Dekker1/Fourmi
Archived
1
0
Fork 0
Code

chemspider source now handles vague search requests

Browse Source
This commit is contained in:
RTB 2014-05-01 12:04:44 +02:00
parent 971552110a
commit 73753a6294
1 changed files with 8 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
FourmiCrawler/sources/ChemSpider.py
View File

@ -205,8 +205,14 @@ class ChemSpider(Source):
sel = Selector(response)
log.msg('chemspider parse_searchrequest', level=log.DEBUG)
sel.register_namespace('cs', 'http://www.chemspider.com/')
csid = sel.xpath('.//cs:int/text()').extract()[0]
# [TODO] - handle multiple csids in case of vague search term
csids = sel.xpath('.//cs:int/text()').extract()
if len(csids) == 0:
log.msg('ChemSpider found nothing', level=log.ERROR)
return
elif len(csids) > 1:
log.msg('ChemSpider found multiple substances, taking first '
'element', level=log.DEBUG)
csid = csids[0]
structure_url = self.website[:-1] + self.structure % csid
extendedinfo_url = self.website[:-1] + self.extendedinfo % csid
log.msg('chemspider URL: %s' % structure_url, level=log.DEBUG)