| --- Log opened Mon Jan 07 00:00:33 2019 |
| 00:25 | | Degi [Degi@Nightstar-dgotpd.dyn.telefonica.de] has quit [Connection closed] |
| 01:51 | | Kindamoody is now known as Kindamoody[zZz] |
| 04:18 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 04:18 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 05:05 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Connection closed] |
| 05:14 | | Derakon is now known as Derakon[AFK] |
| 05:15 | | Vorntastic [uid293981@Nightstar-6br85t.irccloud.com] has joined #code |
| 05:15 | | mode/#code [+qo Vorntastic Vorntastic] by ChanServ |
| 05:52 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has joined #code |
| 05:58 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has quit [Ping timeout: 121 seconds] |
| 06:42 | | celticminstrel is now known as celmin|sleep |
| 07:35 | | macdjord is now known as macdjord|slep |
| 08:59 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 09:01 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has joined #code |
| 09:01 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 09:13 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has quit [Operation timed out] |
| 09:20 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has joined #code |
| 09:20 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 09:39 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has quit [Operation timed out] |
| 09:47 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has joined #code |
| 09:47 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 09:53 | | Kindamoody[zZz] is now known as Kindamoody |
| 10:25 | | * TheWatcher replies to a work email that combines all his favourite subjects into one ball od horribleness: email, html emails, remote image loading, caching, and analytics |
| 10:25 | <@TheWatcher> | *of |
| 10:26 | <@Tamber> | blech |
| 10:40 | <&Reiver> | TheWatcher: A few months ago our comms team came up with an 'adorable' animated GIF and wanted to have it as a signature in all official comms from the council. |
| 10:41 | <&Reiver> | Shockingly, there were a few wee issues with this... |
| 11:57 | <&[R]> | Like the ancient ones that still get their assistants to print their emails for them? |
| 12:47 | | celmin|sleep is now known as celmin|away |
| 13:15 | | himi [sjjf@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 15:38 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 15:38 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 16:51 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has joined #code |
| 17:05 | | Vorntastic [uid293981@Nightstar-6br85t.irccloud.com] has quit [[NS] Quit: Connection closed for inactivity] |
| 18:36 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Connection closed] |
| 18:49 | | Vorntastic [uid293981@Nightstar-6br85t.irccloud.com] has joined #code |
| 18:49 | | mode/#code [+qo Vorntastic Vorntastic] by ChanServ |
| 19:41 | | Degi_ [Degi@Nightstar-nb8e9q.dyn.telefonica.de] has joined #code |
| 21:03 | < Mahal> | https://techcrunch.com/2019/01/07/github-free-users-now-get-unlimited-private-repositories/ |
| 21:04 | <&McMartin> | Huh |
| 21:04 | < Emmy> | but when do they get unlimited free beer!? |
| 21:05 | <&McMartin> | Silly parasites! It's free as in kittens, not as in beer. |
| 21:12 | < Emmy> | kitties! |
| 21:15 | <&[R]> | Wonder if that's their response to the massive exodius |
| 21:16 | <&[R]> | It's rather delayed if that's the case |
| 21:22 | <&Reiver> | What was the exodus? |
| 21:23 | <&[R]> | MS announced they were going to acquire github |
| 21:23 | <&[R]> | A bunch of people immediately left |
| 21:24 | <&[R]> | When you spend over a decade trying to fuck the tech community, you kind of gain a negative amount of goodwill |
| 21:26 | <&Reiver> | ...oh dear |
| 21:27 | <&McMartin> | Yeah |
| 21:27 | <&McMartin> | Most of those people were a cost center anyway though~ |
| 21:29 | <&Reiver> | How big a drop? Like, problematic percentage wise? |
| 21:30 | <&Reiver> | ... actually, scratch that, I now have a Terrible Computer Problem to solve |
| 21:30 | <&Reiver> | We are setting up an ETL process to dump table data out of one system so it can be loaded into a SQL Server database. I have talked on this before. |
| 21:30 | <&Reiver> | We have a new and terrible wrinkle |
| 21:31 | <&Reiver> | Our source system does not allow you to configure the CSV file. It is a simple comma seperated, no quotations, no escape characters, naieve table dump. |
| 21:31 | <&Reiver> | The data we are dumping includes free text comment fields. |
| 21:31 | <&Reiver> | People do in fact use punctuation. >_< |
| 21:32 | <@Tamber> | ...noooo, really?! |
| 21:32 | <&Reiver> | JFC this is meant to be enterprise level software guys what the hell |
| 21:33 | <&Reiver> | I am now exploring our other options. There are not many. |
| 21:33 | <&Reiver> | The most promising one is the existence of a user-specified Delimited File format, in which we can specify what character makes up a delimeter |
| 21:33 | <&McMartin> | re: percentage: zero projects I actually cared about moved, at any rate |
| 21:33 | <&Reiver> | I do not trust this system to be 100% unicode compliant. |
| 21:34 | <&Reiver> | ... does a text file break if I start using, I dunno, the ACK ascii code as a field delimeter? I'm trying to find something that's not on the keyboard, so to speak. |
| 21:38 | < Mahal> | I feel very strongly like you need to bring the CSV file from the Awful System through some kind of programmatic scrub before inserting it into it's new SQL home. |
| 21:38 | <&Reiver> | And how do you propose to do that, though |
| 21:38 | <&Reiver> | Commas in text will be naievely present |
| 21:38 | <&McMartin> | Can you at least rely on newlines being valid |
| 21:39 | <&McMartin> | Like, can you determine where records begin an end |
| 21:39 | <&McMartin> | If so a first pass will let you get the records that *don't* have commas. |
| 21:39 | <&Reiver> | ... oh yes we were having problems with that one too weren't we |
| 21:39 | <&Reiver> | These 'records' are tables with 50+ columns, fwiw |
| 21:40 | <&McMartin> | For the first pass that is ideally an implementation detail |
| 21:40 | <&McMartin> | That *said* |
| 21:40 | <&Reiver> | The enterprise software solved the customisability problem by throwing in, eg, TEXT1 TEXT2 TEXT3... TEXT10 for text, date, value, and sometimes even bit and key |
| 21:41 | <&McMartin> | At what point does "screw the 'export' format, reverse engineer what it actually uses, that knows where records and their fields begin and end, convert from that" become feasible |
| 21:41 | <&Reiver> | Would have already done it were it not for the fact this solution is going to have to go into the ~cloud~ where we have no DB access whatsoever and need to use their own system calls |
| 21:41 | <&Reiver> | Because the ~cloud~ is an upgrade, remember |
| 21:42 | <&Reiver> | I would have sent my CV to that data scientist job were it not that I am woefully unqualified and have waaay too much going on to try and transfer employments -_- |
| 21:42 | <&McMartin> | Right, my question is "do you have no backups" |
| 21:42 | <&McMartin> | Or did you copy it up and now all changes since are Lost Forever |
| 21:42 | <&Reiver> | We will recieve, at maximum, one monthly backup per month |
| 21:43 | <&Reiver> | What I am trying to set up is, effectively, a synchronisation system so we can keep that backup up to date in between so we can run all our /other/ systems off of the ghost skeleton |
| 21:43 | <&[R]> | This sounds enterprise-level quality |
| 21:43 | <&Reiver> | So much quality! |
| 21:43 | <&Reiver> | The official solution is to just use all their modules for their everything-else. |
| 21:43 | <&[R]> | "Let's randomly make absurd restrictions for no sane reason." |
| 21:44 | <&Reiver> | Including the ones that are "Coming in six months", several of which have been coming for quite a few six months now |
| 21:44 | <&[R]> | Ouch |
| 21:44 | <&Reiver> | pft, it's hardly insane to want to set up a walled garden and then control your clients data |
| 21:44 | <&Reiver> | That's how you boost customer retention, right? |
| 21:45 | <&McMartin> | To answer your earlier question: ASCII should preserve all characters in the 0-127 range. |
| 21:46 | <&McMartin> | Be aware, however, that character code 26 has the traditional meaning "End of File" |
| 21:46 | <&McMartin> | If I get to pick my separator character, NUL is the traditional one. |
| 21:47 | <&Reiver> | ... including when loading data into a SQL server? |
| 21:47 | <&Reiver> | 'cuz NULL is not exactly an uncommon character in that particular set |
| 21:47 | <&McMartin> | It doesn't matter in general if it's not used in specific |
| 21:48 | <&McMartin> | But yes, I guess the cheapest result would be "take the CSV export, search the text file for any byte in the 0-127 range that is never used and that isn't 26" |
| 21:48 | | * Reiver nods |
| 21:48 | <&Reiver> | oh hey |
| 21:48 | <&Reiver> | 036 Record Seperator seems like it would have some vermissiltude if it isn't used in any files anywhere |
| 21:49 | <&McMartin> | Wow, octal |
| 21:51 | <&Reiver> | ? |
| 21:59 | <&McMartin> | "036" is being expressed there in base 8. I do not often encounter base 8. |
| 22:00 | <&McMartin> | (dec 36 is "$", and hex 0x36 is "6".) |
| 22:21 | <&ToxicFrog> | Reiver: yeah, I generally use file/group/record/unit separator for this, or stx/etx if they're unavailable |
| 22:22 | <&ToxicFrog> | No-one seems to use them in practice |
| 22:23 | <&McMartin> | Unless your tables include raw binary blobs |
| 22:24 | < Mahal> | This is genuinely sounding like a case for "import data in-house, massage, export to cloud" rather than getting stuck on how to make it work |
| 22:25 | <&Reiver> | lol, right |
| 22:25 | <&Reiver> | Mahal: Except the needed path is the opposite |
| 22:25 | < Mahal> | don't get me wrong, it's an ass of a problem |
| 22:25 | <&Reiver> | We need to get data /from/ the cloud to internal systems |
| 22:25 | <&Reiver> | And teh cloud is the one with the shit tools |
| 22:25 | <&Reiver> | so |
| 22:25 | < Mahal> | whatever, export from cloud, massage inhouse, then export to other internal systems' |
| 22:25 | < Mahal> | the point remains |
| 22:25 | <&Reiver> | Yes, that is what I am doing |
| 22:26 | <&Reiver> | Step 1: Get the stuff down from cloud in a recoverable format~ |
| 22:27 | | * McMartin looks at the "very long term history" stuff from the TIOBE programming language index, which attempts to measure how much programming languages figure into discussions of programming |
| 22:27 | <&McMartin> | COBOL has a gigantic spike right around 1999~ |
| 22:28 | < Mahal> | Y2K right? |
| 22:28 | <@Alek> | yep. ;_; |
| 22:29 | <&McMartin> | It ended up #3 that year behind C and C++ |
| 22:30 | <&McMartin> | (Current top 3 are Java, C, and Python.) |
| 22:30 | <&McMartin> | Assembly language, interestingly enough, has a pretty solid position varying through the 11-20 slots, with occasional forays outside of it |
| 22:31 | <&Reiver> | Some days you just gotta get right into the metal. |
| 22:31 | <&McMartin> | Yep |
| 22:31 | <&McMartin> | Also it's taught as part of the grounding classes for everything else |
| 22:52 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has quit [Operation timed out] |
| 22:57 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has joined #code |
| 22:57 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 23:06 | | himi [sjjf@Nightstar-v37cpe.internode.on.net] has joined #code |
| 23:06 | | mode/#code [+o himi] by ChanServ |
| 23:06 | <~Vorntastic> | I don't know assembly for any real architectures |
| 23:20 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has quit [Ping timeout: 121 seconds] |
| 23:25 | <&McMartin> | ARM32 is pleasingly wacky |
| --- Log closed Tue Jan 08 00:00:35 2019 |