CODEWITHSUNDEEP
pythonselectsplinter
Alex van der Spek
Alex van der Spek

Reputation: 21

splinter select(value) returns Attribute error

Trying to get all the (5) tables from this url.

http://www.molpower.com//VLCWeb/UIAboutMOL/PortScheduleInfo.aspx?pPort=NLRTMDE&pFromDate=01-Oct-2013&pToDate=10-Oct-2013

The drop box for the individual pages I can populate with type(value) but this does not refresh the page. Stepping through the pages with the nextPage button fails after one because the object is not longer attached to the DOM (and I don't know how to get around that in splinter).

Trying to populate the drop down and then selecting it. This returns this error:

Traceback (most recent call last):

File "<stdin>", line 69, in <module>
File "/usr/local/lib/python2.6/dist-packages/splinter/driver/webdriver/__init__.py", line 334, in select
 self.find_by_xpath('//select[@name="%s"]/option[@value="%s"]' % (self["name"], value))._element.click()
File "/usr/local/lib/python2.6/dist-packages/splinter/element_list.py", line 73, in __getattr__
  self.__class__.__name__, name))
AttributeError: 'ElementList' object has no attribute '_element'

I used the code below. Any help most appreciated! 

from splinter import Browser
from lxml.html import parse
from StringIO import StringIO
from time import sleep

url = r'http://www.molpower.com//VLCWeb/UIAboutMOL/PortScheduleInfo.aspx?pPort=NLRTMDE&pFromDate=01-Oct-2013&pToDate=10-Oct-2013'

def _unpack(row, kind = 'td'):
    elts = row.findall('.//%s' %kind)
    return [val.text_content() for val in elts[0:7]]

def parse_schdls_data(table):
    rows = table.findall('.//tr')
    hdrs = _unpack(rows[0], kind = 'th')
    data = [_unpack(r, kind = 'td') for ir, r in enumerate(rows[1:-1]) if ir % 3 == 0]
    return (hdrs, data)

with Browser() as browser:
    browser.visit(url)
    print browser.url

    pages = browser.find_by_tag('option')

    pagevals = [p.value for p in pages]
    maxpagev = max(pagevals)

    inputs = browser.find_by_tag('input')

    '''
    for ip, inp in enumerate(inputs):
        if inp.has_class('btnMRBPageNext'):
            #print ip, inp.value, inp.text
            #Need input 35 for the nextPage 
            inp.click()
    '''

    selects = browser.find_by_tag('select')

    for ns, sel in enumerate(selects):
        if sel.has_class('inputDropDown'):
            print ns, sel.value, sel.text
            sel.type(sel.value)
            sleep(2)

    moldata = list()
    for page in range(len(pagevals)):

        content = browser.html
        parsed = parse(StringIO(content))
        doc = parsed.getroot()
        tables = doc.findall('.//table')
        schdls = tables[91]

        #Get all rows from that table
        rows = schdls.findall('.//tr')
        hdr, data = parse_schdls_data(schdls)
        #print page, data
        moldata.append(data)

        while browser.is_element_not_present_by_tag('select', wait_time = 2):
            pass

        inputs = browser.find_by_tag('input')
        selects = browser.find_by_tag('select')

        #inputs[35].click()

        #selects[0].type(str(page + 1))
        selects[0].select(selects[0].value)

Upvotes: 2

Views: 2206

Answers (2)

Daniel Dai
Daniel Dai

Reputation: 1079

Although you got the reference to the select elements by calling 

selects = browser.find_by_tag('select')

and then call the select method on this elements

selects[0].select(selects[0].value)

But from the traceback we can know, the splinter converts it to find_by_xpath

self.find_by_xpath('//select[@name="%s"]/option[@value="%s"]' % (self["name"], value))._element.click()

Where it assumes the select element has a name, if there is no name for the select, there will be an error.

<html>
<head>
</head>
<body>
<select id='s1'>
<option value='a'> A </option>
<option value='b'> B </option>
<option value='c'> C </option>
</select>

<select id='s2'>
<option value='a'> A </option>
<option value='b'> B </option>
<option value='c'> C </option>
</select>

</body>
</html>

We can reproduce this error with previous HTML page, and after we add names to the 2 select elements, errors gone.

Upvotes: 0

thinker3
thinker3

Reputation: 13301

Because the select element do not have a name, so for example:

browser.find_by_xpath('//select[@id="MRBgvPortScheduleInformation_edlPager"]/option[@value="2"]')._element.click()

Upvotes: 1

Related Questions