[Hours] [Activity] 2013/08/05 Monday 1.18 #117087 - review, test and comment 2.75 #3330 - produce an alternate patch ===== 3.93 2013/08/06 Tuesday 0.22 #117087, #3330 - follow-up 1.05 scan for new and old tickets with patches to 107962 0.50 #114972 - check related tickets 0.28 #119151 - comment 0.23 #116911 - comment on closing in a few days 0.10 #113088 - reject with comment 0.68 #118139 - storable unused variables - comment 0.83 #115928 - start alternate implementation ===== 3.89 2013/08/07 Wednesday 2.97 #115928 - finish base implementation ===== 2.97 2013/08/08 Thursday 0.77 #115928 - public implementation 2.52 #115928 - testing, fixes, comment with patches ===== 3.29 2013/08/09 Friday 1.77 #115928 - follow-up, test on Win32 0.97 #117793 - retest, apply, close ticket 0.13 #118933 - resolve 1.18 #92446 - apply with fixes 0.05 #116911 - close 0.12 #112350 - close 0.67 look for tickets to work on, comment on #2968 ===== 4.89 2013/08/10 Saturday 1.07 look for old tickets with patches (and no patches) to 78066 ===== 1.07 Which I calculate is 20.04 hours Approximately 13 tickets were reviewed or otherwise worked on, and 2 patches were applied This week I prepare alternate patches for two tickets, #3330 and #115928. [perl #3330] This adds a warning when the user attempts to use perl's increment operator on a string that isn't a number and doesn't match /^[A-Za-z]+[0-9]+/. [perl #115928] The original issue is that rand() on Win32 only has a 15-bit result, the minimum expected by C89. This ticket produced various wild ideas (from a "let's make random numbers" standpoint.) I wrote a patch to take Yves's copy of the freebsd implementation and use that as perl's random number generator.