2
I Use This!
Low Activity

Commits : Listings

Analyzed 1 day ago. based on code collected 1 day ago.
Nov 18, 2024 — Nov 18, 2025
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
add FinalizingOption type, convert relevant checks over to optparse based query_cache enabling; in the process, rework the pipelining to use itertools.groupby (faster then what was used now).
ferringb
as Brian Harring
More... about 19 years ago
requirements of checks specified via optparse options; centralizes few hardcoded settings (arches fex), and opens them up to user control
ferringb
as Brian Harring
More... about 19 years ago
use optparse for arg processing.
ferringb
as Brian Harring
More... about 19 years ago
change over requires_profiles to requirement attrib
ferringb
as Brian Harring
More... about 19 years ago
fixup imports.
ferringb
as Brian Harring
More... about 19 years ago
start of shifting over to check 'requirements'
ferringb
as Brian Harring
More... about 19 years ago
typo fix
ferringb
as Brian Harring
More... about 19 years ago
reorganization, checking logic first, then results classes...
ferringb
as Brian Harring
More... about 19 years ago
get_cpvstr, wrapper to pull cpvstr if accessible instead of just stringify'ing a package instance
ferringb
as Brian Harring
More... about 19 years ago
saner help msg, saner -c support
ferringb
as Brian Harring
More... about 19 years ago
add multilib-strict to know restricts
ferringb
as Brian Harring
More... about 19 years ago
fix --list-checks output
ferringb
as Brian Harring
More... about 19 years ago
fixup code to check post_rdepends also since it's now split out.
ferringb
as Brian Harring
More... about 19 years ago
hash strings instead of holding onto the atoms hashed long val for comparison reasons.
ferringb
as Brian Harring
More... about 19 years ago
break up __main__ into seperate functions somewhat
ferringb
as Brian Harring
More... over 19 years ago
remove redundant hash calls via shifting the cache lookup into the loop, shaves another 6s, putting it at 144s.
ferringb
as Brian Harring
More... over 19 years ago
rewrite the modx check a bit to limit the scope of dnf_solutions calls, 20s faster, and picks up cases it was missing before.
ferringb
as Brian Harring
More... over 19 years ago
scan for _all_ non visible breaks, not just the first.
ferringb
as Brian Harring
More... over 19 years ago
dnf_solutions -> iter_dnf_solutions
ferringb
as Brian Harring
More... over 19 years ago
shift the blocks check to after the key check
ferringb
as Brian Harring
More... over 19 years ago
remove daft unstable_unique call. shaves around 30s off (another ~10%), knocking it down to 176s for a full run.
ferringb
as Brian Harring
More... over 19 years ago
SystemExit -> (SystemExit, KeyboardInterrupt)
ferringb
as Brian Harring
More... over 19 years ago
fix bug introduced 3 revs back involving query cache
ferringb
as Brian Harring
More... over 19 years ago
fixup docstrings on result classes
ferringb
as Brian Harring
More... over 19 years ago
fix up xml generation, and punt unused result class
ferringb
as Brian Harring
More... over 19 years ago
final major evaluate_depset/relevant profile identification redundant work elimination work; shaves another 14 seconds off, 3m19s. Kind of curious what the profiling would show now. :)
ferringb
as Brian Harring
More... over 19 years ago
further caching of costly generation. knocks off another 23s down to 3m33s user.
ferringb
as Brian Harring
More... over 19 years ago
refactor the evaluate_depset caching into the feeder
ferringb
as Brian Harring
More... over 19 years ago
generalize cache passing, pass in the feeder instead
ferringb
as Brian Harring
More... over 19 years ago
reuse of evaluate_depset where possible- chops # of evaluate_depset calls from 1.4 million to 150k (caching rate of roughly 80%) So... this puts pkgcore_checks under the 4 minute mile mark. :)
ferringb
as Brian Harring
More... over 19 years ago