server-release.sh doesn't work for a single data set

[Esteban82]

@PaulWessel I run:

make server-release with mars_relief (as suggested by the text in the terminal).

$ make server-release
scripts/server-release.sh
server-release.sh: Enter a specific data set (e.g., mars_relief) or hit return for all data on the candidate: mars_relief 

In the first command of /tmp/release.sh, the names of both paths are wrong.

# 1. Issue rsync command
rsync -al /export/gmtserver/gmt/candidate/server/mars_relief /export/gmtserver/gmt/data/server/mars_relief

They have to include the name of the planet:

server/mars_relief         # wrong
server/mars/mars_relief    # OK

Solution

I think I can fix it.

I am thinking on editing the script. I can add a command that creates the variable {PLANET} from {DATASET}. Let me know if you are ok with this solution. Or let me know how would you like me to do it.

[PaulWessel]

Grep for PLANET and planet in scripts/*.sh. I have solved that elsewhere

[Esteban82]

This ? planet=$(echo ${dataset} | awk -F_ '{print $1}')

[Esteban82]

Is it possible that I have to add a "/" at the end of the first path here?

rsync -al /export/gmtserver/gmt/candidate/server/mars_relief /export/gmtserver/gmt/data/server/mars_relief

[anbj]

If you add a slash, rsync will copy the contents of the directory. No slash will copy the directory (with the content).

A slash basically means dir/*.

[Esteban82]

Ok. I think that explains why I got server/earth/earth_synbath/earth_synbath/*.

[anbj]

And I think adding a slash does make sense. Why would you want a directory named mars_relief inside a directory named mars_relief.. if you’re not into that.