Speaking of viewing URLs in browsers, is it possible to view them in the
windows version of Artemis? I couldn't find 'results_to_netscape' in
etc/, and even though I copyied it there and enabled sanger_options I
never got the 'Send to browser' option. Do you know how to do it?
Regards,
Marcus
On Mon, 2003-09-15 at 12:21, Marcus Claesson wrote:
> Hi all,
>
> I started using the 'results_to_netscape' function that enables web
> browsing of results from the external artemis tools (from the run menu),
> but couldn't get it to work fully. The URLs to the SRS server had to be
> updated and manipulated further in a way the made the
> 'results_to_netscape' quite complicated. Instead I chose to use Entrez
> retrieval system at NCBI. For those interested I attach the changed
> 'results_to_netscape' code here:
>
> #!/bin/sh -
>
> # This script will send a file to netscape with database IDs linked to
> ENTREZ
> # To enable this in Artemis, sanger_options must be set to true in the
> options file.
>
> NETSCAPE=/usr/local/bin/netscape
>
>
> if [ -f "$DIANA_ENVIRONMENT_FILE" ]
> then
> . $DIANA_ENVIRONMENT_FILE
> fi
>
> if [ $# = 0 ]
> then
> echo no argument given 1>&2
> exit 1
> fi
>
> file_arg=$1
>
> unique_bit=$$.`hostname`
>
> if [ -f ./$file_arg ]
> then
> # the file is in the current directory - we need the full path so
> netscape
> # can find the file
> new_file=$PWD/$file_arg.$unique_bit.html
> else
> new_file=$file_arg.$unique_bit.html
> fi
>
> cat <<EOF > $new_file
> <HTML>
> <HEAD>
> <TITLE>
> Results for $file_arg
> </TITLE>
> </HEAD>
> <BODY>
> <PRE>
> EOF
>
> perl -e '
> BEGIN {
>
> $BLAST_START_LINE = "Sequences producing High-scoring Segment Pairs|"
> .
> "Sequences producing significant alignments:";
> $FASTA_START_LINE = "The best scores are";
>
> # the list of IDs we have seen so far
> @ids = ();
>
> # the list of IDs we have made anchors for so far
> @anchored_ids = ();
>
> }
>
> sub hyperlink_to_anchor
> {
> $id = shift;
> qq(<a href="#$id">$id</a>);
> }
>
> sub hyperlink_id
> {
> $id = shift;
> $r = qq#<A TARGET=_blank
> HREF=\"http:\/\/www.ncbi.nlm.nih.gov\/entrez\/viewer.fcgi?val=$id\">$id<\/A>#;
> return $r
> }
>
> $file_name = $ARGV[0];
>
> if ($file_name =~ /\.gz$/) {
> open IN_FILE, "gzip -d < $file_name |" or die "failed to open
> $file_name\n";
> } else {
> open IN_FILE, "$file_name" or die "failed to open $file_name\n";
> }
>
> while (<IN_FILE>) {
>
> s/(http\:\/\/([\S.\/]+))/<a href=\"$1\">$1<\/a>/gi;
>
> # ignore header lines
> if (1..m/$BLAST_START_LINE|$FASTA_START_LINE/) {
> print;
> next;
> }
>
> if (@ids && /^\s*$/) {
> $summary_finished = 1;
> }
>
> if (/\w+\|([^|]+)\|/) {
> $id = $1;
> }
>
> if ($summary_finished) {
> if ((grep {$_ eq $id} @ids) && (!grep {$_ eq $id} @anchored_ids))
> {
> # not anchored yet so make it an anchor
> if (s/\b$id\b/"<a name=\"$id\">" . (hyperlink_id($id)) .
> "<\/a>"/e) {
> push @anchored_ids, $id;
> }
> }
> } else {
> if (!grep {$_ eq $id} @ids) {
> push @ids, $id;
> }
>
> s/$id/hyperlink_to_anchor($id)/ei;
>
> if (!s/ $id/" " . hyperlink_id($id)/ei) {
> # if the id occurs once in the line put a link at end of line
> s/$/" NCBI:" . hyperlink_id($id)/e;
> }
> }
> print;
> }
>
> ' $file_arg >> $new_file
>
> cat <<EOF >> $new_file;
> </PRE>
> </BODY>
> </HTML>
> EOF
>
> # delete it at some point
> echo "rm -f $new_file " | at now + 36 hours
>
> if $NETSCAPE -remote "openURL($new_file)"
> then
> exit 0
> else
> echo starting new netscape
> # netscape isn't running - so start it
> ($NETSCAPE &)&
>
> # now send the URL. we do things this way so that the script
> doesn't exit
> # until netscape has successfully shown the URL
>
> sleep 1
>
> # don't exit the script until the file is successfully displayed or
> until
> # 40 seconds is up
> for i in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
> do
> if $NETSCAPE -remote "openURL($new_file)"
> then
> exit 0
> else
> sleep 2
> fi
> done
>
> exit 1
> fi
>
>
