I'm trying to scrape some data on average GPA and more from a lot of pages similar to this one:
http://www.ptcas.org/ptcas/public/Listing.aspx?seqn=3200&navid=10737426783
require 'mechanize'
agent = Mechanize.new
page = agent.get('http://www.ptcas.org/ptcas/public/Listing.aspx?seqn=3200&navid=10737426783')
gpa_headers = page.xpath('//h3[contains(text(), "GPA")]')
pp gpa_headers
My issue is that gpa_headers is nil but there is at least one h3 element containing "GPA".
What could be causing this issue? I thought it may be that since the page has dynamic elements that Mechanize had some issue with that yet I can puts page.body and the output includes:
... <h3 style="text-align:center;">GPA REQUIREMENT</h3> ...
Which, by my understanding should be found with the xpath I used.
If there is a better approach to this I would like to know that too.
This looks to be a problem with the DOM structure of the site, as it contains a tag named style which isn't being closed and looks like this:
<td colspan='7'><style='text-align:center;font-style:italic'>The
institution has been granted Candidate for Accreditation status by the
Commission on Accreditation in Physical Therapy Education (1111 North
Fairfax Street, Alexandria, VA, 22314; phone: 703.706.3245; email: <a
href='mailto:accreditation#apta.org'>accreditation#apta.org</a>).
Candidacy is not an accreditation status nor does it assure eventual
accreditation. Candidate for Accreditation is a pre-accreditation
status of affiliation with the Commission on Accreditation in Physical
Therapy Education that indicates the program is progressing toward
accreditation.</td>
as you can see, the td tag closes but the inner style never did.
If you don't need this part of the code I would recommend removing this before trying to work with the entire response. I don't have experience with ruby but I would do something like:
Get the raw body of the response.
Replace the part that matches this regex '(<style=\'.*)</td>' with empty string, or close the tag yourself.
Work with this new response body.
Now you would be able to work with xpath selectors.
eLRuLL gives the source of the problem above. Here is an example of how I fixed the issue:
require 'mechanize'
require 'nokogiri'
agent = Mechanize.new
page = agent.get('http://www.ptcas.org/ptcas/public/Listing.aspx?seqn=3200&navid=10737426783')
mangled_text = page.body
fixed_text = mangled_text.sub(/<style=.+?<\/td>/, "</td>")
page = Nokogiri::HTML(fixed_text)
gpa_headers = page.xpath('//h3[contains(text(), "GPA")]')
pp gpa_headers
This will return the header that I was looking for above:
[#<Nokogiri::XML::Element:0x2b28a8ec0c38 name="h3" attributes=[#<Nokogiri::XML::Attr:0x2b28a8ec0bc0 name="style" value="text-align:center;">] children=[#<Nokogiri::XML::Text:0x2b28a8ec0774 "GPA REQUIREMENT">]>]
A more reliable solution is to work with a HTML5 parser like nokogumbo:
require 'nokogumbo'
doc = Nokogiri::HTML5(page.body)
gpa_headers = doc.search('//h3[contains(text(), "GPA")]')
Related
I'm working on a web-scraping solution that grabs totally different webpages and lets the user define rules/scripts in order to extract information from the page.
I started scraping from a single domain and build a parser based on Nokogiri.
Basically everything works fine.
I could now add a ruby class each time somebody wants to add a webpage with a different layout/style.
Instead I thought about using an approach where the user specifies elements where content is stored using xpath and storing this as a sort of recipe for this webpage.
Example: The user wants to scrape a table-structure extracting the rows using a hash (column-name => cell-content)
I was thinking about writing a ruby function for extraction of this generic table information once:
# extracts a table's rows as an array of hashes (column_name => cell content)
# html - the html-file as a string
# xpath_table - specifies the html table as xpath which hold the data to be extracted
def basic_table(html, xpath_table)
xpath_headers = "#{xpath_table}/thead/tr/th"
html_doc = Nokogiri::HTML(html)
html_doc = Nokogiri::HTML(html)
row_headers = html_doc.xpath(xpath_headers)
row_headers = row_headers.map do |column|
column.inner_text
end
row_contents = Array.new
table_rows = html_doc.xpath('#{xpath_table}/tbody/tr')
table_rows.each do |table_row|
cells = table_row.xpath('td')
cells = cells.map do |cell|
cell.inner_text
end
row_content_hash = Hash.new
cells.each_with_index do |cell_string, column_index|
row_content_hash[row_headers[column_index]] = cell_string
end
row_contents << [row_content_hash]
end
return row_contents
end
The user could now specify a website-recipe-file like this:
<basic_table xpath='//div[#id="grid"]/table[#id="displayGrid"]'
The function basic_table is referenced here, so that by parsing the website-recipe-file I would know that I can use the function basic_table to extract the content from the table referenced by the xPath.
This way the user can specify simple recipe-scripts and only has to dive into writing actual code if he needs a new way of extracting information.
The code would not change every time a new webpage needs to be parsed.
Whenever the structure of a webpage changes only the recipe-script would need to be changed.
I was thinking that someone might be able to tell me how he would approach this. Rules/rule engines pop into my mind, but I'm not sure if that really is the solution to my problem.
Somehow I have the feeling that I don't want to "invent" my own solution to handle this problem.
Does anybody have a suggestion?
J.
I need to post data on a website through a program.
To achieve this I am using Mechanize Nokogiri and Selenium.
Here's my code :
def aeiexport
# first Mechanize is submitting the form to identify yourself on the website
agent = Mechanize.new
agent.get("https://www.glou.com")
form_login_AEI = agent.page.forms.first
form_login_AEI.util_vlogin = "42"
form_login_AEI.util_vpassword = "666"
# this is suppose to submit the form I think
page_compet_list = agent.submit(form_login_AEI, form_login_AEI.buttons.first)
#to be able to scrap the page you end up on after submitting form
body = page_compet_list.body
html_body = Nokogiri::HTML(body)
#tds give back an array of td
tds = html_body.css('.L1').xpath("//table/tbody/tr[position()>1]/td")
# Checking my array of td with some condition
tds.each do |td|
link = td.children.first # Select the first children
if link.html = "2015 32 92 0076 012"
# Only consider the html part of the link, if matched follow the previous link
previous_td = td.previous
previous_url = previous_td.children.first.href
#following the link contained in previous_url
page_selected_compet = agent.get(previous_url)
# to be able to scrap the page I end up on
body = page_selected_compet.body
html_body = Nokogiri::HTML(body)
joueur_access = html_body.search('#tabs0head2 a')
# clicking on the link
joueur_access.click
rechercher_par_numéro_de_licence = html_body.css('.L1').xpath("//table/tbody/tr/td[1]/a[1]")
pure_link_rechercher_par_numéro_de_licence = rechercher_par_numéro_de_licence['href']
#following pure_link_rechercher_par_numéro_de_licence
page_submit_licence = agent.get(pure_link_rechercher_par_numéro_de_licence)
body_submit_licence = page_submit_licence.body
html_body = Nokogiri::HTML(body_submit_licence)
#posting my data in the right field
form.field_with(:name => 'lic_cno[0]') == "9511681"
1) So far what do you think about this code, Do you think there is an error in there
2) This part is the one I am really not sure about : I have posted my data in the right field but now I need to submit it. The problem is that the button I need to click is like this:
<input type="button" class="button" onclick="dispatchAndSubmit(document.JoueurRechercheForm, 'rechercher');" value="Rechercher">
it triggers a javascript function onclick. I am triying Selenium to trigger the click event. Then I end up on another page, where I need to click a few more times.. I tried this:
driver.find_element(:value=> 'Rechercher').click
driver.find_element(:name=> 'sel').click
driver.find_element(:value=> 'Sélectionner').click
driver.find_element(:value=> 'Inscrire').click
But so far I have not succeeded in posting the data.
Could you please tell me if selenium will enable me to do what I need to do. If can I do it ?
At a glance your code can use less indentation and more white space/empty lines to separate the internal logic of AEIexport (which should be changed to aei_export since Ruby uses snake case for method names. You can find more recommendations on how to style ruby code here).
Besides the style of your code, an error I found at the beginning of your method is using an undefined variable page when defining form_login_AEI.
For your second question, I'm not familiar with Selenium; however since it does use a real web browser it can handle JavaScript. Watir is another possible solution.
An alternative would be to view the page source (i.e. in Firebug) and understand what the JavaScript on the page does. Then use Mechanize to follow the link manually.
I'm using Mechanize in a Rails 4 application. I created a new agent to scrape a page:
clienturl = #bid.mozs.where(is_main: true).first.attributes['url']
agent = Mechanize.new
#page = agent.get('http://' + clienturl)
#url = #page.uri
I can do things like get the uri, title and meta description. I'd like to now get the count of images on the page and how many of those images are missing alt attributes. Is this possible with Mechanize?
Do something like this:
require 'mechanize'
agent = Mechanize.new
page = agent.get('http://www.iana.org/domains/reserved')
doc = page.parser
img_count = doc.search('img').size # => 2
img_w_alt_count = doc.search('img[#alt]').size # => 1
img_count - img_w_alt_count # => 1
Nokogiri is the parser inside Mechanize. parser returns an instance of the parsed DOM. From that we can ask Nokogiri to search for all nodes matching a selector. I used a CSS selector, but you can use XPath also; CSS tends to be more readable and less verbose.
search returns a NodeSet, so size tells us how many nodes matched.
I am using a YQL query (the standard example query, with GOOG, YHOO, MSFT and AAPL) to generate XML for all of the available fields. I wanted to scrape the YQL site for the XML output once it is generated using a Ruby script, so that I could run it over and over again for different stocks and store the data somewhere. I haven't finished my script yet, but what I have seems to just not run. Here is the code:
yahoo_finance_scrape.rb
require 'rubygems'
require 'nokogiri'
require 'restclient'
PAGE_URL = "http://developer.yahoo.com/yql/console/"
yql_query = 'use "http://github.com/spullara/yql-tables/raw/d60732fd4fbe72e5d5bd2994ff27cf58ba4d3f84/yahoo/finance/yahoo.finance.quotes.xml"
as quotes; select * from quotes where symbol in ("YHOO","AAPL","GOOG","MSFT") '
if page = RestClient.post(PAGE_URL, {'name' => yql_query, 'submit' => 'Test'})
puts "YQL query: #{yql_query}, is valid"
xml_output = Nokogiri::HTML(page)
lines = xml_output.css('#container #layout-doc #yui-gen3000008 #yui-gen3000009 #yui_3_11_0_3_1393417778356_354
#yui-gen3000015 #yui-gen3000016 div#yui_3_11_0_2_1393417778356_10 #centerBottomView
#outputContainer div#output #outputTabContent #formattedView #viewContent #prexml')
lines.each do |line|
puts line.css('span').map{|span| span.text}.join(' ')
sleep 0.03
end
end
When I run the program, it only prints
"YQL query: use "http://github.com/spullara/yql-tables/raw/d60732fd4fbe72e5d5bd2994ff27cf58ba4d3f84/yahoo/finance/yahoo.finance.quotes.xml"
as quotes; select * from quotes where symbol in ("YHOO","AAPL","GOOG","MSFT") , is valid"
And then just stops. Oh, I am using that Github url because yahoo.finance.quotes was not working, and someone else on Stackoverflow suggested to use it.
If you want to check the css tags, just go to http://developer.yahoo.com/yql/console/ and enter my query and do an inspect element on it. I would post it here, but I don't know how.
The output is just the content of your yql_query var. so this does not help much.
You probably should not put the "use xxxx ax quotes" as a string in your code.
Check out what "someone else" had in mind.
The RestClient.post() method returns a response object. With all HTTP operations, always check the response.code, otherwise you don't know about errors.
response = RestClient.post(...)
puts "HTTP Response code: #{response.code}"
if response.code == 200
page = repsonse.to_str
...
end
According to the Nokogiri website the xml_output.css() method filters like it is a css selector. if you have for example "#container #layout-doc", this means "filter elements with the id 'layout-doc' inside elements of the id 'container' and so on. Is this really what you itend to do? if yes, the last "#prexml" should be enough and much less error-prone, as ids should normally be unique.
I have been trying to write a script that may help me to comment from command line.(The sole reason why I want to do this is its vacation time here and I want to kill time).
I often visit and post on this site.So I am starting with this site only.
For example to comment on this post I used the following script
require "uri"
require 'net/http'
def comment()
response = Net::HTTP.post_form(URI.parse("http://www.geeksforgeeks.org/wp-comments-post.php"),{'author'=>"pikachu",'email'=>"saurabh8c#gmail.com",'url'=>"geekinessthecoolway.blogspot.com",'submit'=>"Have Your Say",'comment_post_ID'=>"18215",'comment_parent'=>"0",'akismet_comment_nonce'=>"70e83407c8",'bb2_screener_'=>"1330701851 117.199.148.101",'comment'=>"How can we generalize this for a n-ary tree?"})
return response.body
end
puts comment()
Obviously the values were not hardcoded but for sake of clearity and maintaining the objective of the post i am hardcoding them.
Beside the regular fields that appear on the form,the values for the hidden fields i found out from wireshark when i posted a comment the normal way.I can't figure out what I am missing?May be some js event?
Edit:
As few people suggested using mechanize I switched to python.Now my updated code looks like:
import sys
import mechanize
uri = "http://www.geeksforgeeks.org/"
request = mechanize.Request(mechanize.urljoin(uri, "archives/18215"))
response = mechanize.urlopen(request)
forms = mechanize.ParseResponse(response, backwards_compat=False)
response.close()
form=forms[0]
print form
control = form.find_control("comment")
#control=form.find_control("bb2_screener")
print control.disabled
# ...or readonly
print control.readonly
# readonly and disabled attributes can be assigned to
#control.disabled = False
form.set_all_readonly(False)
form["author"]="Bulbasaur"
form["email"]="ashKetchup#gmail.com"
form["url"]="9gag.com"
form["comment"]="Y u no put a captcha?"
form["submit"]="Have Your Say"
form["comment_post_ID"]="18215"
form["comment_parent"]="0"
form["akismet_comment_nonce"]="d48e588090"
#form["bb2_screener_"]="1330787192 117.199.144.174"
request2 = form.click()
print request2
try:
response2 = mechanize.urlopen(request2)
except mechanize.HTTPError, response2:
pass
# headers
for name, value in response2.info().items():
if name != "date":
print "%s: %s" % (name.title(), value)
print response2.read() # body
response2.close()
Now the server returns me this.On going through the html code of the original page i found out there is one more field bb2_screener that i need to fill if I want to pretend like a browser to the server.But the problem is the field is not written inside the tag so mechanize won't treat it as a field.
Assuming you have all the params correct, you're still missing the session information that the site stores in a cookie. Consider using something like mechanize, that'll deal with the cookies for you. It's also more natural in that you tell it which fields to fill in with which data. If that still doesn't work, you can always use a jackhammer like selenium, but then technically you're using a browser.