Add :smart :noref modes, update README

README.md

@@ -40,6 +40,8 @@ The arguments are:
 - Advanced
 	- `csv`: export to csv
 	- `md`: convert articles to Markdown
+	- `nolink`: drop links, but keeps links' inner text
+	- `noref`: drop items' link
 	- `cache`: only take articles from the cache (ie. don't grab new articles' content), so as to save time
 	- `debug`: to have some feedback from the script execution. Useful for debugging
 	- `theforce`: force download the rss feed

morss/morss.py

@@ -11,6 +11,7 @@ from fnmatch import fnmatch
 import re
 import json
 
+import lxml.etree
 import lxml.html
 
 import feeds
@@ -628,6 +629,16 @@ def After(rss, options):
 			if not options.keep:
 				del item.desc
 
+		if options.nolink and item.content:
+			content = lxml.html.fromstring(item.content)
+			for link in content.xpath('//a'):
+				log(link.text_content())
+				link.drop_tag()
+			item.content = lxml.etree.tostring(content)
+
+		if options.noref:
+			item.link = ''
+
 		if options.md:
 			conv = HTML2Text(baseurl=item.link)
 			conv.unicode_snob = True