AcePerl
view release on metacpan or search on metacpan
will not ordinarily have to call this method.
=back
=head1 BUGS
1. The ACE model should be consulted prior to updating the database.
2. There is no automatic recovery from connection errors.
3. Debugging has only one level of verbosity, despite the best
of intentions.
4. Performance is poor when fetching big objects, because of
many object references that must be created. This could be
improved.
5. When called in an array context at("tag[0]") should return the
current tag's entire column. It returns the current subtree instead.
6. There is no way to add comments to objects.
Ace/Object.pm view on Meta::CPAN
the subtree rooted at the tag. In the current version, an implicit
right() and dereference is performed.
=head2 fetch() method
$new_object = $object->fetch;
$new_object = $object->fetch($tag);
Follow object into the database, returning a new object. This is
the best way to follow object references. For example:
$laboratory = $object->at('Laboratory')->fetch;
print $laboratory->asString;
Because the previous example is a frequent idiom, the optional $tag
argument allows you to combine the two operations into a single one:
$laboratory = $object->fetch('Laboratory');
=head2 follow() method
Ace/Object.pm view on Meta::CPAN
sub kill {
my $self = shift;
return unless my $db = $self->db;
return 1 unless $db->count($self->class,$self->name);
my $result = $db->raw_query("kill");
if (defined($result) and $result=~/write access/im) { # this keeps changing
$Ace::Error = "Write access denied";
return;
}
# uncache cached values and clear the object out
# as best we can
delete @{$self}{qw[.PATHS .right .raw .down]};
1;
}
# sub isTimestamp {
# my $self = shift;
# return 1 if $self->class eq 'UserSession';
# return;
# }
Ace/Sequence/Homol.pm view on Meta::CPAN
=head1 SYNOPSIS
# Get all similarity features from an Ace::Sequence
@homol = $seq->features('Similarity');
# sort by score
@sorted = sort { $a->score <=> $b->score } @homol;
# the last one has the highest score
$best = $sorted[$#sorted];
# fetch its associated Ace::Sequence::Homol
$homol = $best->target;
# print out the sequence name, DNA, start and end
print $homol->name,' ',$homol->start,'-',$homol->end,"\n";
print $homol->asDNA;
=head1 DESCRIPTION
I<Ace::Sequence::Homol> is a subclass of L<Ace::Object> (B<not>
L<Ace::Sequence>) which is specialized for returning information about
a DNA or protein homology. This is a temporary placeholder for a more
README.ACEBROWSER view on Meta::CPAN
database, but most of it is fully generic.
Demos are running at http://stein.cshl.org/elegans/.
REQUIREMENTS:
1. AcePerl 1.76 or higher (available at http://stein.cshl.org/AcePerl/)
2. Perl 5.6.0 or higher
3. CGI.pm 2.77 or higher (available at http://stein.cshl.org/WWW/software/CGI)
4. A Web server
5. sgifaceserver 4.8c or higher. For best results, use the version of
sgifaceserver available at http://www.acedb.org/.
The socket server is generally a better choice than the older RPC-based
server.
INSTALLATION:
1. Read README first. This describes how to install AcePerl.
2. During the installation, you will be asked whether you wish to
install AceBrowser. Answer "yes."
acelib/messubs.c view on Meta::CPAN
#endif
return mess ;
}
/************************* message formatting ********************************/
/* This routine does the formatting of the message string using vsprintf, */
/* it copes with the format string accidentally being our internal buffer. */
/* */
/* This routine does its best to check that the vsprintf is successful, if */
/* not the routine bombs out with an error message. Note that num_bytes is */
/* the return value from vsprintf. */
/* Failures trapped: */
/* num_bytes < 0 => vsprintf failed, reason is reported. */
/* num_bytes + 1 > BUFSIZE => our internal buffer size was exceeded. */
/* (vsprintf returns number of bytes written */
/* _minus_ terminating NULL) */
/* */
static char *uMessFormat(va_list args, char *format, char *prefix,
char *buffer, unsigned int buflen)
acelib/wh/version.h view on Meta::CPAN
/* The function must have this prototype and must return a string that gives */
/* the build date: */
/* */
char *utAppGetCompileDate(void) ;
/* The acedb */
/* makefile is arranged so that the main routine is recompiled every time */
/* the application is relinked. This means that the date represents the */
/* 'build' date of the application. */
/* */
/* Code the macro by simply putting it in the .c file that contains the */
/* main function of the application, it's probably best to put it just */
/* before or after the main function. Do not put a terminating ';' after */
/* the macro, this will cause a compile error. */
/* */
#define UT_COMPILE_PHRASE "compiled on:"
#define UT_MAKE_GETCOMPILEDATEROUTINE() \
char *utAppGetCompileDate(void) { return UT_COMPILE_PHRASE " " __DATE__ " " __TIME__ ; }
docs/ACE_SERVER_TRAPS.HOWTO view on Meta::CPAN
Problem
Most documentation dealing with this software simply tells you to get
it going. The software comes with no documentation whatsoever. No
README at all. There is a manual that comes in /acedocs called
aceserver.html. Its installation instructions don't work on version
4.7g.
Solution
The best installation information is in the README file for
AcePerl-1.54. A few more hints are listed here.
The AcePerl README file implies creating a user called acedb. This
creates permissions problems that we haven't solved yet. We are using
individual user names instead.
One thing that no documentation mentions is that we had to move
gifaceserver.LINUX to /usr/local/bin/gifaceserver. Obvious, but still
makes you wonder while you do it.
docs/ACE_SERVER_TRAPS.HOWTO view on Meta::CPAN
whom the gifaceserver is assigned in the inetd.conf file. We created
one by opening the gifaceserver on a fake port number (12345):
/usr/local/bin/gifaceserver /home/httpd/database/contacts 12345
1200:1200:10
Models
Documentation
The best documentation for models is in /acedocs/exploring/*. The
table of contents is in /acedocs/exploring/toc_models.html.
Unfortunately, like all the ACEDB documentation, it uses absolute
pathnames. We have converted these pathnames to relative ones, and
will make the document available for download on the Spatial Focus
private web page. Although the document is marked "draft" and dated
1994, it is thorough and simple. Doesn't appear to be significantly
out of date.
The moviedb database is the best simple example of a database.
Editors
ACEDB is picky about its ascii. vi works great. Can't vouch for emacs
;-). Don't use anything nasty like a word processor.
White Space
It really likes alignment, and it likes tabs. Combining tabs and
spaces kills otherwise perfectly good models every five seconds.
docs/ACE_SERVER_TRAPS.HOWTO.html view on Meta::CPAN
<p><h5>Consequences of the Solution</h5></p>
<p>The <strong>NOTES</strong> file distributed with this version of ACEDB advises you to move the <em>acedb</em> and <em>textace</em> scripts to /usr/local/bin, and using them to start the program. With your environment variables in place, you can s...
<p><h4>Permissions</h4></p>
<p>Take your permissions seriously. Richard Durbin's Installation Guide is out of date, but gives good advice in this department.</p>
<p><h3>Gifaceserver<br>Warning: this isn't yet working completely</h3></p>
<p><h4>Problem</h4></p>
<p>Most documentation dealing with this software simply tells you to get it going. The software comes with no documentation whatsoever. No README at all. There is a manual that comes in /acedocs called aceserver.html. Its installation instruction...
<p><h4>Solution</h4></p>
<p>The best installation information is in the <strong>README</strong> file for <em>AcePerl-1.54</em>. A few more hints are listed here.</p>
<p>The <em>AcePerl</em> <strong>README</strong> file implies creating a user called <em>acedb</em>. This creates permissions problems that we haven't solved yet. We are using individual user names instead.</p>
<p>One thing that <strong>no</strong> documentation mentions is that we had to move <em>gifaceserver.LINUX</em> to <em>/usr/local/bin/gifaceserver</em>. Obvious, but still makes you wonder while you do it.</p>
<p><h4>Inetd.conf</h4></p>
<p>Our individual <em>inetd.conf</em> files were completely commented out, and the daemon stopped because of our dispersed locations. Append the required line to the file, and enter:</p>
<p><em>Killall -HUP inetd</em></p>
<p><h4>server.log</h4></p>
<p>The server really wants a <em>server.log</em> file, writable by the user to whom the <em>gifaceserver</em> is assigned in the <em>inetd.conf</em> file. We created one by opening the <em>gifaceserver</em> on a fake port number (12345):</p>
<p><em> /usr/local/bin/gifaceserver /home/httpd/database/contacts 12345 1200:1200:10</em></p>
<p><h2>Models</h2></p>
<p><h3>Documentation</h3></p>
The <strong>best</strong> documentation for models is in <em>/acedocs/exploring/*.</em> The table of contents is in <em>/acedocs/exploring/toc_models.html</em>. Unfortunately, like all the ACEDB documentation, it uses absolute pathnames. We have c...
<p>The moviedb database is the best simple example of a database.</p>
<p><h3>Editors</h3></p>
<p>ACEDB is picky about its ascii. <em>vi</em> works great. Can't vouch for <em>emacs</em> ;-). <strong>Don't</strong> use anything nasty like a word processor.</p>
<p><h3>White Space</h3></p>
<p>It really likes alignment, and it likes tabs. Combining tabs and spaces kills otherwise perfectly good models every five seconds.</p>
<p><h4>To Do</h4></p>
Solve the mysteries of the failure of AceBrowser. Every other means of access works now.</p>
</body>
</html>
( run in 0.856 second using v1.01-cache-2.11-cpan-4e96b696675 )