wos

SOAP Client for querying the Web of Science database

Description

Web of Science (previously Web of Knowledge) is an online subscription-based scientific citation indexing service maintained by Thomson Reuters.

wos is a python SOAP Client (both API and command-line tool) to query the WOS database in order to get XML data from a query using the WWS access.

Installation

The package has been uploaded to PyPI, so you can install the package using pip:

pip install wos

Usage

You can use the wos command to query the Web of Science API. If you want to access data that needs to be accessed using the premium API, you also have to authenticate using your username and password.

usage: wos [-h] [--close] [-l] [-u USER] [-p PASSWORD] [-s SID]

{query,doi,connect} ...

Query the Web of Science.

positional arguments:

{query,doi,connect} sub-command help

query query the Web of Science. doi get the WOS ID from the DOI. connect connect and get an SID.

optional arguments:

-h, --help	show this help message and exit
--close	Close session.
--proxy PROXY	HTTP proxy
-l, --lite	Wos Lite
-v, --verbose	Verbose

authentication:

API credentials for premium access.

-u USER, --user USER -p PASSWORD, --password PASSWORD -s SID, --sid SID

You can use the WOS Lite API using the --lite parameter (for each query).

You can also authenticate using the session id (SID). In fact the sessions are not closed by the command line utility. Example:

$ wos --user JohnDoe --password 12345 connect
Authenticated using SID: ABCDEFGHIJKLM

$ wos --sid ABCDEFGHIJKLM query 'AU=Knuth Donald' -c1
Authenticated using SID: ABCDEFGHIJKLM
<?xml version="1.0" ?>
<records>
    <REC r_id_disclaimer="ResearcherID data provided by Thomson Reuters">
        <UID>WOS:000287850200007</UID>
        <static_data>
            <summary>
                <EWUID>
                    <WUID coll_id="WOS"/>
                    <edition value="WOS.SCI"/>
                </EWUID>
                <pub_info coverdate="MAR 2011" has_abstract="N" issue="1"
                          pubmonth="MAR" pubtype="Journal" pubyear="2011"
                          sortdate="2011-03-01" vol="33">
                    <page begin="33" end="45" page_count="13">33-45</page>
                </pub_info>
                <titles count="6">
                    <title type="source">MATHEMATICAL INTELLIGENCER</title>
....

$ wos --sid ABCDEFGHIJKLM doi '10.1007/s00283-010-9170-7'
10.1007/s00283-010-9170-7

Check the user_query documentation to understand how to create query strings.

Example

Obviously you can also use the python client programmatically:

from wos import WosClient
import wos.utils

with WosClient('JohnDoe', '12345') as client:
    print(wos.utils.query(client, 'AU=Knuth Donald'))

APIs

In wos 0.1.11+, the WosClient class can access the following APIs.

• retrieve [lite / premium]
• retrieveById [lite / premium]
• search [lite / premium]
• citedReferences [premium]
• citedReferencesRetrieve [premium]
• citingArticles [premium]
• relatedRecords [premium]

[FAQ] I cannot connect ...

I am not affiliated with Thomson Reuters. The library leverages the Web of Science WWS API (Web Services Premium or Lite), which is a paid service offered by Thomson Reuters. This means that your institution has to pay for the Web of Science Core Collection access. The simple registration to Web of Knowledge / Web of Science does not entitle you to access the WWS API service.

So if you receive errors like No matches returned for Username or No matches returned for IP, these errors are thrown directly by the WWS API server. This means that the library is correctly communicating with the server, but you do not have access to the Web Services API. If you think this is an error and you should be entitled to access the services, please do not open an issue here but contact Thomson Reuters support, because the problem is on their side.