X-Git-Url: http://www.chiark.greenend.org.uk/ucgi/~yarrgweb/git?p=ypp-sc-tools.db-live.git;a=blobdiff_plain;f=yarrg%2Fyppedia-ocean-scraper;h=476c1cd9b381bb0d5d5ec1979cc4928bff9fa9f4;hp=5f9c5f007b46fda4265ccc7492dc2b3d58e079f0;hb=d7465beff921821bf120a25a30a35ef06bddfc0e;hpb=ff016739a385102c8e11e80757d5404745bd9a39

diff --git a/yarrg/yppedia-ocean-scraper b/yarrg/yppedia-ocean-scraper
index 5f9c5f0..476c1cd 100755
--- a/yarrg/yppedia-ocean-scraper
+++ b/yarrg/yppedia-ocean-scraper
@@ -42,6 +42,23 @@ import re as regexp
 from optparse import OptionParser
 from BeautifulSoup import BeautifulSoup
 
+
+# For fuck's sake!
+import codecs
+import locale
+def fix_stdout():
+    sys.stdout = codecs.EncodedFile(sys.stdout, locale.getpreferredencoding())
+    def null_decode(input, errors='strict'):
+        return input, len(input)
+    sys.stdout.decode = null_decode
+# From
+#  http://ewx.livejournal.com/457086.html?thread=3016574
+#  http://ewx.livejournal.com/457086.html?thread=3016574
+# lightly modified.
+# See also Debian #415968.
+fix_stdout()
+
+
 ocean = None
 soup = None
 opts = None
@@ -62,8 +79,7 @@ def fetch():
 	debug('fetching',url)
 	dataf = urllib2.urlopen(url)
 	debug('fetched',dataf)
-	soup = BeautifulSoup(dataf,
-		convertEntities=BeautifulSoup.HTML_ENTITIES)
+	soup = BeautifulSoup(dataf)
 
 
 title_arch_re = regexp.compile('(\\S.*\\S) Archipelago \\((\\S+)\\)$')
@@ -86,13 +102,17 @@ def title_arch_ok(t):
 def parse_chart():
 	ta = soup.find('textarea')
 	debug('ta',ta)
-	rc = ta.renderContents()
-	debug('rc',rc)
-	return rc
+	s = ta.string
+	debug('s',s)
+	s = regexp.sub(r'\&lt\;', '<', s)
+	s = regexp.sub(r'\&gt\;', '>', s)
+	s = regexp.sub(r'\&quot\;', '"', s)
+	s = regexp.sub(r'\&amp\;', '&', s)
+	debug('s',s)
+	return s
 
 def parse_ocean():
-	firstarch = soup.find('a', attrs = {'title': title_arch_ok})
-	debug('fa',firstarch)
+	content = soup.find('div', attrs = {'id': 'content'})
 
 	def findall_title_arch_ok(t):
 		return t.findAll('a', attrs = {'title': title_arch_ok})
@@ -101,7 +121,7 @@ def parse_ocean():
 		if u.name != 'table': return False
 		return len(findall_title_arch_ok(u)) > 1
 
-	archestable = firstarch.findParent('table', attrs={'border':'1'})
+	archestable = content.findChild('table', attrs={'border':'1'})
 	debug('at',archestable)
 
 	archsoups = []