Added wildcard to bika listing search #8
Conversation
Sample print sheet: new fields added
| @@ -594,7 +594,7 @@ def _process_request(self): | |||
| continue | |||
| ##logger.info("Or: %s=%s"%(index, value)) | |||
| if idx.meta_type in('ZCTextIndex', 'FieldIndex'): | |||
| self.Or.append(MatchRegexp(index, value)) | |||
| self.Or.append(MatchRegexp(index, '%s*' % value)) | |||
There was a problem hiding this comment.
How about
\b
Matches the empty string, but only at the beginning or end of a word.
I'm probably missing something, but * doesn't seem to be the right one to match at start of word:
'*'
Causes the resulting RE to match 0 or more repetitions of the preceding RE
Unless you mean to use MatchGlob. It looks like MatchRegexp wants a re object as parameter, but didn't dig to see if it accepts strings too. I'm guessing it does ..
There was a problem hiding this comment.
Thanks for reviewing this Jean. MatchRegexp seems to handle * correctly and when it makes no difference to the result if I use MatchGlob. BTW, this works well for text but not so well for numbers. This client has strings like "1,1,1-Methyl Hydride" but advanced query is not a friend of numbers. Any ideas?
There was a problem hiding this comment.
Hmm that sounds surprising. A regexp like term* would match any string that contains ter with or without m, anywhere in the string (for re.search()) or at the start (for re.match()). That doesn't sound like what was intended.
A glob search for term* would match strings starting with term. (The regexp would also match this, which looks correct, but it would also match terq, which is wrong).
I had a quick look at AdvancedQuery but it doesn't import re or glob. I don't know where the regexp evaluation actually happens.
Regarding numbers, that's on the level of the index implementation, below AdvancedQuery. Text indexes normally filter out things like numbers and stopwords. To nicely match chemical names, you'll need a custom index. This is probably much better done using collective.solr.
I see Bika doesn't use ManagableIndex ..
AdvancedQueryworks best when used together with
Products.ManagableIndexanddm.incrementalsearch.
Some of its features depend on these products, e.g. matching
and incremental filtering. Furthermore, these additional
components can speed up queries by several orders of magnitude.
Added a wildcard to listing search so that words starting with the given text will be matched. This does not solve the issues of (a) matches anywhere in the text or (b) numbers in the text.