Backslash Escape Sequences

[Andy Anderson-3]

On the list pgsql-general, I had this exchange:

>>> Andy Anderson <aanderson@...> writes:
>>>> In a test I just did, the sequence \" (backslash double-quote) is  
>>>> interpreted as just a " inside of the E'...' string constant  
>>>> expression. This is great, since PHP's addslashes() sticks them  
>>>> in along with the other stuff I really need to quote like ' and  
>>>> \. But I see that \" isn't documented in the manual in section  
>>>> 4.1.2.1.  I assume this is, in fact, standard behavior for  
>>>> Postgres? (Perhaps a comprehensive table might be a good idea at  
>>>> this point in the manual.)
>>
>> Tom Lane wrote:
>>> Huh?  It says
>>>
>>>   Any other character following a backslash is taken literally.
>>>
>>>
>
>  Andy Anderson wrote:
>> Ah, slight ambiguity here. Perhaps this might best say "Any other  
>> character following a backslash is taken literally, and the  
>> backslash is removed."

In any case, here's a contribution to the manual, a short table with  
this information, in a format that might help make the subject  
clearer. Modify at will!

-- Andy

<!--<TABLE BORDER="1" CLASS="CALSTABLE"> <COL><COL><COL> <THEAD> <TR>  
<TH>Backslash Escape Sequence</TH> <TH>Interpretation</TH> <TH>Notes</
TH> </TR> </THEAD> <TBODY> <TR> <TD><TT CLASS="TOKEN">\b</TT></TD>  
<TD>backspace</TD> <TD></TD> </TR> <TR> <TD><TT CLASS="TOKEN">\f</TT></
TD> <TD>form feed</TD> <TD></TD> </TR> <TR> <TD><TT CLASS="TOKEN">\n</
TT></TD> <TD>newline</TD> <TD></TD> </TR> <TR> <TD><TT CLASS="TOKEN">
\r</TT></TD> <TD>carriage return</TD> <TD></TD> </TR> <TR> <TD><TT  
CLASS="TOKEN">\t</TT></TD> <TD>tab</TD> <TD></TD> </TR> <TR> <TD> <TT  
CLASS="TOKEN">\<var>o</var></TT><br> <TT CLASS="TOKEN">\<var>oo</var></
TT><br> <TT CLASS="TOKEN">\<var>ooo</var></TT><br> </TD> <TD> octal  
byte value </TD> <TD> <var>o</var> = 0 - 7 </TD> </TR> <TR> <TD> <TT  
CLASS="TOKEN">\x<var>h</var></TT><br> <TT CLASS="TOKEN">\x<var>hh</
var></TT><br> </TD> <TD> hexadecimal byte value </TD> <TD> <var>h</
var> = 0 - F </TD> </TR> <TR> <TD><TT CLASS="TOKEN">\</TT><var>a</
var></TD> <TD><var>a</var></TD> <TD><var>a</var> = any other  
character. Commonly used to include \ and '</TD> </TR> </TBODY> </
TABLE>-->

--
Sent via pgsql-docs mailing list (pgsql-docs@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-docs

[Alvaro Herrera-7]

Andy Anderson wrote:

> In any case, here's a contribution to the manual, a short table with  
> this information, in a format that might help make the subject clearer.
> Modify at will!

FWIW if you really want to make a contribution to the docs, please see
the SGML sources instead of the HTML output (which is machine-generated).

http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/doc/src/sgml/

--
Alvaro Herrera                                http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support

--
Sent via pgsql-docs mailing list (pgsql-docs@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-docs

[Andy Anderson-3]

>> Andy Anderson wrote:
>>
>>> In any case, here's a contribution to the manual, a short table with
>>> this information, in a format that might help make the subject  
>>> clearer.
>> Modify at will!

> On Apr 30, 2008, at 6:04 PM, Alvaro Herrera wrote:
>> FWIW if you really want to make a contribution to the docs, please  
>> see
>> the SGML sources instead of the HTML output (which is machine-
>> generated).

OK, then, I hope the following is accurate enough SGML to be more  
useful. Submitted here per the instructions on <http://www.postgresql.org/docs/8.3/interactive/bug-reporting.html 
 >.

-- Andy

********************************************************************

<table id="sql-backslash-table">
        <title>Backslash Escape Sequences</title>
        <tgroup cols="2">
                <thead>
                        <row>
                                <entry>Backslash Escape Sequence</>
                                <entry>Interpretation</entry>
                        </row>
                </thead>
                <tbody>
                        <row>
                                <entry><literal>\b</literal></entry>
                                <entry>backspace</entry>
                        </row>
                        <row>
                                <entry><literal>\f</literal></entry>
                                <entry>form feed</entry>
                        </row>
                        <row>
                                <entry><literal>\n</literal></entry>
                                <entry>newline</entry>
                        </row>
                        <row>
                                <entry><literal>\r</literal></entry>
                                <entry>carriage return</entry>
                        </row>
                        <row>
                                <entry><literal>\t</literal></entry>
                                <entry>tab</entry>
                        </row>
                        <row>
                                <entry>
                                        <literallayout>
                                                <literal>\<replaceable>o</replaceable></literal>
                                                <literal>\<replaceable>oo</replaceable></literal>
                                                <literal>\<replaceable>ooo</replaceable></literal>
                                                (<replaceable>o</replaceable> = 0 - 7)
                                        </literallayout>
                                </entry>
                                <entry>octal byte value</entry>
                        </row>
                        <row>
                                <entry>
                                        <literallayout>
                                                <literal>\x<replaceable>h</replaceable></literal>
                                                <literal>\x<replaceable>hh</replaceable></literal>
                                                (<replaceable>h</replaceable> = 0 - F)
                                        </literallayout>
                                </entry>
                                <entry>hexadecimal byte value</entry>
                        </row>
                        <row>
                                <entry>
                                        <literallayout>
                                                <literal>\<replaceable>a</replaceable></literal>
                                                (<replaceable>a</replaceable> = any other character than the  
above,
                                                e.g. <literal>\</literal> and <literal>'</literal>.)
                                        </literallayout>
                                </entry>
                                <entry><replaceable>a</replaceable></entry>
                        </row>
                </tbody>
     </tgroup>
</table>

--
Sent via pgsql-docs mailing list (pgsql-docs@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-docs

[Bruce Momjian-5]

Andy Anderson wrote:

> >> Andy Anderson wrote:
> >>
> >>> In any case, here's a contribution to the manual, a short table with
> >>> this information, in a format that might help make the subject  
> >>> clearer.
> >> Modify at will!
>
> > On Apr 30, 2008, at 6:04 PM, Alvaro Herrera wrote:
> >> FWIW if you really want to make a contribution to the docs, please  
> >> see
> >> the SGML sources instead of the HTML output (which is machine-
> >> generated).
>
> OK, then, I hope the following is accurate enough SGML to be more  
> useful. Submitted here per the instructions on <http://www.postgresql.org/docs/8.3/interactive/bug-reporting.html 

Good idea --- a table is certainly clearer.  I modified your patch and
applied the attached version.  Thanks much.

Your documentation changes can be viewed in five minutes at:

        http://developer.postgresql.org/pgdocs/postgres/index.html

--
  Bruce Momjian  <bruce@...>        http://momjian.us
  EnterpriseDB                             http://enterprisedb.com

  + If your life is a hard drive, Christ can be your backup. +

Index: doc/src/sgml/syntax.sgml
===================================================================
RCS file: /cvsroot/pgsql/doc/src/sgml/syntax.sgml,v
retrieving revision 1.122
diff -c -c -r1.122 syntax.sgml
*** doc/src/sgml/syntax.sgml 20 Mar 2008 21:42:47 -0000 1.122
--- doc/src/sgml/syntax.sgml 26 Jun 2008 22:24:11 -0000
***************
*** 286,304 ****
       quote.)
       Within an escape string, a backslash character (<literal>\</>) begins a
       C-like <firstterm>backslash escape</> sequence, in which the combination
!      of backslash and following character(s) represents a special byte value.
!      <literal>\b</literal> is a backspace,
!      <literal>\f</literal> is a form feed,
!      <literal>\n</literal> is a newline,
!      <literal>\r</literal> is a carriage return,
!      <literal>\t</literal> is a tab.
!      Also supported are
!      <literal>\<replaceable>digits</replaceable></literal>, where
!      <replaceable>digits</replaceable> represents an octal byte value, and
!      <literal>\x<replaceable>hexdigits</replaceable></literal>, where
!      <replaceable>hexdigits</replaceable> represents a hexadecimal byte value.
!      (It is your responsibility that the byte sequences you create are
!      valid characters in the server character set encoding.) Any other
       character following a backslash is taken literally. Thus, to
       include a backslash character, write two backslashes (<literal>\\</>).
       Also, a single quote can be included in an escape string by writing
--- 286,348 ----
       quote.)
       Within an escape string, a backslash character (<literal>\</>) begins a
       C-like <firstterm>backslash escape</> sequence, in which the combination
!      of backslash and following character(s) represent a special byte
!      value:
!
!      <table id="sql-backslash-table">
!       <title>Backslash Escape Sequences</title>
!       <tgroup cols="2">
!       <thead>
!        <row>
!         <entry>Backslash Escape Sequence</>
!         <entry>Interpretation</entry>
!        </row>
!       </thead>
!
!       <tbody>
!        <row>
!         <entry><literal>\b</literal></entry>
!         <entry>backspace</entry>
!        </row>
!        <row>
!         <entry><literal>\f</literal></entry>
!         <entry>form feed</entry>
!        </row>
!        <row>
!         <entry><literal>\n</literal></entry>
!         <entry>newline</entry>
!        </row>
!        <row>
!         <entry><literal>\r</literal></entry>
!         <entry>carriage return</entry>
!        </row>
!        <row>
!         <entry><literal>\t</literal></entry>
!         <entry>tab</entry>
!        </row>
!        <row>
!         <entry>
!          <literal>\<replaceable>o</replaceable></literal>,
!          <literal>\<replaceable>oo</replaceable></literal>,
!          <literal>\<replaceable>ooo</replaceable></literal>
!          (<replaceable>o</replaceable> = 0 - 7)
!         </entry>
!         <entry>octal byte value</entry>
!        </row>
!        <row>
!         <entry>
!          <literal>\x<replaceable>h</replaceable></literal>,
!          <literal>\x<replaceable>hh</replaceable></literal>
!          (<replaceable>h</replaceable> = 0 - 9, A - F)
!         </entry>
!         <entry>hexadecimal byte value</entry>
!        </row>
!       </tbody>
!       </tgroup>
!      </table>
!
!      It is your responsibility that the byte sequences you create are
!      valid characters in the server character set encoding. Any other
       character following a backslash is taken literally. Thus, to
       include a backslash character, write two backslashes (<literal>\\</>).
       Also, a single quote can be included in an escape string by writing


--
Sent via pgsql-docs mailing list (pgsql-docs@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-docs

[Alvaro Herrera-7]

Bruce Momjian wrote:

> > OK, then, I hope the following is accurate enough SGML to be more
> > useful. Submitted here per the instructions on
> > <http://www.postgresql.org/docs/8.3/interactive/bug-reporting.html 
>
> Good idea --- a table is certainly clearer.  I modified your patch and
> applied the attached version.  Thanks much.

Hmm, I think the \\ and \' stuff should be part of that table too -- the
\<replaceable>a</> as in the patch seemed fine.

--
Alvaro Herrera                                http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.

--
Sent via pgsql-docs mailing list (pgsql-docs@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-docs