Hi there!
I use this little program on my linux netbook (gentoo, intel 1.6GHz with multithreading and 2 GB RAM) and bash 4.x for completion. After rebuilding the cache, the completion is so damn slow i can`t use it :( If I press tab i have to wait several seconds to gain suggestions. I've tried to profiling the bottlenecks with ruby-prof, a single "mite"-call took me about two seconds and most of it are used for the selects and the maps (mostly regex-stuff)
Total: 1.980000
%self total self wait child calls name
17.17 0.44 0.34 0.00 0.10 465 Array#select
12.63 0.46 0.25 0.00 0.21 353 Array#map
9.60 0.75 0.19 0.00 0.56 134 Kernel#gem_original_require(d5)
9.09 0.18 0.18 0.00 0.00 50790 String#to_s
6.57 0.13 0.13 0.00 0.00 1847 Module#constants
4.04 0.18 0.08 0.00 0.10 27 Kernel#gem_original_require(d7)
4.04 1.14 0.08 0.00 1.06 53 Kernel#gem_original_require(d4)
3.54 0.42 0.07 0.00 0.35 882 Array#each
3.54 0.07 0.07 0.00 0.00 50475 Hash#key?
3.03 0.43 0.06 0.00 0.37 31 Kernel#gem_original_require(d6)
2.53 0.05 0.05 0.00 0.00 159 Array#-
2.02 0.04 0.04 0.00 0.00 136 <Class::Dir>#[]
...
That is way to much!
Hi there!
I use this little program on my linux netbook (gentoo, intel 1.6GHz with multithreading and 2 GB RAM) and bash 4.x for completion. After rebuilding the cache, the completion is so damn slow i can`t use it :( If I press tab i have to wait several seconds to gain suggestions. I've tried to profiling the bottlenecks with ruby-prof, a single "mite"-call took me about two seconds and most of it are used for the selects and the maps (mostly regex-stuff)
That is way to much!