| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA1 |
| 3 |
|
| 4 |
Steve Long wrote: |
| 5 |
[...] |
| 6 |
>>> for ((i=0;i<100000;i++)); do echo /usr/share/doc/${P}/examples > |
| 7 |
>>> /dev/null; |
| 8 |
>>> real 11.25 |
| 9 |
>>> real 9.24 |
| 10 |
>> So that's what, on the order of 20 microseconds faster for each iteration? |
| 11 |
>> |
| 12 |
> Or ~18%. (You shouldn't use the first iteration in general, btw.) |
| 13 |
|
| 14 |
I've not really got an opinion on the topic, per se, but fwiw, this is |
| 15 |
really not a meaningful statistic. *If* parsing strings in the ebuild is |
| 16 |
not a trivial part of the overall ebuild parsing process, then yes, this |
| 17 |
is a significant gain and should be treated as such. I find it unlikely |
| 18 |
that this would be the case. |
| 19 |
|
| 20 |
I'm not sure how one can go about measuring the impact of this on ebuild |
| 21 |
parsing as a whole. Maybe make take a few "typical" ebuilds, change |
| 22 |
quoting style, and run it through ebuild.sh in a loop. All the inherited |
| 23 |
eclasses would need to change too, though, I guess. |
| 24 |
|
| 25 |
Regards, |
| 26 |
Arun |
| 27 |
-----BEGIN PGP SIGNATURE----- |
| 28 |
Version: GnuPG v2.0.9 (GNU/Linux) |
| 29 |
|
| 30 |
iEYEARECAAYFAkj2sxgACgkQ+Vqt1inD4uzW3wCfancNcJxcyHerjSZdZfK9UKb7 |
| 31 |
k5oAn0186/lLTAS2+n1Z7egzhAP1kISV |
| 32 |
=CkaZ |
| 33 |
-----END PGP SIGNATURE----- |
Archives