corrected issue when searches with "q" gets encoded twice #93

Merged

leandroferreira merged 1 commit from master into master

2012-06-24 12:15:10 -07:00

leandroferreira commented

2012-05-31 07:01:44 -07:00

(Migrated from github.com)

When doing a search, the q parameter gets encoded twice.

Example:

code:
twitter = twython.Twython()
results = twitter.search(q='from:username')

result:
request via GET sending q=from%253Ausername

expected result:
request via GET sending q=from%3Ausername

The value is encoded on search:
kwargs['q'] = urllib.quote_plus(Twython.unicode2utf8(kwargs['q']))

and then on _request:
url = '%s?%s' % (url, urllib.urlencode(params))

I fixed It by passing a safe character to quote_plus:
kwargs['q'] = urllib.quote_plus(Twython.unicode2utf8(kwargs['q']), ':')

When doing a search, the q parameter gets encoded twice. Example: code: twitter = twython.Twython() results = twitter.search(q='from:username') result: request via GET sending q=from%253Ausername expected result: request via GET sending q=from%3Ausername The value is encoded on search: kwargs['q'] = urllib.quote_plus(Twython.unicode2utf8(kwargs['q'])) and then on _request: url = '%s?%s' % (url, urllib.urlencode(params)) I fixed It by passing a safe character to quote_plus: kwargs['q'] = urllib.quote_plus(Twython.unicode2utf8(kwargs['q']), ':')

ryanmcgrath commented

2012-06-24 12:15:14 -07:00

(Migrated from github.com)

Apologies for the delayed merge, just now getting back to normal workflows. Merged and good to go. Thanks!

leandroferreira commented

2012-06-24 14:47:09 -07:00

(Migrated from github.com)

Thanks for the merge, this app is turning out to be really useful for my sites!