Add collecting no-prio issues. Restructure. * Move usage examples from collect_issues.py to examples/. Also add a section about this to the readme. This should make it easier to understand how it all works together. As we have two examples now, hardcode the database filenames. * Add a function to create the necessary database command from a list of columns. This makes it easier to work with tracker that have customized their priorities, * Add an example how to collect no-prio issues in a database. Note that the display part is not included. * Cleanup roundup_content_data by removing db file name suggestions and unused "reference db" command. This could be re-added for specific installations if they need it, however it probably should be separated as configuration.

--- a/README.creole	Mon Jul 09 11:44:52 2018 +0200
+++ b/README.creole	Mon Jul 09 12:28:28 2018 +0200
@@ -4,7 +4,26 @@
 
 It is Free Software, check out the file headers.
 
+=== Example
+Run ./demo.py from a roundup-tracker to have a running tracker.
+
+{{{
+cp  examples/collect_demo1.py c1.py
+./c1.py
+# create or change some issues
+./c1.py
+
+# to inspect the database contents
+sqlite3 demo1.db 'select * from issues;'
+
+./display_issues_demo.py >demo1.html
+chromium demo1.html
+}}}
+
+
 === Notes
+examples/collect_demo2.py builds database column names dynamically
+and tracks issues with priorities. The display part is still missing.
 
 When migrating to 3:99c68ebfb3b9, Nov 30 17:46:22 2015
 you need to add the print statements for the content-type header