| | <<<Back 1 day (to 2012/09/26) | 2012/09/27 |
| marcosw | I need to take casper down for a few minutes. | 03:57.02 |
| sebras | marcosw: I managed to catch your last message before the reboot. :) | 07:54.51 |
| chrisl | kens: ping | 08:17.02 |
| kens | chrisl pong | 08:17.10 |
| chrisl | SO, flights: are you also heading out on the 6th and back on the 9th? | 08:17.36 |
| kens | Yeah I think so, may as well do one day at the show | 08:17.49 |
| | I'm considering flying premium though | 08:17.58 |
| chrisl | Both ways, or just return? | 08:18.23 |
| kens | Both ways ;-) | 08:18.30 |
| chrisl | Using miles or cash? | 08:19.01 |
| kens | Not sure, might use miles if I have enough (7,500 each way) | 08:19.22 |
| chrisl | I'm not seeing an option to use miles - I'll try again.... | 08:19.54 |
| kens | You need to check the 'miles plus money' radio button | 08:20.15 |
| | at the 'select detination' screen | 08:20.27 |
| chrisl | Yeh, they seem to have dumbed down the default search screen.... | 08:20.45 |
| kens | Oh yeah, I was fdoing it while logged in | 08:21.17 |
| chrisl | I am logged in, and I still had to use the "more options" link.... | 08:21.45 |
| kens | Oh odd, I didn't.... | 08:22.04 |
| chrisl | Okay, so 918ukp + miles both ways. Not bad..... | 08:22.50 |
| kens | Yes, it looks like its £300 more than the economy fare. | 08:23.21 |
| | More or less | 08:23.32 |
| chrisl | For me the economy fair came in at 705ukp | 08:23.51 |
| kens | I think its 600 odd if you have miles+money for that too. | 08:24.17 |
| | I meant 300 less fo rhte equivalent economy fare | 08:24.34 |
| | So I'd need 7 tier points to keep silver, by 28th Feb. | 08:24.56 |
| | One flight to go.... | 08:25.04 |
| chrisl | Oh, okay. How does it work out expensing Miles for part of the final ticket tally? | 08:25.32 |
| kens | Take a scrren shot or printout of the economy fare | 08:25.56 |
| chrisl | Great! I might as well use some of these air miles! | 08:26.21 |
| kens | Send that to Miles along with the actual ticket, and only claim the economy potion | 08:26.23 |
| | I did use a load of miles flying us to New York, but then gained 10,000 flying to SFO, so I think I have enough | 08:27.23 |
| | Even if I paid for premium btoh ways it would only be 6 tier points, so there's no way I'm getting 7 by February. | 08:27.53 |
| | So might as well use air miles. | 08:28.04 |
| chrisl | I'm not bothered about staying silver - it doesn't seem to benefit us much normally | 08:28.38 |
| kens | Gets you more miles, but not much more, no | 08:28.48 |
| chrisl | I'd just got to know my credit card details by memory, and they sent me a new bloody card! | 08:30.31 |
| kens | :-) | 08:30.46 |
| chrisl | Worse still, the new card is only valid until next year..... | 08:31.12 |
| kens | That's odd, why did they do that ? | 08:31.24 |
| | Dammit where do I find out how many miles I have ? .... | 08:32.20 |
| chrisl | Based on previous experience, it probably means someone had their card cloned, so they send new cards "just in case", but only valid until the previous card would expire | 08:32.44 |
| kens | Oh D'oh, my miles is at the top... I have 21,000 | 08:33.17 |
| | More than enough for this flight | 08:33.23 |
| | Hmm, must remember to do a new ESTA as well, mine expires on Dec 8th | 08:35.08 |
| chrisl | Okay, booked and saved the economy quote to send for expenses.... | 08:36.43 |
| kens | OK so I'm also using Miles+Money and going premium economy both ways. Flying on the 6th and back on the 9th. | 08:36.54 |
| chrisl | Yep, VS019 and VS020 | 08:37.11 |
| kens | Indeed, just doing my booking now | 08:37.21 |
| chrisl | And I've selected 36C both out and return | 08:40.39 |
| kens | Not quite there yet, the site seems a little slow today | 08:40.55 |
| chrisl | Oh, I just assumed it was because I'm using my parents' broadband..... | 08:41.30 |
| kens | Huh, that's a new check box, agree to pay government post-purchase price increases... | 08:41.32 |
| chrisl | I thought I remembered that from last time. I remember reading about it somewhere | 08:42.02 |
| kens | Well, I have a terribel memory.... | 08:42.33 |
| chrisl | IIRC, it's the latest nonsense from Government that increases in duty can be levied on all future flights, not just future bookings | 08:43.55 |
| kens | money gouging again. | 08:44.10 |
| | Grrr, somethign went wrong with my credit card authorisation, lokos like I have to restart from square one. | 08:44.28 |
| chrisl | I was half expecting mine to be put "on hold" with it being a brand new card. | 08:45.12 |
| kens | I probably got the expiry wrong. Virgin keep on putting in a named month, and I have trouble with that because my card has a numerical expiry date. | 08:45.47 |
| | I'm not good at converting one into the other | 08:46.08 |
| chrisl | For a while my debit card had the named month, and my credit card had it numeric, that caused me confusion..... | 08:47.33 |
| kens | Hmm, transaction failed again.... | 08:48.35 |
| | time to follow the on-screen directions and phone them | 08:48.58 |
| chrisl | My laptop is nagging me to update, so I'm going to start that rolling and go make a coffee | 08:50.29 |
| kens | plenty of time for that ;-) | 08:50.45 |
| chrisl | and now reboot..... | 09:01.16 |
| kens | grrr nationwide put a block on mt card, now I have to sort out why. | 09:11.23 |
| | Right back to step 1 with virgin.... | 09:15.47 |
| chrisl | Right back to breaking the tiff devices :-) | 09:27.24 |
| Robin_Watts | Turned out the 603 price I was looking at yesterday (miles+money) was Economy both ways, not PE on return, which was a shame. | 09:30.55 |
| | So I just went for Economy both ways. | 09:31.11 |
| kens | Well I have *finally* finished my booking.... | 09:31.26 |
| | Bloody stupid verification procedure 'Can you tell us the balance of one of your other accoutns ?@ well, no, I don't have that information memorised.... | 09:32.04 |
| chrisl | Robin_Watts: so, for once, I won't be the furthest back on the plane! I wonder how I'll contrive to still be the last off..... | 09:32.27 |
| kens | chrisl you reckon you're in 32c back ? | 09:37.57 |
| chrisl | Er, yeh | 09:38.06 |
| kens | Hmm, seat map shows that as available | 09:38.17 |
| | Until I just tok it | 09:38.31 |
| | I now have 33c out, 32c back | 09:39.01 |
| mace | Robin_Watts: about? | 09:39.06 |
| Robin_Watts | mace: hi | 09:39.12 |
| mace | any idea if Ray got busy again? | 09:39.28 |
| kens | ray is always busy ;-) | 09:39.38 |
| mace | never got a reply from the other week :) | 09:39.38 |
| | hehe | 09:39.44 |
| chrisl | kens: ah, no, sorry 36c there and back | 09:40.44 |
| Robin_Watts | He's away from home at the moment due to it being remodelled, but I don't think he's off at customers too much. | 09:41.00 |
| kens | OK well I'm currently in 33c out and 32c back. So I can't kick your seat ;-( | 09:41.14 |
| Robin_Watts | mace: Was it an email you were expecting a reply to? | 09:41.21 |
| | Send it again, and copy it to tech@artifex.com - that way we'll all see it, and can hassle him if he doesn't reply. | 09:42.09 |
| mace | Robin_Watts: well there was an email months back which was left with him to check server specs etc, but there was that very brief discussion on irc last week | 09:42.13 |
| Robin_Watts | mace: Right. He's probably just forgotten :( | 09:42.32 |
| mace | np, will drop him a line, didn't want to hassle him :) | 09:42.50 |
| kens | tor8 Robin_Watts there is someone on Stack Exchange asking about adding annotations in MuPDF, don't know if you want to respond, the URL is: | 09:44.25 |
| | http://stackoverflow.com/questions/12608761/ink-annotations-in-mupdf | 09:44.25 |
| Robin_Watts | ew. Nasty. | 09:46.44 |
| kens | You can always ignore it | 09:47.15 |
| Robin_Watts | paulgardiner: I pushed tor8's git submodule changes yesterday. | 10:08.06 |
| | And I updated the cluster to expect thirdparty to be in that form. | 10:08.45 |
| paulgardiner | ah right. Did you figure out how to make the submodule updates automatic | 10:08.46 |
| Robin_Watts | No. | 10:08.58 |
| | "git submodule update --init" is the magic command. | 10:09.12 |
| | So, you 'll need to pull that in before you can use the cluster. | 10:09.52 |
| | v8 is NOT a git submodule though. | 10:10.10 |
| tor8 | all you need is that command (git submodule update --init) in the .git/hooks/post-checkout and post-merge hooks | 10:10.16 |
| Robin_Watts | so keep that in there as a directory. | 10:10.27 |
| paulgardiner | So. Move v8 out somewhere safe; delete thirdparty; do the git sm stuff; put v8 back? | 10:11.35 |
| tor8 | git diff and git status will also show if you're not up to date, so you probably don't even need the hooks | 10:11.39 |
| Robin_Watts | paulgardiner: Or just delete everything in thirdparty except v8, then do git sm stuff. | 10:12.25 |
| paulgardiner | ah yes right | 10:12.35 |
| | tor8: I guess we still wont be updating thirdparty very often, so it might be more efficient to leave out the hooks, and see if it often runs me into trouble | 10:13.31 |
| | I imagine this works only when far along the master branch, right? | 10:14.52 |
| Robin_Watts | backtracking to before the point it was introduced won't work, indeed. | 10:15.22 |
| tor8 | Robin_Watts: actually, it'll just complain about not being able to remove the directories | 10:16.49 |
| | so it mostly works | 10:16.57 |
| Robin_Watts | right, but you'll be left with no code in them. | 10:17.04 |
| tor8 | the submodules will still be there, just unmanaged | 10:17.16 |
| | sort of like the situation with the third party zip | 10:17.28 |
| Robin_Watts | really? | 10:17.31 |
| tor8 | argh, autocorrect | 10:17.33 |
| Robin_Watts | OK, well, the include paths will be wrong on windows anyway. | 10:17.52 |
| tor8 | yeah, it just warns and leaves the directories with the submodules checked out | 10:17.56 |
| | the unix makefile uses wildcards so that will work, the windows projects won't because of the version numbers | 10:18.16 |
| paulgardiner | So is --init always needed, or is that just for the first time? | 10:19.02 |
| Robin_Watts | paulgardiner: --init is needed for the first time. | 10:19.33 |
| | but the nice thing is that if you put it there, it's only used if it's needed. | 10:19.50 |
| | so if you were setting an alias, you'd set it with --init present. | 10:20.02 |
| paulgardiner | right | 10:20.53 |
| | He he! Done that, and it all just worked. | 10:21.25 |
| Robin_Watts | I need someone to invent a really good reason why I can't go running today... | 10:22.02 |
| paulgardiner | You shouldn't go running because it doesn't really matter if you miss a run out of laziness and for no other good reason... sorry that wasn't a very good effort was it?! :-) | 10:23.40 |
| Robin_Watts | tor8: So, with mesh shadings, we load them, and do all the subdivision at load time, not at render time. | 12:43.55 |
| tor8 | Robin_Watts: oh, you found a test case that actually uses tensor patches? | 12:44.21 |
| | and not one of the stupid illustrator ones that subdivides the patches into smaller patches before saving... | 12:45.01 |
| Robin_Watts | I have a case htere that uses Type 7 shadings, yes. | 12:45.12 |
| sebras | Robin_Watts: hello? :) | 12:47.18 |
| | is there a problem with the shadings? | 12:47.34 |
| Robin_Watts | sebras: memory explosion. 750Meg for this page. | 12:47.48 |
| sebras | ouch. | 12:47.51 |
| | ok, so you want to postpone the subdivision until rendering it? | 12:48.08 |
| Robin_Watts | sebras: I think that would be preferable. | 12:48.17 |
| | but I'm wondering whether we shouldn't leave the shading as a compressed stream and do all the parsing/subdivision at render time. | 12:48.58 |
| sebras | that's an option of course. | 12:49.12 |
| | like we do for images nowadays. | 12:49.30 |
| kens | Robin_Watts : I think I've killed the cluster | 12:49.32 |
| | Its killing nodes individually | 12:50.17 |
| Robin_Watts | kens: It'll fix itself in 5 mins. | 12:50.18 |
| kens | OK | 12:50.26 |
| | 5 nodes to kill :-) | 13:14.32 |
| chrisl | Right, heading home - back in an hour or so...... | 13:17.41 |
| Robin_Watts | kens: I'm assuming it's retrying? | 13:48.25 |
| kens | Looks like it | 13:49.07 |
| | I htink I should kill it | 13:49.13 |
| | But I would like to knwo what is killing it | 13:49.20 |
| | I just sent an abort | 13:50.14 |
| | I think its one file, it causes a server to timeout, which causes the others to rety those jobs, one of which | 13:50.41 |
| | times out | 13:50.45 |
| | So the remainder retry, and one times out | 13:50.58 |
| | Etc | 13:51.00 |
| Robin_Watts | It can't be a single file. | 13:51.17 |
| kens | Why not ? | 13:51.22 |
| Robin_Watts | If a single file times out then we ignore it. | 13:51.26 |
| kens | Oh... | 13:51.30 |
| Robin_Watts | It's only if we start to have too many timeouts in general that it gets confused. | 13:51.48 |
| kens | What if it causes server to stop responding ? | 13:51.50 |
| Robin_Watts | BUT... marcos updated the cluster code yesterday. | 13:52.01 |
| kens | Well, I woudl still like to know what 's wrong :-( | 13:52.16 |
| Robin_Watts | I don't know what to suggest, sorry. | 13:52.42 |
| | I wonder if we could ask marcos to upload the cluster logs even in the case of an abort... | 13:53.03 |
| kens | I'll wait for marcos and see if he can tell me | 13:56.00 |
| | marcosw you there ? | 14:27.04 |
| | Well, I guess that's a no then | 14:30.34 |
| Robin_Watts | ray_laptop: ping? | 14:37.52 |
| ray_laptop | Robin_Watts: pong | 14:38.03 |
| Robin_Watts | mace: ping? | 14:38.08 |
| | ray_laptop: mace was on here earlier looking for you. | 14:38.16 |
| ray_laptop | no mace here. I sent Miah a reply to an email earlier. BTW 'he' or 'she' ? | 14:38.50 |
| Robin_Watts | He. | 14:39.07 |
| | mace is here, at least according to my chatzilla. | 14:39.17 |
| kens | mine has him marked as 'away' | 14:39.30 |
| Robin_Watts | ah, well if you've answered the email, that's probably all that matters :) | 14:39.55 |
| ray_laptop | Robin_Watts: do you know how I can check if our web server has all of the components available he expected (why this didn't get discussed prior to genning up a website, I don't know) | 14:41.13 |
| Robin_Watts | ray_laptop: I didn't see his email, did I ? | 14:41.32 |
| | What did he need? | 14:41.46 |
| | mysql and php ? | 14:41.51 |
| | who are we hosted with? godaddy? | 14:41.59 |
| ray_laptop | Robin_Watts: maybe not -- I wasn't sure how you knew him, and where he came from | 14:42.01 |
| Robin_Watts | I know mace from Picsel. | 14:42.11 |
| ray_laptop | Robin_Watts: no, XO communications | 14:42.12 |
| | mace/miah wrote me and said: "it requires an Apache/MySQL/PHP environment" | 14:43.16 |
| Robin_Watts | ok, that doesn't sound unusual. | 14:43.29 |
| ray_laptop | no, but the server is NOT a linux server and we DON'T have the ability to install packages. | 14:44.01 |
| Robin_Watts | It's a windows based host? | 14:44.14 |
| ray_laptop | in particular I wasn't sure why we needed mysql | 14:44.15 |
| | nope, | 14:44.23 |
| Robin_Watts | I assume he has a CMS of some kind for miles/scott to post news articles. | 14:44.38 |
| | (Content Management System, not Color Management System :) ) | 14:44.50 |
| ray_laptop | Robin_Watts: SunOShttp://www.artifex.com 5.8 VDE sun4u | 14:45.25 |
| Robin_Watts | If XO communications can't do it, then we just move to a system that can. Like ix or godaddy or .... | 14:45.27 |
| ray_laptop | Robin_Watts: maybe, but I just asked him for a tarball of the website so we could just see if it runs | 14:46.17 |
| Robin_Watts | ray_laptop: OK. so XO lists MySQL as an update option. | 14:46.33 |
| | MySQL without PHP is bonkers. | 14:47.19 |
| ray_laptop | if I type 'mysql' it says: ERROR 2001: Can't create UNIX socket (99) | 14:47.27 |
| Robin_Watts | You have shell access? | 14:48.01 |
| | artifex.com appears down to me. | 14:48.11 |
| ray_laptop | Robin_Watts: yes. | 14:48.27 |
| Robin_Watts | kwality hosting. | 14:48.36 |
| ray_laptop | http://www.artifex.com | 14:48.51 |
| | it works for me | 14:49.04 |
| Robin_Watts | not for me :( | 14:49.29 |
| mace | ray_laptop: hi | 14:49.33 |
| chrisl | Robin_Watts: it works for me, too | 14:49.49 |
| ray_laptop | mace: hi. I assume you have caught up on the recent logs ??? | 14:49.54 |
| mace | just did yeah, good to finally catch you :) | 14:51.27 |
| | reason for mysql is as robin noted, one of miles key requirements was for a site he can update without hacking on html | 14:51.59 |
| | we probably could get it working with sqlite but mileage may vary etc | 14:52.25 |
| Robin_Watts | In this day and age, ANY decent webhosting company should provide MySQL/PHP. | 14:55.39 |
| | mace: Are you tied particularly to Apache? | 14:55.48 |
| | (It seems like the default choice for most hosters, but...) | 14:56.01 |
| mace | nope, it just happens to be the server we have the most experience with. anything that'll run php and is /reasonably/ mainstream should work | 14:56.30 |
| Robin_Watts | Right. I have had apache/MySQL/PHP based stuff working off a windows box before. | 14:57.20 |
| mace | ray_laptop: simplest test to do would be to create a test.php file somewhere in a web accessible directory and put <?php phpinfo(); ?> in it | 14:57.28 |
| | if it executes correctly it'll output all manner of detail about php etc | 14:58.21 |
| Robin_Watts | http://www.wss.co.uk/test.php <- That's ix. | 15:00.34 |
| mace | forgive my ignorance.. i take it thats not where artifex.com is hosted? | 15:04.10 |
| Robin_Watts | mace: No. Currently artifex.com is with XO communications. | 15:04.27 |
| mace | nods | 15:04.37 |
| Robin_Watts | We have some websites on godaddy, and we did for a while have a downloads site on ix, until our traffic blew it away. | 15:05.02 |
| | I don't have access to the artifex.com site to try adding that file though, ray will need to do that. | 15:05.38 |
| | (at the moment, I can't even reach artifex.com in my browser :( ) | 15:05.54 |
| kens | me neither | 15:06.25 |
| | can't find... | 15:06.31 |
| Robin_Watts | kens: good to know it's not just me. | 15:08.22 |
| kens | Yeqh, thought you would feel that way :-) | 15:08.32 |
| mace | i'm not entirely clear on how artifex.com is hosted, but certainly there are 4 webserver ip addresses being advertised for artifex.com and 4 forhttp://www.artifex.com, however they are not the same ips | 15:08.40 |
| | once of the advertised ips is 207.155.252.18 and it is down | 15:09.38 |
| | actually three of them are down/unresponsive, 207.155.252.18, 207.155.248.31 and 207.155.248.47 | 15:10.25 |
| | so i guess you're just unlucky.. 75% chance of hitting a duff server :( | 15:10.48 |
| kens | Oh,oh, there it goes again | 15:32.35 |
| | I wonder if this is actually a network problem I'm seeing | 15:32.50 |
| | Weird, and now its says everything is just fine.... | 15:33.36 |
| mvrhel | good morning | 15:56.40 |
| kens | morning Michael | 15:57.06 |
| henrys | hi mvrhel | 15:57.13 |
| | well I have xps zip container writing working now. The simple driver should be easy now. It will be much like gdevsvg.c I suppose. | 15:58.08 |
| | it's amazing all the crap they put in an XPS file - schema etc. | 15:59.11 |
| mvrhel | thats good news | 15:59.21 |
| | I just need to do a bit of documenation on this black point compensation stuff. It all seems to be working so I can close my last customer ehancement. although I have 2 things for customer 330 | 16:00.36 |
| | that should not take too long | 16:00.52 |
| | kens: then I can work on V2 profile stuff | 16:01.03 |
| kens | mvrhel that would be great, I#'m hoping to sinish up the linearisation stuff soon | 16:01.22 |
| | finish... | 16:01.30 |
| | Then the colour code is my next priority | 16:01.43 |
| henrys | chrisl:please don't drop any of our customer names to Karen and friends. | 16:14.10 |
| kens | Ah, *much* better, no pdfwrite differences with working files and down to ~30 seg faults, a bit improvment. | 16:15.55 |
| | Oddly almost all the seg faults are with PCL input. | 16:16.18 |
| henrys | yeah we got paid I guess I'll stick around | 16:18.41 |
| kens | wonders if we Europeans did.... | 16:18.57 |
| Robin_Watts | kens: I got paid 30th August. | 16:20.15 |
| kens | OK robin thanks | 16:25.47 |
| | Goodnight all. | 16:26.40 |
| mvrhel | looks like cia is broken again | 17:19.12 |
| Robin_Watts | chrisl fixed it last time. | 17:19.29 |
| henrys | chrisl:am I missing something I tried autoconf'ing (puke) using --with-drivers=ALL and did not get svgwrite. Maybe a bug. | 17:20.16 |
| Robin_Watts | svgwrite doesn't work currently, right? | 17:21.24 |
| henrys | it has bugs but it works, it looks like configure.ac intends to include it with the ALL option. | 17:23.01 |
| Robin_Watts | I only see svgwrite in the gs stuff, not the pcl stuff. | 17:26.10 |
| | s/stuff/configure.ac/ | 17:26.30 |
| henrys | I'm working with gs - it is in the windows build? | 17:26.32 |
| Robin_Watts | no. | 17:26.59 |
| | does it work if you --with-drivers=SVG ? | 17:28.58 |
| | Ah. | 17:29.40 |
| | ALL = PRINTERS,FILES,X11 | 17:29.48 |
| | Probably FILES should gain SVG_DEVS. And PNG_DEVS for that matter. | 17:30.43 |
| henrys | I got it to work with --with-drivers=ALL,svgwrite - but I'm not happy with our definition of ALL. | 17:31.45 |
| Robin_Watts | and JBIG2DEVS | 17:31.46 |
| | possibly ALL should be made PRINTERS,FILES,X11,OBSCURE :) | 17:32.15 |
| henrys | seems to me svgwrite is a subset of ALL | 17:32.16 |
| Robin_Watts | seems reasonable to me too. | 17:32.23 |
| mvrhel | henrys: so Stefan does not seem to reply to my emails that I have sent him over the past few weeks as I add in these enhancements | 18:03.24 |
| | not sure what is going on there | 18:03.33 |
| Robin_Watts | mvrhel: Send him a mail asking if your mails are getting through? | 18:06.19 |
| mvrhel | probably a good idea | 18:07.50 |
| | ok. I guess I should add in the capability for black preservation in cmyk to cmyk conversions. yet another option.... | 18:11.18 |
| | bbiab | 18:11.36 |
| henrys | mvrhel:actually I'll talk to Miles and he'll probably just call him and touch base, unless you object. | 18:27.58 |
| chrisl | henrys: sorry, I forgot to mark myself "away" - we removed svgwrite from all the "standard" lists because of its propensity to crash, so it has to be requested explicitly in the configure options. | 18:40.47 |
| henrys | chrisl:okay no problem. | 18:53.55 |
| chrisl | henrys: I do have a "to do" item, to review that part of configure - the way it "works" is too convoluted just now, handling the dependencies | 18:55.15 |
| henrys | I guess we are going to have to figure out the svg crashes the xps device is going to look very similar. | 19:00.35 |
| chrisl | Well, I can add svg device back in - I need to do a bit of configure hacking to put tiffsep1 back in (I accidentally removed it) | 19:02.34 |
| marcosw1 | Robin_Watts: http://git.ghostscript.com/?p=ghostpdl.git;a=summary shows the relative date that each commit was pushed (i.e. 109 min ago). other than scraping the web page how can I get that information? | 19:11.15 |
| Robin_Watts | git log ? | 19:11.51 |
| marcosw1 | not as far as I can figure out. it just gives me the commit date/time and not the push date/time | 19:12.58 |
| chrisl | marcosw1: I'm pretty sure the time on the web page is the commit time, not the push time | 19:15.27 |
| marcosw1 | the google machine suggests that I have to run a git reflog command on git.ghostscript.co. | 19:15.31 |
| Robin_Watts | git log --pretty=format:"%cr" | 19:15.36 |
| | git log --pretty=format:"%cr %ar" <- That shows both dates in relative format | 19:16.14 |
| | cr is the 'commit' date. | 19:17.06 |
| | ar is the author date. | 19:17.13 |
| | Neither of those is the push date though. | 19:17.26 |
| | I'm not sure there is a concept of a push date. | 19:17.34 |
| marcosw1 | Robin_Watts: clearly there is, git.ghostcript.com uses it to sort the commits. | 19:17.59 |
| Robin_Watts | Not at all. | 19:18.06 |
| | We always pull --rebase before we push, right? | 19:18.27 |
| saper | there might something in the repository reflog, probably not accessible remotely | 19:18.30 |
| Robin_Watts | which means any commits we push up will have a newer commit date than the ones that were there before. | 19:18.59 |
| | because the commit dates are reset as part of the reapplication process in the rebase. | 19:19.18 |
| | I think. | 19:21.05 |
| chrisl | That's my understanding, yes | 19:21.44 |
| marcosw1 | Robin_Watts and chrisl: okay, thanks. | 19:22.10 |
| Robin_Watts | http://stackoverflow.com/questions/3357219/expose-the-date-a-commit-was-pushed-to-a-repository | 19:22.24 |
| | That suggests that the time you were hoping for isn't provided as standard even internally within bare repos. | 19:22.37 |
| | marcosw: Do you care, other than for the fact that you want to get the commits sorted in the right order? | 19:23.11 |
| marcosw1 | Robin_Watts: yes, but based on your pull --rebase comment I believe the order of the commits sorted by commit date should match the order sorted by push date, which is what I'm looking for. | 19:23.28 |
| Robin_Watts | And actually, why are you needing to sort the commits? | 19:23.32 |
| marcosw1 | I want to sort the commits so I can figure out the order they were regression tested in. | 19:24.21 |
| Robin_Watts | For any SHA, you know it was tested after SHA~1 | 19:24.43 |
| marcosw1 | yes. that gives me the order, but I figured getting the date/time would be possible as well. | 19:27.35 |
| Robin_Watts | so: git "rev-list SHA~1 -1" will always give you the SHA that this one was tested against. | 19:27.44 |
| | fair enough. | 19:27.46 |
| marcosw1 | I can use the timestamp of the log files, but that fails if the files are touched or copied without -p. | 19:28.14 |
| chrisl | Well, regression order == commit order == temporal order because we always rebase before a push | 19:28.59 |
| Robin_Watts | assuming nothing goes wrong in the cluster :) | 19:29.26 |
| marcosw1 | chrisl: yeah, I figured that out a couple of comments ago (12:23:28 PM, my time). | 19:29.48 |
| chrisl | And assuming we never have to back out a commit... but in general | 19:30.01 |
| Robin_Watts | If we rerun git jobs because of a cluster node failing spuriously for example... | 19:30.20 |
| | dinner time. back soon. | 19:30.25 |
| marcosw1 | I've decided I'm happy with 'git log --format="%H %cd" --date=short' | 19:32.07 |
| chrisl | I use the commit order for the changelog | 19:32.48 |
| mvrhel | bbiaw | 20:20.10 |
| henrys | mvrhel:still there | 20:20.45 |
| | ? | 20:20.46 |
| mvrhel | henrys: yes | 20:26.10 |
| | just getting ready to head out for a bit | 20:26.16 |
| henrys | I said above:actually I'll talk to Miles and he'll probably just call him | 20:26.33 |
| | and touch base, unless you object. | 20:26.33 |
| mvrhel | oh yes. I think that is a good idea henrys | 20:27.07 |
| henrys | okay | 20:27.15 |
| mvrhel | I had probably been a little slow at getting some of his enhancements because of all the planar sep work. but that is all stuff he is going to use also | 20:28.01 |
| | ok bbiaw | 20:28.19 |
| radistao | Hi. | 20:31.30 |
| | i convert JPEG to PDF using [viewjpeg.ps JPEG_FILE_NAME viewJPEG showpage] | 20:32.39 |
| | this creates PDF with default page size | 20:33.14 |
| | how to fit PDF page to input image dimensions? | 20:33.38 |
| chrisl | radistao: viewjpeg.ps doesn't do that, you would need to modify the Postscript to make it work that way | 20:38.51 |
| radistao | do you have any idea? Is it possible at all? can i change viewjpeg.ps to make varable PDF page size (i mean 1 JPEG -> 1 PDF with single page) | 20:56.52 |
| chrisl | radistao: I can't see any reason that it couldn't be made to work as you want, but I'm a bit too tired to look into it just now. | 21:02.24 |
| radistao | ok. thx any way. will continue later | 21:04.40 |
| | i'll try myself first... | 21:04.56 |
| chrisl | Oops, I just realised why my initial thought didn't work, and he's gone.... oh, well... | 21:18.15 |
| | Forward 1 day (to 2012/09/28)>>> | |
IRC Logs