python and http POST
in [Python]
|
From: galileo228 on 13 Feb 2010 15:16 Thank you all for your responses, and Javier thank you for your longer response. I've just downloaded mechanize and beautifulsoup and will start to play around. From a pure learning standpoint, however, I'd really like to learn how to use the python post method (without mechanize) to go to a webpage, fill in a form, click 'submit', follow the redirect to the results page, and download content. For example, if I go to google.com, use firebug and click on the search bar, the following HTML is highlighted: <input value="" title="Google Search" class="lst" size="55" name="q" maxlength="2048" onblur="google&&google.fade&&google.fade()" autocomplete="off"> So if I were to use the 'post' method, how can I tell from the code above what the ID of the searchbar is? Is it 'value', 'name', or neither? Assuming that the ID is 'name', then to search google for the term 'olypmics' would the proper code be: import httplib2 import urllib data = {'q':'olympics'} body = urllib.urlencode(data) h = httplib2.Http() resp, content = h.request("http://www.google.com", method="POST", body=body) print content; Does content return the content of the 'search results' page? And if not, how do I tell python to do that? Finally, must I transmit headers, or are they optional? Thanks all for your continued help! Matt On Feb 12, 2:59 am, Javier Collado <javier.coll...(a)gmail.com> wrote: > Hello, > > I haven't used httplib2, but you can certainly use any other > alternative to send HTTP requests: > - urllib/urllib2 > - mechanize > > With regard to how do you find the form you're looking for, you may: > - create the HTTP request on your own with urllib2. To find out what > variables do you need to post, you can use tamperdata Firefox addon as > suggested (I haven't used that one) or httpfox (I have and it works > great). > - use mechanize to locate the form for you, fill the data in and click > on the submit button. > > Additionally, you may wan to scrape some data that may be useful for > your requests. For that BeautifulSoup is good solution (with some > Firebug help to visually locate what you're looking for). > > Best regards, > Javier > > P.S. Some examples here:http://www.packtpub.com/article/web-scraping-with-pythonhttp://www.packtpub.com/article/web-scraping-with-python-part-2 > > 2010/2/11 galileo228 <mattbar...(a)gmail.com>: > > > Hey All, > > > Been teaching myself Python for a few weeks, and am trying to write a > > program that will go to a url, enter a string in one of the search > > fields, submit the search, and return the contents of the search > > result. > > > I'm using httplib2. > > > My two particular questions: > > > 1) When I set my 'body' var, (i.e. 'body = {'query':'search_term'}), > > how do I know what the particular key should be? In other words, how > > do I tell python which form on the web page I'm visiting I'd like to > > fill in? Do I simply go to the webpage itself and look at the html > > source? But if that's the case, which tag tells me the name of the > > key? > > > 2) Even once python fills in the form properly, how can I tell it to > > 'submit' the search? > > > Thanks all! > > > Matt > > -- > >http://mail.python.org/mailman/listinfo/python-list > >
From: galileo228 on 13 Feb 2010 15:37
Thank you all for your responses, and Javier thank you for your longer response. I've just downloaded mechanize and beautifulsoup and will start to play around. From a pure learning standpoint, however, I'd really like to learn how to use the python post method (without mechanize) to go to a webpage, fill in a form, click 'submit', follow the redirect to the results page, and download content. For example, if I go to google.com, use firebug and click on the search bar, the following HTML is highlighted: <input value="" title="Google Search" class="lst" size="55" name="q" maxlength="2048" onblur="google&&google.fade&&google.fade()" autocomplete="off"> So if I were to use the 'post' method, how can I tell from the code above what the ID of the searchbar is? Is it 'value', 'name', or neither? Assuming that the ID is 'name', then to search google for the term 'olypmics' would the proper code be: import httplib2 import urllib data = {'q':'olympics'} body = urllib.urlencode(data) h = httplib2.Http() resp, content = h.request("http://www.google.com", method="POST", body=body) print content; Does content return the content of the 'search results' page? And if not, how do I tell python to do that? Finally, must I transmit headers, or are they optional? Thanks all for your continued help! Matt On Feb 12, 2:59 am, Javier Collado <javier.coll...(a)gmail.com> wrote: > Hello, > > I haven't used httplib2, but you can certainly use any other > alternative to send HTTP requests: > - urllib/urllib2 > - mechanize > > With regard to how do you find the form you're looking for, you may: > - create the HTTP request on your own with urllib2. To find out what > variables do you need to post, you can use tamperdata Firefox addon as > suggested (I haven't used that one) or httpfox (I have and it works > great). > - use mechanize to locate the form for you, fill the data in and click > on the submit button. > > Additionally, you may wan to scrape some data that may be useful for > your requests. For that BeautifulSoup is good solution (with some > Firebug help to visually locate what you're looking for). > > Best regards, > Javier > > P.S. Some examples here:http://www.packtpub.com/article/web-scraping-with-pythonhttp://www.packtpub.com/article/web-scraping-with-python-part-2 > > 2010/2/11 galileo228 <mattbar...(a)gmail.com>: > > > Hey All, > > > Been teaching myself Python for a few weeks, and am trying to write a > > program that will go to a url, enter a string in one of the search > > fields, submit the search, and return the contents of the search > > result. > > > I'm using httplib2. > > > My two particular questions: > > > 1) When I set my 'body' var, (i.e. 'body = {'query':'search_term'}), > > how do I know what the particular key should be? In other words, how > > do I tell python which form on the web page I'm visiting I'd like to > > fill in? Do I simply go to the webpage itself and look at the html > > source? But if that's the case, which tag tells me the name of the > > key? > > > 2) Even once python fills in the form properly, how can I tell it to > > 'submit' the search? > > > Thanks all! > > > Matt > > -- > >http://mail.python.org/mailman/listinfo/python-list > > |