Merge remote-tracking branch 'jon/silence-generated'

diff --git a/doc/bugs/ikiwiki_lacks_a_--quiet.mdwn b/doc/bugs/ikiwiki_lacks_a_--quiet.mdwn
index fc9644843b08355c9d0ec86b14bb2108bbb5094b..9c90f82a68dc7ab983bcb19b16afacfe4e0494e3 100644 (file)
--- a/doc/bugs/ikiwiki_lacks_a_--quiet.mdwn
+++ b/doc/bugs/ikiwiki_lacks_a_--quiet.mdwn
@@ -18,3 +18,12 @@ report errors would be nice. -- RichiH
     done
 
 Yes, I can simply redirect the output, but an option would be cleaner, imo. -- Richard
+
+> The output above looks like verbose mode output to me (the scanning/building lines, at least).  Check you haven't enabled it in your setup file by accident. I get the following:
+
+    $ ikiwiki --setup setup
+    successfully generated [cgi]
+    successfully generated [post-update]
+    skipping bad filename [...]
+
+> I've written a patch ([here](https://github.com/jmtd/ikiwiki/tree/silence-generated), pull request sent) that fixes the 'generated...' lines. -- [[Jon]]

diff --git a/doc/bugs/table_plugin_does_not_handle___92__r__92__n_lines_in_CSV_files.mdwn b/doc/bugs/table_plugin_does_not_handle___92__r__92__n_lines_in_CSV_files.mdwn
new file mode 100644 (file)
index 0000000..07a7afb
--- /dev/null
+++ b/doc/bugs/table_plugin_does_not_handle___92__r__92__n_lines_in_CSV_files.mdwn
@@ -0,0 +1,3 @@
+The table plugin seems to be unable to read a CSV file that uses \r\n for line delimiters.
+The same file with \r works fine. The error message is "Empty data".
+--liw

diff --git a/doc/forum/Processing_non-pages.mdwn b/doc/forum/Processing_non-pages.mdwn
new file mode 100644 (file)
index 0000000..23af417
--- /dev/null
+++ b/doc/forum/Processing_non-pages.mdwn
@@ -0,0 +1,7 @@
+I'd like to be able to write a plugin that minifies CSS pages, but the whole plugin mechanism appears to be oriented towards generating HTML pages.  That is, all files appear to be split into "pages with page types" and "pages without page types".  Pages without page types are copied from the source to the destination directory and that's all.  Pages *with* page-types go through the whole gamut: scan, filter, preprocess, linkify, htmlize, sanitize, format, and then they're written as "foo.html".
+
+I could be mistaken, but I don't think registering "css" as a page-type would work.  Sure, I could then process the content to my heart's content, but at the end, my foo.css file would be saved as foo.html, which is NOT what I want.
+
+What I would like would be something in-between, where one could take `foo.css`, process it (in this case, run a minify over it) and output it as `foo.css`.
+
+How?

diff --git a/doc/forum/tags_acting_strangely:_not_picking_up_all_pages__44___not_populating_feeds/comment_2_a47884ffd749df980cd62f4c1e3167ce._comment b/doc/forum/tags_acting_strangely:_not_picking_up_all_pages__44___not_populating_feeds/comment_2_a47884ffd749df980cd62f4c1e3167ce._comment
new file mode 100644 (file)
index 0000000..683c6a5
--- /dev/null
+++ b/doc/forum/tags_acting_strangely:_not_picking_up_all_pages__44___not_populating_feeds/comment_2_a47884ffd749df980cd62f4c1e3167ce._comment
@@ -0,0 +1,10 @@
+[[!comment format=mdwn
+ username="http://lovesgoodfood.com/jason/"
+ nickname="Jason Riedy"
+ subject="Less painful to clone."
+ date="2011-04-21T02:34:15Z"
+ content="""
+http://lovesgoodfood.com/jason/git/JasonsChatter.git
+
+(and maybe this time I'll remember to save my setup)
+"""]]