"Unable to retrieve datasets" error for Static Website server method
jwtobias opened this issue · 9 comments
I am trying to use your instructions for making a static website deployment on an Ubuntu 20.04 server. When I visit my site, I see the cirrocumulus home page, but I get an "Unable to retrieve datasets" banner in the lower left (screenshot attached).
I think I have the files I need where they should be, but here are a look into my datasets.json file and the referenced dataset that was produced by cirro prepare_data:
(base) jtobias@WS8:~/cirrocumulus/build$ cat datasets.json
[
{
"id": "hum_cbn_forCirro",
"name": "hum_cbn_forCirro",
"url": "hum_cbn_forCirro/hum_cbn_forCirro.jsonl"
}
]
(base) jtobias@WS8:~/cirrocumulus/build$ ls hum_cbn_forCirro/
hum_cbn_forCirro.jsonl hum_cbn_forCirro.jsonl.idx.json uns
That's a little tricky for me for this headless server. Let me see if I can do something like that via Xwindows.
I have discovered a solution... If I copy the the datasets.json file to the directory above build (my web root), this seems to solve the problem.
The script seems not to be looking for that file in the build directory itself.
This will fix the issue:
mv static/js/main.c57698e7.js temp
cat temp | sed 's/\/datasets/datasets/g' > static/js/main.c57698e7.js