Sophie: maradns-1.4.13-2.mga4 i586

maradns-1.4.13-2.mga4.i586.rpm

<!-- Do *not* edit this file; it was automatically generated by ej2html
     Look for a name.ej file with the same name as this filename -->
<!-- Last updated Mon Mar 24 09:52:33 2008 -->

<HTML><HEAD>



<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=utf-8">
</HEAD><BODY >

<h1>NAME</h1>
csv2_txt - Description of txt and raw resource records in the csv2 zone file
<h1>DESCRIPTION</h1>

Due to the complexity of TXT and RAW records, this man page is
dedicated to describing the csv2 format of this RR.

<p>

TXT and RAW rrs in MaraDNS' csv2 zone files can store any arbitrary 
binary data.  Additionally, it is possible to arbitrarily divide up
TXT records in to chunks (chunks, which RFC1035 call "character-string"s,
are described below).

<h2>ASCII AND UTF-8 DATA</h2>

If a given TXT field or RAW record contains only ASCII data, creating a
record is easy: Place the full data between single quotes, like
this:

<pre>
a.example.com. TXT 'This is some text' ~
</pre>

It is also possible, to place almost any printable ASCII characters 
between quotes.  The '~' (tilde) character is not allowed unless 
csv2_tilde_handling has a value of 0; the '|' (pipe), '#' (hash) and
non-printable ASCII control characters are not allowed in TXT data
if the ~ is used to separate records.  If there are any bytes
with a value of 0x80 or more, the data must be UTF-8 encoded Unicode.

For example:

<pre>
b.example.com. TXT 'This is an example UTF-8 character: I ♥ MaraDNS'
</pre>
 
(If your font does not have this Unicode symbol, it is a heart)


<p>

The printable ASCII characters not allowed in quotes are the ' character,
the '|' character, the '~' (tilde) character, and the '#' character.
See BACKSLASH ESCAPE SEQUENCES below for information on adding these
characters to TXT or RAW fields.

<h2>UNQUOTED DATA</h2>

Note that the record does not have to be quoted.  As long as the
record only contains ASCII alphanumeric data, and/or the characters
'-', '_', '+', '%', '!', '^', and '=', the data can be unquoted
as follows:

<pre>
c.example.com. TXT This_is_100%_unquoted_text_+symbols!
</pre>

It is also possible to mix quoted and unquoted text, such as this:

<pre>
d.example.com. TXT This' is a mix 'of_unquoted' and quoted 'text!
</pre>

Which will have its data look like this:

<pre>
This is a mix of_unquoted and quoted text!
</pre>

When mixing quoted and unquoted data, it is important to have all whitespace
<i>inside</i> quotes.

<h2>BACKSLASH ESCAPE SEQUENCES</h2>

In order to accommodate storing non-UTF-8 high bit characters, the
single quote character, non-printable ASCII control codes, the '|',
'~', and '#' characters, and to permit multi-line TXT/RAW records
(with comments allowed mid-record), the TXT/RAW RR allows backslashes.
These backslashes only have significance <i>outside</i> of quoted text;
if they are placed inside single quotes, they are not interpreted and
result in a literal backslash being added to the resource record data.

<p>

The following characters can be backslashed:

<dl>
<dt>
'
<dd>
When backslashed, the adds a literal quote to the resource record.
<dt>
whitespace
<dd>
When any whitespace is backslashed (space, newline, cr, and tab), this
indicates that the record has not ended, and that more data for this
resource will follow.  This also allows comments to be placed in 
TXT and RAW resource records.  What happens is that the backslash
indicates that any whitespace characters (space, tab, carriage return,
and line feed) are to be ignored until
the next non-whitespace character that is not a # (hash).  If a
# is seen, this indicates that we ignore any and all characters until the
next carriage return or line feed, and continue to ignore everything
until the next non-whitespace character.  
See the section on multi-line and
commented records for examples.
<dt>
0123
<dd>
When a number between 0 and 3 is backslashed, this indicates the 
beginning of a three-digit octal number.
<dt>
x
<dd>
When an x is backslashed, this indicates the beginning of a two-digit
hexadecimal number.
</dl>

Note that, with the exception of the single quote, the backslash character
is <i>not</i> used to remove the meta-significance of a given character.
In particular, unlike other environments, it is not possible to backslash
spaces.  Spaces can be represented either as ' ' in quotes, \x20, or 
as \040.

<p>

Here are some examples of backslashed data.

In this example, we see backslash sequences being used to store non-UTF-8
hi-bit data:

<pre>
e.example.com. TXT \x80\x81\x82\x83 ~
</pre>

This same data can also be created as follows:

<pre>
f.example.com. TXT \200\201\202\203 ~
</pre>

Octal and hex information can be mixed:

<pre>
g.example.com. TXT \200\x81\202\x83 ~
</pre>

Literal single quotes can be placed in resource records:

<pre>
h.example.com. TXT 'perl -e '\''print "A Perl of a TXT record!\n"'\' ~
</pre>

The above example produces this record:

<pre>
perl -e 'print "A Perl of a TXT record!\n"' ~
</pre>

To render the '~' character, use the escape sequence \x7e (outside of
quotes).  For example:

<pre>
h1.example.com. TXT 'http://ocf.berkeley.edu/'\x7e'set' ~
</pre>

Produces this record:

<pre>
http://ocf.berkeley.edu/~set
</pre>

To render the '|' character, use the escape sequence \x7c:

<pre>
h2.example.com. TXT 'ls '\x7c' more' ~
</pre>

Produces this record:

<pre>
ls | more
</pre>

To render the '#' character, use the escape sequence \x23:

<pre>
h3.example.com. TXT 'Press '\x23' for customer service' ~
</pre>

Produces this record:

<pre>
Press # for customer service
</pre>

<h2>MULTI-LINE AND COMMENTED RECORDS</h2>

By utilizing backslashes followed by comments, it is possible to have
multi-line and commented TXT and RAW records.  The following resource
record will span more than one line on an 80-column display:

<pre>
i.example.com. TXT 'Not only did the quick brown fox jump over the lazy dog, but the lazy dog jumped over the cat.' ~
</pre>

Without affecting this resource record, the same data can be split over 
multiple lines:

<pre>
j.example.com. TXT 'Not only did the quick brown fox jump '\
                   'over the lazy dog, but the lazy dog'\
                   ' jumped over the cat.' ~
</pre>

Some points:

<ul>
<li>The backslash must be outsize of the quotes (or a literal backslash will
    be added to the record)
<li>The backslash must be present <i>before</i> any unquoted white space.  
    Usually, the backslash is placed immediately after the
    quote character.
<li>Unlike other environments, it does not matter whether or not there
    is invisible whitespace after the backslash.
</ul>

It is also possible to add comments after such a backslash as follows:

<pre>
k.example.com. TXT 'Not only did the quick brown fox jump '\ # The fox
                   'over the lazy dog, but the lazy dog'\    # The dog
                   ' jumped over the cat.' ~                 # The cat
</pre>

Note that, since the third comment is not preceded by a backslash, this
indicates the end of the resource record.  

<p>

There can also be multiple lines dedicated to 
comments (and, optionally, even blank lines) in the middle 
of TXT and RAW record data:

<pre>
k2.example.com. TXT 'This is some data '\
# Here we have some comments followed by a blank line

# Now we have some more comments, 
# followed by the rest of the data
    'and this is the rest of the data' ~
</pre>

<p>

<h2>MULTIPLE TXT CHUNKS</h2>

TXT RRs may be divided up in to multiple "chunks" (RFC1035 calls these
"character-string"s).  A single chunk can be anywhere from zero to 255
bytes long.  The default is to have one chunk, as follows:

<pre>
o.example.com. TXT 'TXT record with only one chunk' ~
</pre>

It is also possible to have a record with multiple chunks.  Chunks
are delimited by an unquoted ';' character:

<pre>
p.example.com. TXT 'This is chunk one';'This is chunk two' ~
</pre>

Or:

<pre>
q.example.com. TXT 'This is chunk one';\   # Our first chunk
                    This_is_chunk_two;\    # Our second chunk
                   'This is chunk three' ~ # Our final chunk
</pre>

Quoted ; characters simply add a ; to the record data.

<p>

If a single TXT chunk is longer than 255 bytes long, the csv2 parser will
report an error in the zone file: Single TXT chunk too long

<p>

In order to resolve this, place unquoted ; characters in the record data
so that each chunk is under 255 octets (bytes or characters) in length.

<p>

It is possible to have zero length chunks:

<pre>
r.example.com. TXT 'chunk one';;'chunk three' ~ # Chunk two zero-length
</pre>

In particular, is is possible to have zero length chunks at the
beginning and end of a TXT record:

<pre>
s.example.com. TXT ;'chunk two'; ~ # Chunks one and three zero-length
</pre>

Do not place semicolons at the beginning nor end of TXT records unless
you wish to have these zero-length chunks.

<p>

Chunk support only exists for TXT records.  An unquoted ; character will
cause a syntax error in a RAW record.

<h2>RAW RECORDS</h2>

With the exception of no support for chunk delimiters, and the addition of a
numeric record type before the record data, the format for RAW records is
identical to text records.  For example, if we wish to have a "Kitchen Sink"
RR record, which has the 8-bit binary numbers "16", "1", and "2", followed
by the ASCII string "Kitchen sink+ data", we can specify this in any of the
following manners:

<pre>
t1.example.com. RAW 40 \x10\x01\x02'Kitchen sink'\x2b' data' ~
</pre>

<pre>
t.example.com. RAW 40 \020\001\002Kitchen' sink+ data' ~
</pre>

<pre>
u.example.com. RAW 40 \x10\x01\x02Kitchen\x20sink+\x20data ~
</pre>

<pre>
v.example.com. RAW 40 \x10\001\x02\
                      'Kitchen sink+ data' ~
</pre>

<pre>
w.example.com. RAW 40 \x10\ # Meaning: 16
                      \x01\ # Coding: 1
                      \x02\ # Sub-coding: 2
                      'Kitchen sink+ data' ~ # Data: 'Kitchen sink+ data'
</pre>

<h1>LEGAL DISCLAIMER</h1>
THIS SOFTWARE IS PROVIDED BY THE AUTHORS ''AS IS'' AND ANY EXPRESS 
OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 
ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHORS OR CONTRIBUTORS BE 
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR 
BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, 
WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE 
OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, 
EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 

<h1>AUTHOR</h1>
Sam Trenholme 
<A href=http://www.samiam.org/>http://www.samiam.org/</a>
</BODY></HTML>