将包含unicode的字符串传递给RESTful API

Question

I am firing off 3 separate queries against a RESTful API (I am using python/urllib2/pandas) each query containing one of the following string variants:

(1) 'Caveolin-1 suppresses Human Immunodeficiency Virus-1 replication by inhibiting acetylation of NF-\xce\xbaB'

(2) 'Caveolin-1 suppresses Human Immunodeficiency virus-1 replication by inhibiting acetylation of NF-κB'

(3) 'Caveolin-1 suppresses Human Immunodeficiency virus-1 replication by inhibiting acetylation of NF

Outcomes are:

(1) doesnt return any results (when done programmatically from python)

(2) Works and returns the expected result - the matching record (query fired manually using a Chrome plugin for making RESTful API calls and just paste the string as is into the appropriate part of the API call)

(3) works the same as (2)

Since I have the source data and am doing (1) programmatically from python by reading string from a dataframe, is there any way of doing something (have no idea what) with the unicode characters in my source data (I am guessing thats what '\\xbaB' etc are) to make them passable to the API? '\\xce\\xbaB' seems to be the encoding for 'κB'based on the above.

Or this is this something I am going to have to look at API documentation for (which for this bit doesnt exist I dont think...).

If this is hard/its easier - whats the best way to just get rid of any unicode characters from the string before passing the query (ie fallback to (3))?

Thanks in advance!

REF: from python I am executing the following to come with the API

 api_call = 'http://some_api/index:ABCDE?query=title(' + str(title_string) + ')' headers = {'APIKey': API_key, 'accept':'text/xml, application/atom+xml'} 

request = urllib2.Request(api_call, headers = headers, )
response = urllib2.urlopen(request,'' , 30)

return response.read()

Answer 1

\\xce and \\xba are characters with the hex values ce and ba respectively. Without knowing more about how you're talking to the API or what it's expecting I would think you could do something like this to make the string passable:

>>> urllib.quote('an Immunodeficiency Virus-1 replication by inhibiting acetylation of NF-\xce\xbaB')
'an%20Immunodeficiency%20Virus-1%20replication%20by%20inhibiting%20acetylation%20of%20NF-%CE%BAB'

EDIT:

Normally in python this is how I would add parameters to the URL:

params = {'query' : 'title(' + title_string + ')'}
api_call = 'http://some_api/index:ABCDE?' + urllib.urlencode(params)

So I would lean towards that instead of my earlier urllib.quote suggestion (which I think would be applicable if title_string was part of the path), but I'm not sure if it's enough with the hex values in title_string. I think it will depend on how it's being handled on the server side.

Answer 2

尝试将“ \\”转换为“ \\\\”，因为“ \\ x”表示以十六进制数字开头的十六进制字符代码