[13:07] Odd_Bloke, do you think my returning "......" is silly ? [13:07] for dmi data. [13:11] smoser: In what case? [13:12] recent commit to trunk, i made it return ..... (what dmi decode would return) when /sys is full of \xff [13:12] but i relally think "" seems better, as otherwise the user has to specifically know about this "..." [13:13] (and worse, i return the length of the value in sys, basically .replace('\xff', ".") [13:14] smoser: I've been seeing the recent activity around here on dev of new cloud-init function around networking, which is where the majority of my interest lies. What dev is going on? Is this on a 0.7.x branch? It seems the v2.0 work died off. [13:14] 0.7.x [13:14] my 'net1' branch is where i'm working on it, and i want to get at least part of that into ubuntu today. [13:15] and then you'll merge that back to master? [13:15] oh. well, yes. it will all land in trunk [13:15] before ubuntu [13:15] thanks [13:18] smoser: Yeah, I definitely think returning '' or None would be better. [13:18] '' seems better than None to me. [13:18] in order to differenciate between "NO VALUE THERE" and "value not set" [13:19] smoser: Oh, sorry, this is for all Fs, not all 0s? [13:19] Then, yeah, agreed on empty-string. [13:20] Of course, the question then is if we modify dmidecode's output in that case? :p [13:20] Odd_Bloke, i'm modifying dmidecode's output [13:23] Odd_Bloke, review? [13:23] https://code.launchpad.net/~smoser/cloud-init/trunk.dmidecode-null/+merge/288925 [13:28] smoser: I'm heading in to a (hopefully) half-hour meeting; will review after. :) [13:28] meh. you can review while you wait for everyone else to show up [13:28] really quick [13:29] but your suggestion is also ok [13:29] smoser: Cool; +1'd. === Guest12587 is now known as mfisch === mfisch is now known as Guest62254 === Guest62254 is now known as mfisch