search

bbcf / gdv

Genome Data Viewer
http://bbcf.epfl.ch/twiki/bin/view/BBCF/GDV
1 stars 1 forks source link

DNA text files not rendered by the browser #12

Open yjarosz opened 13 years ago

yjarosz commented 13 years ago

The DNA track is not visible from GDV : (DNA track rendering works with text files that have to be imported by the browser)

two solutions : 1) add another path in the javascript like we did for 'styleRoot' and take text files from JBrowsoR

2) replace JBrowsoR and create sequences directly from GDV

fabdavid commented 13 years ago

Jbrowsor is sending back the right relative path to get the sequences in browserRoot. You need to store the address of the Jbrowsor server somewhere in your project anyway.

On May 9, 2011, at 10:49 AM, yjarosz wrote:

The DNA track is not visible from GDV : (DNA track rendering works with text files that have to be imported by the browser)

  • cause : the browser root of JBrowse search for text files in GDV directories - and text files are from JBrowsoR => no matches

two solutions : 1) add another path in the javascript like we did for 'styleRoot' and take text files from JBrowsoR

2) replace JBrowsoR and create sequences directly from GDV

Reply to this email directly or view it on GitHub: https://github.com/bbcf/gdv/issues/12

yjarosz commented 13 years ago

yes the path send back by jbrowsor is so like : /data/seq/3/1637.txt and my browserRoot is : ../tracks so the final path where jbrowse want to fetch DNA is ../tracks/data/seq/3/1637.txt which is not a good path because under tracks, I access the tracks made by GDV and there is no data directory or symlink to jbrowsor data (which is not possible to do if GDV is not on the same server as jbrowsor) so the solution 1) is to make another path 'DNARoot' that will tell GDV (or apache) to fetch data from jbrowsoR server.

P.S: I think it will not be possible to load it directly from javascript because jbrowsor can be on a different server, and javascript cannot load data from other domain.

fabdavid commented 13 years ago

Yes, exactly, we have to hardcode somewhere the url of the jbrowsor server. And send it to the javascript. If it doesn't work, I suggest to find out a solution that makes jbrowsor evolve.

On May 9, 2011, at 11:17 AM, yjarosz wrote:

yes the path send back by jbrowsor is so like : /data/seq/3/1637.txt and my browserRoot is : ../tracks so the final path where jbrowse want to fetch DNA is ../tracks/data/seq/3/1637.txt which is not a good path because under tracks, I access the tracks made by GDV and there is no data directory or symlink to jbrowsor data (which is not possible to do if GDV is not on the same server as jbrowsor) so the solution 1) is to make another path 'DNARoot' that will tell GDV (or apache) to fetch data from jbrowsoR server.

P.S: I think it will not be possible to load it directly from javascript because jbrowsor can be on a different server, and javascript cannot load data from other domain.

Reply to this email directly or view it on GitHub: https://github.com/bbcf/gdv/issues/12#comment_1121931