VS-1368 The tarball is too damn big #8829
Conversation
Co-authored-by: Miguel Covarrubias <[email protected]>
|
It reduces the size of just the header lines in the interval list from 581689 bytes to 3976. So 0.0068 smaller. |
There was a problem hiding this comment.
I still feel that a whitelist vs blacklist approach would be cleaner, as it would allow us to avoid the grep regex filtering. It would let us keep what we want by doing it in a way that is genomically meaningful--specifying the intervals that we want to keep and tossing everything that isn't in them by intersecting our whitelist with the intervals we are passed--instead of relying on string manipulation. But I'm willing to give this a thumb as long as we ensure that we go back and do it in a cleaner way later. We want to get something workable in place sooner rather than later, after all.
Also, I'd probably be compelled to give it a thumb anyway just for the name of the PR.
* Compress the tarball saves a bit. * Remove unused contigs from interval_list files by grepping. --------- Co-authored-by: Miguel Covarrubias <[email protected]>
* Compress the tarball saves a bit. * Remove unused contigs from interval_list files by grepping. --------- Co-authored-by: Miguel Covarrubias <[email protected]>
* Compress the tarball saves a bit. * Remove unused contigs from interval_list files by grepping. --------- Co-authored-by: Miguel Covarrubias <[email protected]>
This PR does 2 things to address the size of the tarball:
Example run here.