callat-qcd / lattedb

Lattice QCD database interface using EspressoDB as the content manager.
https://ithems.lbl.gov/lattedb
BSD 3-Clause "New" or "Revised" License
1 stars 0 forks source link

ff4d tables #40

Closed walkloud closed 4 years ago

walkloud commented 4 years ago

We need the following tables to monitor data collection

1 - FormFac_4D 2 - FormFac_4D_tslice 3 - FormFac_4D_tslice_src_avg 4 - ConcatFormFac_4D (this is the that exists now)

3 has the same dictionary with configuration_range -> configuration and add at t_sep

2 has the same dictionary as 3, but add src (xXyYzZtT)

1 has the same dictionary as 2

walkloud commented 4 years ago

only 3 and 4 have a TAPE dictionary to accompany

ckoerber commented 4 years ago

Changes modulo adjustments have been implemented with 9e3cd878dfb41cdf35a85fba573c47b2afcc3934.

Important Note

Commit a9db882305cd2f6bb97886aa1ddcd9c15f6c5469 introduces the new table FormFactor4D which is not the Concatenated FormFactor4D table. However the Concatenated FormFactor4D table was previously named FormFactor4D table. This renaming was executed with previous commits. However in both SQLITE and POSTGRES still indices storing, e.g., user and formfactor4d relations where present and I had to manually remove them, e.g.,

DROP INDEX lattedb_project_formfac_fromfactor4dfile_user;

and similar commands. After this, migration worked without issues.

ckoerber commented 4 years ago

Since tables are getting quite large, I was wondering if one can still use jQuery DataTables in conjecture with a server side rendering.

Here is a lightweight Python package which promises to help: https://github.com/peppelinux/django-datatables-ajax and also I have read about the Django rest framework in some comments.

ckoerber commented 4 years ago

PR #41 closes this issue.