax.com>


[Quoted] On Mon, 10 Feb 2003 08:40:24 -0800, DA Morgan <damorgan_at_exesolutions.com> wrote:

>bill wrote:
>
>> DA Morgan <damorgan_at_exesolutions.com> wrote in message news:<3E468856.4F52A011_at_exesolutions.com>...
>> > bill turner wrote:
>> >
>> > > On Sat, 08 Feb 2003 06:35:25 +0100, Sybrand Bakker
>> > > <gooiditweg_at_sybrandb.demon.nl> wrote:
>> > >
>> > > >On Fri, 07 Feb 2003 22:33:22 +0000, Alexxx12
>> > > ><member14441_at_dbforums.com> wrote:
>> > > >
>> > > >>
>> > > >>The path is correct. Why am I getting this error??
>> > > >
>> > > >
>> > > >Utl_file can't work with files on a client as it is running on the
>> > > >server
>> > > >
>> > > >
>> > > >Sybrand Bakker, Senior Oracle DBA
>> > > >
>> > > >To reply remove -verwijderdit from my e-mail address
>> > > I have a similar problem as the original poster. I did some research.
>> > > From what I read in the Oracle documenation, it is possible to use
>> > > UTL_FILE on the client side. Unless I completely misinterpreted what
>> > > was being said, you could use the funciton on both server and client
>> > > side. For the the server side, however, an entry was required in
>> > > init.ora (or some such thing). Since this is Sunday, and I do not have
>> > > the documentation with, I cannot quote from it. I did read this in
>> > > both the PL/SQL documentation and Supplied Products (?) documentation.
>> > > It was exactly for the reason to understand what Oracle meant by
>> > > client and server side or to get other input as to why I received the
>> > > "invalid path" error. I DO think I understand what is meant by client
>> > > side. So, assuming that I did understand the documentation properly,
>> > > does it mean that some package needs to be installed on the client?
>> > >
>> > > bill
>> >
>> > UTL_FILE is only server side. You can dump data to the client ... but not
>> > using UTL_FILE.
>> >
>> > Probably the only possible exception would be if you mapped the client to
>> > look like a drive on the server. Something I would not advise.
>> >
>> > Daniel Morgan
>>
>> Of course, I am likely running on less than full information. However,
>> as the quotes below show, it is understandable that I would be
>> confused.
>>
>> The following is a direct quote from the PL/SQL User's Guide and
>> Reference Release 8.1.5:
>>
>> "PL/SQL file I/O is available on both the client and server sides.
>> However, on the server side, file access is restricted to those
>> directories explicitly listed in the accessible directories list,
>> which is stored in the Oracle initialization file"
>>
>> The following quote is from Oracle8i Supplied Packages Reference
>> Release 8.1.5:
>>
>> "The PL/SQL file I/O feature is available for both client side and
>> server side PL/SQL. The client implementation (text I/O) is subject to
>> normal operating system file permission checking, and it does not need
>> any additional security constraints. However, the server
>> implementation might be running in a privileged mode, and will need
>> additional security restrictions that limit the power of this
>> feature."
>>
>> I am actually using 9i. This documentation was handy. Since it seems
>> as if I interpreted this incorrectly, what does it mean to have this
>> feature available for both client and server side?
>>
>> What is TEXT_IO? I could not find any information on that. I could I
>> use that? If not, how, then, can I write a file to the client (Win
>> 2000) rather than the server (Sun Solaris)?
>>
>> Bill
>
>Yes and no. PL/SQL I/O is not UTL_FILE. I'm not sure how one would assume that they were synonymous.
>
>TEXT_IO is a built-in package in Oracle Forms.
>
>Daniel Morgan

Okay, well, both of those quotes were in sections discussing UTL_FILE specifically, there was no indication that the topic was changed to some other form of output. Therefore, it was quite easy to assume they were synonymous. In fact, I still read it that way, though, undoubtedly you are correct. Rather, it is just poorly written documentation. Received on Tue Feb 11 2003 - 07:31:58 CET

Original text of this message