Beefy Boxes and Bandwidth Generously Provided by pair Networks
Just another Perl shrine
 
PerlMonks  

last hour of cb

by Tanktalus (Canon)
on Jan 26, 2007 at 20:59 UTC ( #596792=document: print w/ replies, xml ) Need Help??

Updates more-or-less every 5 minutes (except when there is no activity). Extracted from Tanktalus' CB Stats' database. Feedback
Shows the last hour or so, but never more than will fit in 64k, nor over two hours. Other sources of cb history
Last update: Jul 22, 2014 at 22:00 UTC

Previous message (not shown): Jul 22, 2014 at 20:26 UTC
[RonW]davido One of our older projects uses Perl 5.8.8 for building it
[tye]we just barely got away from 5.8.mumble (when we moved the DB off Solaris)
[Voronich]Broke in to a new bottle of Mezcal with $mgr and $coworker and all was right with the world.
[RonW]Current build process uses 5.16.3
[RonW]V, good to hear that
[RonW]I suspect all the builds could be updated to 5.20, but testing that the builds were that same as those done with the older versions of Perl would be a huge project of its own
[tye]Upgrading from 5.8.x was such a big project that we chose to go with a very modern version of Perl. We only recently finished that project. So now we are on 5.10.1. :)
[davido]tye: Well, early adopters pay the price by being the ones who discover the bugs. :)
[davido]I've just been working out a patch to fix the namespace bug in Inline::CPP, and wanted to find a lightweight way to do a preliminary check that users using a new feature will be handing me a legal package name.
[tye]I'd think you could use a trick I like that would work for old Perl and new Perl: /^[^\W\D]\w*$/
[tye]er, /^[^\W\d]\w*$/
[davido]have to add :: in there...
[davido]or just split on :: and verify what comes between.
[davido]er... split on :: and '
[davido]But I think we've verified that \p{ID_Start} and \p{ID_Continue} existed in 5.8. Anyway, it was this article that got me thinking of more complexity than your suggestio
[tye]yeah, I read that and the stack overflow thing it referenced. I think they may have both overlooked the possibility of the simple [^\W\d]. Though I don't understand all of the esoteric bits being discussed.
[tye]or perhaps it should be [^\W0-9]
[davido]I think in support of your suggestion, most people don't understand all of the esoteric bits because they are not used and will probably never end up in a bug report.
[davido]I followed back through the SO post as well, and even the Java-oriented one. So now I'm less of an expert than when I started. :)
[tye]if stackoverflow didn't really hate prolonging discussions (which contributes to me hating to use stackoverflow) I might post such a suggestion to that item
[RonW]The first article was saying that \w amtched some things that weren't valid for Perl identifiers
[davido]Do a meditation here and ping tchrist. We'll certainly get his opinion of it. :)
[davido]For my case, catching an invalid package name would allow me to carp a useful message. Letting a crazy (and unforseen) one slip through would only result in a slightly less helpful message, so no big deal, not worth a ton of complexity.

↑Previous Hour↑
↓Current Hour↓

[tye]I thought it was only saying that \w matches things (specifically 0-9) that can't *start* a Perl identifier.
[davido]that's what I took from it.
[SuicideJunkie]Prolonging discussions? SO doesn't have discussions, just context free answers.
[davido]It would have been really nice if 30 years ago someone had decided that \w should be \i, for "identifier", rather than \w for not a word, but maybe not an identifier either. :)
[davido]alright, got to run.
[tye]tye replied to the first article
[RonW]The system is being slow to show your reply. Else it's waiting for a moderator
[tye]"Your comment is awaiting moderation."
[tye]If I'd made that web site, the message would be "Your comment lacks moderation." (:
[RonW]ok
[thezip]Hmm... postgres has done something unexpected for me... I have a query that does a join on a few tables, to gather costs for various activities. The vendor has also provided a canned report that provides the same information. So I was double checking ...
[thezip]... my query against the canned report, and found that the number of records was different between the two. After further inspection, I found that the dollar amounts for my query were somehow aggregated, and that my totals == the total of the canned ...
[thezip]report for the same activities. The reason I had fewer row was because they were aggregated, but the dollar amounts matched. I did not aggregate anything myself, so why could this have happened?
[thezip]Ah... I think I see it now
[RonW]Way beyond my experience with Postgres
[thezip]I think I've selected the wrong table... One table has line items, and one has aggregations. I chose poorly, I think
[thezip]Actually, I've chosen well, because now I do not need to aggregate, which saves me a step. I do not need the line details for anything
[thezip]... and now I know why one table is named 'receivable', and one is named 'receivableline'
[RonW]Congrats, thezip

Log In?
Username:
Password:

What's my password?
Create A New User
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others imbibing at the Monastery: (10)
As of 2014-07-22 22:07 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    My favorite superfluous repetitious redundant duplicative phrase is:









    Results (129 votes), past polls