You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
take binary/compile into account when predicting ETA
show binary/compiled counts in stats
Available info
/var/cache/edb/mtimedb has a binpkgssection and distinhuishes ebuild from binary in the mergelist section
emerge.log shows something like this
1733743818: >>> emerge (3 of 29) dev-libs/simdjson-3.11.0 to /
1733743818: === (3 of 29) Merging Binary (dev-libs/simdjson-3.11.0::/var/cache/binpkgs/dev-libs/simdjson-3.11.0.gpkg.tar)
1733743821: >>> AUTOCLEAN: dev-libs/simdjson:0
1733743821: === Unmerging... (dev-libs/simdjson-3.10.1)
1733743824: >>> unmerge success: dev-libs/simdjson-3.10.1
1733743826: === (3 of 29) Post-Build Cleaning (dev-libs/simdjson-3.11.0::/var/cache/binpkgs/dev-libs/simdjson-3.11.0.gpkg.tar)
1733743826: ::: completed emerge (3 of 29) dev-libs/simdjson-3.11.0 to /
Which means that we need to parse new lines and handle state to know if a logged merge was binary or not (so far we only needed to parse the >>> emerge line). Should be easy enough but might slow things down. On the plus side, this would allow us to show the current merge step even to non-root users.
Not sure how, algorithmically, we should separate bin merges from compile merges. I'd like to find something compatible with #56.
The text was updated successfully, but these errors were encountered:
Whishlist
log
predict
stats
Available info
/var/cache/edb/mtimedb
has abinpkgs
section and distinhuishesebuild
frombinary
in the mergelist sectionemerge.log
shows something like thisWhich means that we need to parse new lines and handle state to know if a logged merge was binary or not (so far we only needed to parse the
>>> emerge
line). Should be easy enough but might slow things down. On the plus side, this would allow us to show the current merge step even to non-root users.Not sure how, algorithmically, we should separate bin merges from compile merges. I'd like to find something compatible with #56.
The text was updated successfully, but these errors were encountered: