Opened 15 years ago
Closed 15 years ago
#220 closed defect (fixed)
MeV doesn't download files correctly
Reported by: | Nicklas Nordborg | Owned by: | Nicklas Nordborg |
---|---|---|---|
Priority: | blocker | Milestone: | MEV Launcher v1.3 |
Component: | net.sf.basedb.mev | Keywords: | |
Cc: |
Description
There is a strange behaviour with MeV. It has some kind of cache for downloaded (TDMS and CGH) files, but once there is a file of a certain type (TDMS or CGH) the same file is used also for other datasets. For example, if we have an experiment with two bioassay sets and I create a TDMS file with the export plug-in for each one of those. Then, I start MeV with the TDMS file from the first bioassay set and everything is good, but when I start MeV with the TDMS file from the second bioassay set MeV doesn't download the other TDMS file, it uses the TDMS file from the first bioassay set already in the cache.
One theory is that the last part of our file names are not unique, a typical download url is: http://localhost:8080/2.12-stable/extensions/mev-launcher.jar/MevExport/8fa27ade945e4dc9/35/mev.tdms.servlet
where 8fa...
is the session id and 35
is the bioassay set id. The part after the last / is the same for all TDMS files and it may be that the cache in MeV only uses this as a key.
(In [1098]) Fixes #220: MeV doesn't download files correctly
Added the file id and last update time to the exported filename. This seems make the cache work as expected and cause a re-download if the file has been updated.