09:32 <guardian> morning

09:32 <guardian> if you're into Git CLI but want to integrate with P4Merge: http://goo.gl/YKGvnX

13:15 <guardian> ddfreyne: you removed explanations about regex based rules

13:15 <guardian> ?

13:15 <guardian> can't seem to find that part on http://nanoc.ws/docs/

13:16 <guardian> or is it http://nanoc.ws/docs/reference/rules/ ?

13:16 <guardian> "It can also be a string that contains the * wildcard, which matches zero or more characters. Additionally, it can be a regular expression." <-- imho fails to convey regular expressions have to be given using %r{}

13:35 <ddfreyne> guardian: They don't need to be given with %r{}. You can use // as well

13:36 <ddfreyne> guardian: %r{} is easier because you do not need to escape the / in the identifier

13:42 <guardian> really?

13:50 <guardian> irb(main):017:0> identifier_to_regex '/foo/bar/xxx-.*\.svg/'

13:50 <guardian> => /^\/foo\/bar\/xxx-.(.*?)\.svg\/$/

13:50 <guardian> the fuck

13:51 <guardian> ah yeah you meant identifier_to_regex /foo\/bar\/xxx-.*\.svg\//

13:51 <guardian> because having .* in a string doesn't fly

13:52 <guardian> and now I remember I told you I've witnessed people using .* in strings for rules, seen this in various nanoc sites which sources are available on GitHub

14:11 <ddfreyne> guardian: Well you can have .* in a string, but it doesn't really make much sense

14:13 <ddfreyne> guardian: nanoc 4 patterns will be better in any case :)

14:13 <ddfreyne> guardian: http://nanoc.ws/development/nanoc-4-upgrade-guide/#new-string-pattern-syntax

14:20 <guardian> indeed

14:33 <guardian> ddfreyne: did you review the pull requests further btw?

14:33 <ddfreyne> guardian: Had no time to look into it properly yet

14:33 <guardian> k

14:40 <guardian> first matching rule wins right? and others are ignored

14:40 <guardian> do I still get this right?

14:41 <ddfreyne> guardian: yes

14:41 <guardian> can you spot a defect in this fragment? http://pastebin.com/GbdG826m

14:42 <guardian> I get svg routed :(

14:42 <travis-ci> [travis-ci] nanoc/nanoc-external/dev 31b7af8 Lifepillar: The build passed.

14:42 <ddfreyne> guardian: "/#{images_dir}/*.svg/" likely won't work (unless you use the static data source)

14:42 <guardian> ah yeah

14:42 <guardian> that's the mistake

14:43 <guardian> I'm so stupid at times

14:43 <guardian> thank you

14:43 <guardian> in fact I was doing it with extensions, then went "it's better if I let nanoc rules come into play"

14:43 <guardian> thank you denis

14:44 <ddfreyne> That'll work in nanoc 4 only :)

14:44 <guardian> oh yeah I'm doing so much work around nanoc 3 these days

14:44 <guardian> I'm pushing that nanoc 3 factory into production at work

14:44 <guardian> to document all the things

14:44 <ddfreyne> Nice :)

14:44 <guardian> they seem to be very happy about it

14:45 <ddfreyne> I may use this at work as well!

15:22 <ddfreyne> Hmmm

15:22 <ddfreyne> http://nanoc.ws/development/nanoc-4-upgrade-guide/

15:22 <ddfreyne> What do you think of the indicator?

16:01 <guardian> what do you mean by indicator?

16:22 <ddfreyne> the red/black sign in the side bar

16:22 <guardian> red/black?!

16:22 <guardian> yellow black here

17:14 <ddfreyne> Yellow black I mean, heh

17:26 <guardian> is there a difference between xml = Nokogiri::XML(contents) and xml = Nokogiri::XML.parse(contents)

17:26 <guardian> I find reading Nokogiri's doc difficult

17:26 <guardian> http://nokogiri.org/Nokogiri/XML.html doesn't seem to mention the former syntax

18:30 <guardian> damn I'm in a situation where nanoc compile doesn't exit

18:30 <guardian> and I don't know how to debug

18:54 <guardian> I managed to put nanoc into an infinite loop by when inlining another item's compiled content

18:59 <guardian> I think I'm "achieving" this because I'm trying to inject foo.compiled_content into an item, but foo is filtered with a buggy filter that returns nil

19:14 <guardian> ddfreyne: where does it happen, the fact that changing a helper marks all items as outdated?

19:28 <ddfreyne> guardian: Any change to lib/ marks everything as outdated

19:28 <ddfreyne> That's done in the OutdatednessChecker

19:42 <guardian> even if in my case? http://pastebin.com/kypU63NS

19:42 <guardian> will any change to _factory/lib mark everything as outdated?

19:44 <ddfreyne> guardian: Yes

19:44 <ddfreyne> guardian: There is no way for nanoc to figure out what the impact of a change in lib/ is, so it has to recompile everything

19:51 <guardian> ok

19:51 <guardian> then that's good

19:54 <guardian> well I'm modifying _factory/lib/filters/tidy_svg.rb but my site doesn't recompile stuff

19:54 <guardian> need to find out why

21:30 <guardian> ddfreyne: I found why I don't have my _factory/lib/**/*.rb helper causing recompile

21:31 <guardian> ddfreyne: http://pastebin.com/kypU63NS despite cs.load, I can't find a way to register them into @site.@code_snippets

21:44 <guardian> that's because my fragment that adds more code snippets is eval'ed in the TOPLEVEL_BINDING I guess