Camelia, the Perl 6 bug

IRC log for #november-wiki, 2009-03-22

| Channels | #november-wiki index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:14 Tene zarah: tell masak http://blogs.gurulabs.com/stephen/2009/​03/tagspm-for-the-perl-6-web-proj.html
01:14 zarah Tene's link is also http://tinyurl.com/d75ccl
01:14 zarah Consider it noted.
03:20 ihrd joined #november-wiki
03:24 ihrd left #november-wiki
04:04 ihrd joined #november-wiki
05:49 ihrd left #november-wiki
08:54 ihrd joined #november-wiki
08:55 masak joined #november-wiki
09:47 ihrd hi massk
09:48 ihrd I write my first blog post
09:48 masak hey ihrd
09:48 zarah masak: You have new messages. Write '/msg zarah @messages' to read them.
09:48 masak @messages
09:48 zarah masak: 8 h 34 m 19 s ago, Tene said http://blogs.gurulabs.com/stephen/2009/​03/tagspm-for-the-perl-6-web-proj.html
09:48 ihrd http://ihrd.livejournal.com/70022.html
09:48 zarah ihrd's link is also http://tinyurl.com/d32m5p
09:48 masak ihrd: seems you're not the only one who wrote something :)
09:48 * masak reads both posts
09:49 ihrd my perl6.ru account blocked, I do not know why. send mail to the owner. post to my blog for now
09:50 masak ihrd: looks good.
09:53 ihrd nothing new, but while I was writing that I make my suggestions a litle bit clear
09:53 masak re the name Web.pm, I know many people consider it too general and vague, but I still find it very clear and descriptive.
09:54 masak ihrd: aye, it often helps to put things down into words.
09:54 ihrd yes and yes
09:54 masak ihrd: for some reason, Google Translate makes a fantastic job translating your blog entry.
09:54 masak perhaps it's the lack of slang :P
09:55 ihrd :)
09:56 masak I have a mind to work on three things in the coming week: Request/Response from Rack, HTML manipulation from Hpricot, and XML-based templates from Genshi.
09:57 masak at least the first and the third one will be of use in November. the second one will be of use for other things.
10:03 ihrd a lot for one week
10:04 ihrd In our propousal we reserved first week for specification
10:04 ihrd and second two weeks to implement f/w basics
10:06 masak right.
10:07 masak I intend to put the focus on the Rack porting.
10:07 ihrd we need more specification in place to go forward
10:07 masak the other two are indeed too big to fit in only one week.
10:07 masak ihrd: yes, additional specification would be nice.
10:07 masak especially, how the parts will fit together.
10:07 ihrd yes, and Rack port one of the core part
10:07 masak aye.
10:08 masak I want to look a bit more at Rack to see what it does and how it works.
10:08 masak will do that today.
10:08 ihrd good
10:09 ihrd I added 'bedrocks'
10:09 masak I will have two goals: porting Request and Response, and writing something about specification.
10:09 * masak looks
10:09 ihrd fill free to add you ideas
10:10 ihrd (and fix my English)
10:10 masak will do.
10:12 ihrd My plan to the next week: rewiright forest as complite example, try to implement simple routines, review Templete
10:12 masak the English is good; only a slight type s/possibale/possible/.
10:13 masak ihrd: sounds good.
10:13 ihrd I planned Template as simple text-based perl6-embeded template
10:13 masak do you still have plans to do a templater with HTML and code mixed?
10:13 ihrd yes
10:14 masak ah, you just answered that :)
10:14 ihrd not just html, but text
10:14 masak yes, good.
10:14 masak is it OK if I reneame the file 'doc/bedrocks' to 'doc/fundamentals'?
10:14 masak I'm not familiar with the former metaphor.
10:15 ihrd thi is not metaphor
10:15 ihrd just word I found in dictionary
10:15 ihrd 'основы'
10:15 masak I suspect it's a metaphor in Russian, but not in English.
10:15 ihrd it is from geology
10:16 masak aye, but to me it signifies nothing more than the geological meaning.
10:16 masak which makes me wonder why you are talking about stone :)
10:16 ihrd to get down to bedrock — добраться до сути дела  A bedrock principle underlying the First Amendment is that the government may not prohibit the expression of an idea simply because society finds the idea itself offensive — Базовый принцип, лежащий в основе Первой Поправки состоит в том, что правительство не вправе препятствовать выражени
10:16 masak ihrd: 'bedrock principles' is fine.
10:16 ihrd ah
10:16 masak but not just 'bedrocks'.
10:17 ihrd okay, thank you
10:17 masak it's the difference between a noun and an adjective.
10:17 masak np
10:17 ihrd yes
10:17 masak I'm learning Chinese, I know exactly this is the hard part of mastering a foreign language.
10:17 masak the subtle meanings of wprds.
10:17 ihrd but 'to get down to bedrock'
10:18 ihrd here it is noun, no?
10:18 masak hm. yes.
10:18 masak but that is not a metaphor I'm used to hearing.
10:18 masak 'get down to basics', perhaps.
10:20 Tene_ joined #november-wiki
10:20 ihrd okay, dictionary always containt something not so popular in real life
10:21 masak yes, it doesn't reflect how much and when a phrase is used.
10:24 masak http://github.com/masak/web/commit/802​4163531d782e51f9363e6f5c58601a3a81bfb
10:24 zarah masak's link is also http://tinyurl.com/db4lyk
10:28 ihrd > my %d = a => 1, b=>2; my @a; @a.push(%d); @a.perl.say;
10:28 ihrd ["a" => 1, "b" => 2]
10:28 ihrd > my %d = a => 1, b=>2; my @a; @a.push(\%d); @a.perl.say;
10:28 ihrd [{"a" => 1, "b" => 2}]
10:28 ihrd shor answer -- because it works
10:29 masak ah. :)
10:29 masak I'll remove the comment again.
10:29 ihrd but you are right, I think this is should work without \
10:29 masak I'll ask on #perl6
10:44 ihrd I was look on the last November today
10:45 ihrd review Dispatcher, and throw out a lot of old code
10:45 ihrd now it looks very clear
10:46 ihrd but I see a lot of strange "Use of unitialized value" when I try ./test_wiki.sh
10:46 ihrd and cant find out what produce it
10:47 masak ihrd: aye, that mystery has been plaguing us for several months now.
10:47 masak I think the solution will be interesting, but viklund++ has looked really hard and found nothing.
10:51 ihrd looks like something in the BEGIN {}
10:51 masak we have a BEGIN {} ? :P
10:51 * masak looks
10:51 ihrd because I put debug messages on the top of november and it is still throw that first
10:52 masak also, during PIR compilation of November.pm
10:54 ihrd yes, I add BEGIN { say 1; } on the top of wiki, and it is say it first, but say after that do thet after all this messages
10:55 masak does that give any clue about what produces the messages?
10:56 masak I fear that the only way to find out what's causing them is to selectively remove parts of modules until nothing can be removed without making the messages disappear.
10:56 ihrd use November; produce first four
10:57 ihrd simple program with only 'use November;'
10:58 masak well, November.pm pulls in a lot of stuff.
10:59 ihrd yes, I am trying to found who exactly doing this
11:01 ihrd hm... when I put all November 'use something' in one file it produce only three
11:01 masak congratulations on the driver's license.
11:01 ihrd thank you
11:01 ihrd I like to drive car
11:01 masak :)
11:01 ihrd mb markup use produce one
11:02 masak the rakudobug about this will be hilarious, I bet.
11:03 ihrd ah! Tags.pm produce three!
11:05 ihrd my $.page_tags_path  is rw = $r ~ 'data/page_tags/';
11:05 ihrd do that
11:05 masak ah!
11:05 masak and $r, for some reason, isn't initialized?
11:07 ihrd yes
11:07 masak wonder why.
11:07 ihrd my $r = Config.server_root;
11:08 ihrd change to the Config.new.server_root and it works properly
11:09 ihrd ah, thi is because someone change Config
11:09 ihrd old days we have method server_root
11:09 masak rakudo: class A { has $.c = "foo" }; class B { my $.d = A.c }; say B.d
11:09 p6eval rakudo de786f: OUTPUT«Use of uninitialized value␤␤»
11:09 ihrd and can call it on tha proto-object
11:09 masak rakudo: class A { has $.c = "foo" }; class B { my $.d = A.new.c }; say B.d
11:09 p6eval rakudo de786f: OUTPUT«foo␤»
11:09 ihrd but now server_root is argument
11:10 masak ihrd: aye.
11:10 masak ok, so it's not a rakudobug.
11:10 ihrd still have anothe three messages, but I fell I found something similar
11:10 ihrd will find I mean
11:11 masak me too.
11:17 ihrd yes, we have commented         $!config = Config.new;
11:17 ihrd this is in November init
11:20 ihrd hm... not this is wrong point
11:20 masak lunch &
11:54 ihrd fix all this stuff
11:55 ihrd no more "Use of unitialized value"
11:58 ihrd hm... current November x3 slow then it be when I check this last time :(
13:05 masak ihrd: that's not good :/ are you checking via the main page? if so, the slowness is a combination of much content and a slow markup parser.
13:05 masak viklund++ is working on caching.
13:09 ihrd masak: yes, main page
13:09 ihrd but no, markup do not slow it a lot
13:10 ihrd with minimal minus 0.2-0.3 s
13:10 masak ah.
13:10 masak good to hear.
13:10 ihrd November.pm do not compile to .pir
13:10 ihrd ehan I do make
13:11 ihrd I do not see why
13:11 * masak tries
13:14 ihrd a!
13:14 ihrd this is because it do not find ::Escape
13:14 masak :/
13:14 masak that'd be my fault.
13:15 masak I kinda moved it to HTML::Template
13:16 ihrd fixed
13:21 masak thank you.
14:09 ihrd left #november-wiki
15:24 ihrd joined #november-wiki
15:51 Tene joined #november-wiki
15:51 masak Tene: greetings.
16:17 sri_kraih_ joined #november-wiki
16:48 Tene hi masak
16:53 masak Tene: I wrote a Happle test before I went to sleep last night.
16:54 masak haven't committed it yet, but I can easily do that.
17:03 masak there, pushed.
17:04 masak I suspect it'll be quite a big task to port Hpricot, but it'll also start to pay off pretty quickly.
17:04 masak having a HTML parser would be really nice.
17:04 masak but I'll put it on low-prio for the next few weeks.
17:05 masak Tene: anothing thing I thought of, about Tags, is that it'd be really useful if one could work validation in there somewhere.
17:05 masak checking that a <head> element is made directly under a <html> element, and so on.
17:13 Tene Interesting idea.
17:15 Tene I'm still trying to decide how much of a problem it is for me that it's a singleton / non-thread-safe.
17:19 masak I thought of that too.
17:19 masak you can't have the cake and eat it.
17:19 masak either you need to create an object, or you'll have thread issues.
17:20 masak if you go with the object, you could call everything as methods, that might be nice.
17:20 masak given Tags.new { .html { .head { ... }; ... } }
17:20 masak though the risk of clobbering $_ is significant.
17:21 * masak needs to go home in a few minutes
17:24 masak will port more of Rack this evening.
18:20 p6eval joined #november-wiki
18:45 ihrd YAy!
18:52 ihrd left #november-wiki
20:26 szabgab joined #november-wiki

| Channels | #november-wiki index | Today | | Search | Google Search | Plain-Text | summary