13:08 <tom[]> can i name data sources?

14:34 <ddfreyne> tom[]: They only have a typo

14:34 <ddfreyne> type

14:35 <ddfreyne> hah

14:35 <ddfreyne> So no

14:35 <ddfreyne> tom[]: What would you need it for?

14:35 <tom[]> dunno

14:35 <tom[]> can't remember

14:36 <tom[]> something odd going with my setup for a static source

14:36 <tom[]> only 5 of the files in the source are being copied

14:36 <tom[]> looking at -d output now

14:40 <tom[]> nanoc appears to just not visit most of the files in the static data source

14:41 <tom[]> it visits the same 5 items many times

15:08 <tom[]> i don't know what to try next

16:21 <bobthecow> tom[]: are you sure you routed all your static data source items?

16:21 <bobthecow> remember that you route them with extensions.

16:21 <tom[]> i got it sorted

16:22 <tom[]> found the undocumented :prifix config item

16:22 <tom[]> i misunderstood the items_root

16:22 <bobthecow> ahh. yeah.

16:23 <bobthecow> i ran into that yesterday with someone else.

16:37 <ddfreyne> Yeah, it is confusing. I shouldn't have based myself off unix mount paths when I implemented that idea

16:38 <ddfreyne> tom[]: assigning names to data sources would be useful to identify where information comes from and how to handle it, I guess

16:39 <tom[]> i'm not sure how it would be better than comments

17:00 <ddfreyne> tom[]: if @item.data_source.name == "cli" ; ... ; end

17:00 <ddfreyne> would be nice to have

17:00 <tom[]> true

17:01 <tom[]> especially if you're getting fancy in custom data sources

18:17 <bobthecow> tom[]: if you're doing fancy things with custom data sources there's nothing stopping you from implementing name in them :)

22:13 <bbs_> hey folks. is there a way to force nanoc to recompile now that the --force flag has been deprecated?

22:14 <bobthecow> bbs_: delete the contents of tmp/

22:14 <bobthecow> having to do that is a sign that you've got something wrong with your site, btw.

22:15 <bbs_> heh. it's because i have data files that affect the output but are not individual content items per se

22:15 <bobthecow> yep. that would be what's wrong.

22:15 <bobthecow> make 'em items.

22:17 <bbs_> 2 arguments against doing that. A) having a mess of individual files is less manageable than a single data file, B) it's easier to build things like a nav or listing parital based on a full data file

22:18 <bobthecow> 1 argument for it: you don't have to make them individual files :)

22:19 <bobthecow> https://gist.github.com/bobthecow/e8327d3b7aa73b8b1c06

22:19 <bobthecow> that stores all my github projects in a single github_items.yaml file.

22:19 <bobthecow> reads them out, and puts makes items out of them.

22:20 <bobthecow> in my case i've got `sync` defined to actually populate the yaml file, since it's just a duplication of github data.

22:20 <bobthecow> but for you, you could have a much simpler data source.

22:20 <bbs_> yeah that's what i'm doing with this sucker. wrote a preprocess method to generate multiple items from a single data file

22:20 <bobthecow> make it an actual data source.

22:20 <bobthecow> like this.

22:21 <bbs_> oooh, ok ok i think i see what you're saying. so make the directory that contains my data file a data source?

22:22 <bobthecow> it doesn't have to be a directory even.

22:22 <bobthecow> look at that github data source...

22:22 <bobthecow> it's just reading out of a single yaml file in my project directory.

22:23 <bobthecow> and generates 121 items out of that file.

22:23 <bobthecow> you *can* do that in a preprocess block, but a proper data source is cleaner...

22:24 <bobthecow> if you just want to fix your preprocess block, the problem is likely that you're not setting the mtime on your items created.

22:24 <bobthecow> so they're never being treated as "changed"

22:24 <bobthecow> i'd set 'em all to the mtime of the data file.

22:27 <bbs_> ah, i see, yeah i would bet mtime is my issue. thanks for the options, did not know what you could do with data sources (besides static). will probably opt for the quick fix now and the actual data source later

22:27 <bobthecow> data sources are pretty rad. i've got several on my site...

22:27 <bobthecow> i believe nanoc.ws has even more.

22:28 <bobthecow> https://github.com/bobthecow/genghisapp.com/blob/master/lib/data_sources/changelog.rb

22:28 <bobthecow> that's probably my favorite data source so far :)

22:29 <bbs_> nifty. i've been very impressed with nanoc's extensibility

22:29 <bobthecow> https://github.com/nanoc/nanoc.ws/tree/master/lib/data_sources

22:29 <bobthecow> yeah, nanoc.ws has a data source that parses yardoc and creates items out of it.

22:33 <bbs_> yeah the more i look at it the more it's just going to be best to make a proper data source. esp since doing so looks straightfoward

22:47 <bobthecow> mind if i ask what your data is?

22:52 <bbs_> actually this is pretty boring, porting over some yml data that was backing a Sinatra site, so just some custom formatted page data basically

22:54 <bobthecow> ahh, gotcha.