c# - How to separate this code in two classes? -


i want code html scraper in c# can links or other targetted strings page want to.

i began , ran straight problem: have no idea how seperate code in classes, can use different search engines.

this current code:

using system; using system.collections.generic; using system.linq; using system.text; using system.threading.tasks; using system.diagnostics; using system.net; using htmlagilitypack;  namespace scraper.components {     class scraper     {         public scraper()         {             webclient client = new webclient();             client.headers.add("user-agent", "mozilla/4.0 (compatible; msie 7.0; windows nt 5.1; .net clr 2.0.50727; .net clr 1.1.4322; .net clr 3.0.04506.30; .net clr 3.0.04506.648)");              htmldocument doc = new htmldocument();             doc.load(client.openread("xxx"));              htmlnode rootnode = doc.documentnode;              htmlnodecollection adnodes = rootnode.selectnodes("//a[@class='ad-title']");              foreach(htmlnode adnode in adnodes) {                 debug.writeline(                     adnode.attributes["href"].value                 );             }         }     } }

my intention seperate whole code below client.headers.add independent class, call example:

scraper scraper = new scraper(new googlese('http://google.com/...'));

or that.

thanks in advance.

just starter idea far, should work.

this beginning of the strategy pattern. may want @ factory pattern generating search engine objects.

namespace scraper.components {     class scraper     {         public scraper(isearchengine engine)         {             webclient client = new webclient();             client.headers.add("user-agent", "mozilla/4.0 (compatible; msie 7.0; windows nt 5.1; .net clr 2.0.50727; .net clr 1.1.4322; .net clr 3.0.04506.30; .net clr 3.0.04506.648)");             engine.search(client);         }      }     class googlese: isearchengine     {         public void search(webclient client){             htmldocument doc = new htmldocument();             doc.load(client.openread("http:\\google.com"));              htmlnode rootnode = doc.documentnode;              htmlnodecollection adnodes = rootnode.selectnodes("//a[@class='ad-title']");              foreach(htmlnode adnode in adnodes) {                 debug.writeline(                     adnode.attributes["href"].value                 );             }         }     class bingse: isearchengine     {         public void search(webclient client){             htmldocument doc = new htmldocument();             doc.load(client.openread("http:\\bing.com"));              htmlnode rootnode = doc.documentnode;              htmlnodecollection adnodes = rootnode.selectnodes("//a[@class='ad-title']");              foreach(htmlnode adnode in adnodes) {                 debug.writeline(                     adnode.attributes["href"].value                 );             }         }     }      interface isearchengine{         void search();     } }

Comments

Post a Comment

Popular posts from this blog

html - How to style widget with post count different than without post count -

i have border below li elements in wordpress widget types appropriate. element inside set block gives them nice easy click full width. the problem, when choose option "show post count" when adding widget such categories or archives displays post count plain text after tag , because tag block pushes post number text line below. whether choose show post count or not show post count still gives exact same element identifyer example "widget_categories". how style post count not on next line still keep links non post count list items full width? here's example on sidebar www.bbmthemes.com/themes/modular/blog-standard/ in order solve problem need remove display:block; for ul.widgets ul a , add in it's place line-height: 33px; what going achieve have same visual effect. what going lose doing links going work on text , not on "box". the other way out of find post count coming , make changes this. example change text count i...
Read more

How to remove text and logo OR add Overflow on Android ActionBar using AppCompat on API 8? -

have researched death , cannot find answer. i have actionbar using appcompat. supporting api v7 (api v8 do). actionbar works api v11. api < v11 has problem of removing icon (and feature) actionbar without supplying overflow. big picture have app logo, app title, , 3 icons squidging same actionbar on low end phone. trying make room! i happy 1 of 2 solutions. either: a method of getting overflow functionality can retrieved pull down. (preferred) method of removing icon , text actionbar below current xml api 11+: <resources> <!-- base application theme api 11+. theme replaces appbasetheme res/values/styles.xml on api 11+ devices. --> <style name="appbasetheme" parent="theme.appcompat.light.darkactionbar"> <item name="android:actionbarstyle">@style/liteactionbarstyle</item> <item name="actionbarstyle">@style/liteactionbarstyle</item> </st...
Read more

IIS->Tomcat Redirect: multiple worker with default -

i trying configure multiple instances of tomcat (7) behind single domain in iis. tomcat instances both run on separate ports. have installed tomcat isapi_rediect.dll , configured iis. everything works redirector, except (you knew coming) being told "all other traffic must mapped 1 of 2 instances (they different applications). more accurate requests except specific pattern need mapped 1 worker , specific pattern (/rpt) needs go other worker. i have 2 workers defined in workers.properties: worker.list=ajp13_1,ajp13_2 worker.ajp13_1.port=8009 worker.ajp13_2.port=8010 and following in uriworkermap.properties: /wxx|/*=ajp13_1 /rpt|/*=ajp13_2 with this, seems fine. except (considerable) number of request not include wxx scope. need map rpt , sub content ajp3_2 , other requests ajp3_1. if add /|*=ajp13_1 then ajp3_2 becomes unreachable , wildchar match maps /rpt/* uri ajp3_1. is there way trying do? appreciate help- have spent time reading same references ...
Read more