Rosetta Code/Count examples
You are encouraged to solve this task according to the task description, using any language you may know.
Find the total number of programming examples for each task and the total for all tasks.
Essentially, count the number of occurrences of =={{header| on each task page.
Output:
<lang>100 doors: 20 examples. 99 Bottles of Beer: 29 examples. Abstract type: 10 examples.
Total: X examples.</lang>
AutoHotkey
<lang AutoHotkey>UrlDownloadToFile
, http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&cmtitle=Category:Programming_Tasks&cmlimit=500&format=xml , tasks.xml
FileRead, tasks, tasks.xml pos = 0 quote = " ; " regtitle := "<cm.*?title=" . quote . "(.*?)" . quote While, pos := RegExMatch(tasks, regtitle, title, pos + 1) {
UrlDownloadToFile , % "http://www.rosettacode.org/w/index.php?title=" . title1 . "&action=raw" , task.xml FileRead, task, task.xml RegExReplace(task, "\{\{header\|", "", count) current := title1 . ": " . count . " examples.`n" output .= current TrayTip, current, % current
} MsgBox % output Return</lang>
C#
Object-oriented solution.
<lang csharp>using System; using System.Collections.Generic; using System.Linq; using System.Text.RegularExpressions; using System.Net;
class Task {
private string _task; private int _examples;
public Task(string task, int examples) { _task = task; _examples = examples; }
public string Name { get { return _task; } }
public int Examples { get { return _examples; } }
public override string ToString() { return String.Format("{0}: {1} examples.", this._task, this._examples); }
}
class Program {
static List<string> GetTitlesFromCategory(string category, WebClient wc) { string content = wc.DownloadString( String.Format("http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&cmtitle=Category:{0}&cmlimit=500&format=json", category) );
return new Regex("\"title\":\"(.+?)\"").Matches(content).Cast<Match>().Select(x => x.Groups[1].Value.Replace("\\/", "/")).ToList(); }
static string GetSourceCodeFromPage(string page, WebClient wc) { return wc.DownloadString( String.Format("http://www.rosettacode.org/w/index.php?title={0}&action=raw", page) ); }
static void Main(string[] args) { WebClient wc = new WebClient(); List<Task> tasks = new List<Task>(); List<string> tasknames = GetTitlesFromCategory("Programming_Tasks", wc);
foreach (string task in tasknames) { string content = GetSourceCodeFromPage(task, wc); int count = new Regex("=={{header", RegexOptions.IgnoreCase).Matches(content).Count; Task t = new Task(task, count);
Console.WriteLine(t); tasks.Add(t); }
Console.WriteLine("\nTotal: {0} examples.", tasks.Select(x => x.Examples).Sum()); }
}</lang>
D
<lang D> import tango.io.Stdout; import tango.net.http.HttpClient; import tango.net.http.HttpHeaders; import tango.text.xml.Document; import tango.text.Util;
alias HttpHeader.ContentLength CL;
auto url = "http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&cmtitle=Category:Programming_Tasks&cmlimit=500&format=xml"; void main() {
auto client = new HttpClient (HttpClient.Get, url); client.open(); char[] mainData, tmp; int total, i;
void cat(void[] content) { tmp ~= cast(char[]) content; }
if (client.isResponseOK) { client.read(&cat, client.getResponseHeaders.getInt(CL)); mainData = tmp; tmp = null;
auto doc = new Document!(char); doc.parse(mainData); foreach (n; doc.query.descendant("cm").attribute("title")) { auto subClient = new HttpClient(HttpClient.Get, "http://www.rosettacode.org/w/index.php?title=" ~ replace(n.value.dup, ' ', '_') ~ "&action=raw"); subClient.open(); if (! subClient.isResponseOK) { Stderr (client.getResponse); break; } subClient.read(&cat, subClient.getResponseHeaders.getInt(CL)); foreach (segment; patterns(cast(char[])tmp, "=={{header|")) i++; --i; if (i) --i; Stdout.formatln ("{0,-40} - {}", n.value, i); total += i; tmp = null; i = 0; } Stdout("total examples: ", total).newline; } else { Stderr (client.getResponse); }
} </lang>
J
Solution:
Using getHTTP
from Web Scraping and utilities from Find unimplemented tasks.
<lang j>
getAllTaskSolnCounts=: monad define
tasks=. getTasks 'Programming_Tasks' counts=. getTaskSolnCounts &> tasks tasks;counts
)
getTaskSolnCounts=: monad define
makeuri=. 'http://www.rosettacode.org/w/index.php?title=' , ,&'&action=raw' wikidata=. getHTTP makeuri urlencode y ([: +/ '{{header|'&E.) wikidata
)
formatSolnCounts=: monad define
'tasks counts'=. y tasks=. tasks , &.>':' res=. ;:^:_1 tasks ,. (8!:0 counts) ,. <'examples.' res , 'Total examples: ' , ": +/counts
) </lang>
Example Usage: <lang j>
formatSolnCounts getAllTaskSolnCounts
100 doors: 46 examples. 99 Bottles of Beer: 62 examples. Abstract type: 17 examples. Ackermann Function: 52 examples. ... </lang>
Java
<lang java5> import java.util.ArrayList; import java.util.Iterator; import ScreenScrape;
public class CountProgramExamples { private static final String baseURL = "http://rosettacode.org/wiki/"; private static final String rootURL = "http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&cmtitle=Category:Programming_Tasks&cmlimit=500&format=xml"; private static final String taskBegin = "title=\""; private static final String taskEnd = "\""; private static final String exmplBegin = ""; private static final String exmplEnd = ""; private static final String editBegin = "";
/** * @param args */ public static void main(String[] args) { // Setup variables int exTotal = 0; int exSubTot = 0; String title = ""; String taskPage = ""; int startPos = 0; String countStr = ""; try { // Get root query results ArrayList<String> tasks = new ArrayList<String>(); ScreenScrape ss = new ScreenScrape(); String rootPage = ss.read(rootURL); while(rootPage.contains(taskBegin)){ rootPage = rootPage.substring(rootPage.indexOf(taskBegin)+taskBegin.length()); title = rootPage.substring(0, rootPage.indexOf(taskEnd)); if (!title.contains("Category:")) { tasks.add(title); } rootPage = rootPage.substring(rootPage.indexOf(taskEnd)); } // Loop through each task and print count Iterator<String> itr = tasks.iterator(); while(itr.hasNext()) { title = itr.next().replaceAll("'","'"); taskPage = ss.read(baseURL+title.replaceAll(" ", "_")); if (taskPage.contains(exmplBegin)) { startPos = taskPage.lastIndexOf(exmplBegin)+exmplBegin.length(); countStr = taskPage.substring(startPos, taskPage.indexOf(exmplEnd, startPos)); exSubTot = Integer.parseInt(countStr.contains(".") ? countStr.substring(0,countStr.indexOf(".")) : countStr); }else{ exSubTot = 0; while(taskPage.contains(editBegin)) { taskPage = taskPage.substring(taskPage.indexOf(editBegin)+editBegin.length()); exSubTot++; } } exTotal += exSubTot; System.out.println(title+": "+exSubTot+" examples."); } // Print total System.out.println("\nTotal: "+exTotal+" examples."); }catch(Exception e){ System.out.println(title); System.out.println(startPos+":"+taskPage.indexOf(exmplEnd, startPos)); System.out.println(taskPage); e.printStackTrace(System.out); } } } </lang>
This is the ScreenScrape class imported in the above class.
OCaml
execute with:
ocaml str.cma unix.cma -I +pcre pcre.cma -I +netsys netsys.cma -I +equeue equeue.cma \ -I +netstring netstring.cma -I +netclient netclient.cma -I +xml-light xml-light.cma countex.ml
<lang ocaml>let repl_quote s =
let reg = Str.regexp_string "'" in (Str.global_replace reg "%27" s)
let repl_space s =
let s = String.copy s in for i = 0 to pred(String.length s) do if s.[i] = ' ' then s.[i] <- '_' done; (s)
let count_ex s =
let pat = Str.regexp_string "=={{header|" in let rec aux n p = try let p = Str.search_forward pat s p in aux (n+1) (p+1) with Not_found -> (n) in aux 0 0
let () =
let url = "http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&\ cmtitle=Category:Programming_Tasks&cmlimit=500&format=xml" in
let str = Http_client.Convenience.http_get url in let xml = Xml.parse_string str in
let total = ref 0 in at_exit (fun () -> Printf.printf "\n Total: %d\n" !total);
let f = function | Xml.Element ("cm", attrs, _) -> let _title = List.assoc "title" attrs in let title = repl_space _title in let title = repl_quote title in let url = "http://www.rosettacode.org/w/index.php?title="^ title ^"&action=raw" in let str = Http_client.Convenience.http_get url in let n = count_ex str in Printf.printf "%s: %d\n%!" _title n; total := n + !total; | _ -> () in
match xml with | Xml.Element ("api", [], [Xml.Element ("query", [], [Xml.Element ("categorymembers", [], cms)])]) -> List.iter f cms | _ -> ()</lang>
Perl
<lang Perl>
- !/usr/bin/perl -w
use strict ; use LWP::UserAgent ; use HTML::Parser ; use constant DOCROOT => "http://www.rosettacode.org/wiki" ; use constant SOLUTIONROOT => "http://www.rosettacode.org/w/index.php?title=" ; my %tasklist = ( ) ; #key: last part of solution list URL, value: title of solution my $ua = new LWP::UserAgent ; my $url = DOCROOT . "/Category:Programming_Tasks" ; my $request = HTTP::Request->new( 'GET' => "$url" ) ; my $response = $ua->request( $request ) ; my $counted = 0 ; my $total_examples = 0 ; my $solresponse ; my $p = HTML::Parser->new( api_version => 3 ) ; #parser for list of tasks my $q = HTML::Parser->new( api_version => 3 ) ; #parser for solutions by task $p->handler( start => \&process , "tagname , attr" ) ; $q->handler( text => \&langfinder, "text" ) ;
if ( $response->is_success( ) ) {
$p->parse( $response->content( ) ) ; foreach my $task( keys %tasklist ) { $request->uri( SOLUTIONROOT . "$task" . "&action=edit" ) ; $solresponse = $ua->request( $request ) ; if ( $solresponse->is_success( )) { $q->parse( $solresponse->content( ) ) ; if ( $tasklist{$task} ) { print "$tasklist{$task} : $counted examples!\n" ; } $counted = 0 ; $q->eof( ) ; } else { print "Error: " . $solresponse->code( ) . " " . $solresponse->message( ) . "\n" ; } } $p->eof( ) ; print "\nTotal: $total_examples examples.\n" ;
} else {
print "Error " . $response->code( ) . " " . $response->message( ) . "\n" ;
} sub process( ) {
return if shift ne "a" ; my $props = shift ; if ( $props->{href} && $props->{href} =~ m,/wiki/([^:]+), ) { if ( $1 !~ /Category/ ) { $tasklist{ $1 } = $props->{title} ; } }
} sub langfinder( ) {
my $text = shift ; while ( $text =~ /header\|.+\}/g ) { $counted++ ; $total_examples++ ; }
} </lang>
Python
<lang python>import urllib, xml.dom.minidom
x = urllib.urlopen("http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&cmtitle=Category:Programming_Tasks&cmlimit=500&format=xml")
tasks = [] for i in xml.dom.minidom.parseString(x.read()).getElementsByTagName("cm"):
t = i.getAttribute('title').replace(" ", "_") y = urllib.urlopen("http://www.rosettacode.org/w/index.php?title=%s&action=raw" % t) tasks.append( y.read().lower().count("{{header|") ) print t.replace("_", " ") + ": %d examples." % tasks[-1]
print "\nTotal: %d examples." % sum(tasks)</lang>
R
<lang R> library(XML) library(RCurl) doc <- xmlInternalTreeParse("http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&cmtitle=Category:Programming_Tasks&cmlimit=500&format=xml") nodes <- getNodeSet(doc,"//cm") titles = as.character( sapply(nodes, xmlGetAttr, "title") ) headers <- list() counts <- list() for (i in 1:length(titles)){ headersi <- getURL( paste("http://rosettacode.org/mw/index.php?title=", gsub(" ", "_", titles[i]), "&action=raw", sep="") ) countsi <- strsplit(headersi,split=" ")1 countsi <- grep("\\{\\{header", countsi) cat(titles[i], ":", length(countsi), "examples\n") } cat("Total: ", length(unlist(counts)), "examples\n") </lang>
Ruby
First, a RosettaCode module, saved as rosettacode.rb: <lang ruby>require 'open-uri' require 'rexml/document'
module RosettaCode
def self.rc_url(page, query) url = "http://www.rosettacode.org/w/%s?%s" % [ URI.escape(page), URI.escape(query.map {|k,v| "%s=%s" % [k,v]}.join("&")) ] url.gsub(/\+/, '%2B') end
def self.rc_tasks(category) query = { "action" => "query", "list" => "categorymembers", "cmtitle" => "Category:#{category}", "format" => "xml", "cmlimit" => 500, } while true url = rc_url "api.php", query doc = REXML::Document.new open(url)
REXML::XPath.each(doc, "//cm") do |task| yield task.attribute("title").value end
continue = REXML::XPath.first(doc, "//query-continue") break if continue.nil? cm = REXML::XPath.first(continue, "categorymembers") query["cmcontinue"] = cm.attribute("cmcontinue").value end end
end</lang>
Then, we implement the task with: <lang ruby>require 'rosettacode'
total_examples = 0
RosettaCode.rc_tasks("Programming_Tasks") do |task|
url = RosettaCode.rc_url("index.php", {"action" => "raw", "title" => task}) examples = open(url).read.scan("=={{header").length puts "#{task}: #{examples}" total_examples += examples
end
puts puts "Total: #{total_examples}"</lang>
Scala
This was writen for Scala 2.8, but Scala 2.7 can be used with slight modifications to the IO library.
Different than the example for other languages, it parallelizes the reading and counting, and it also encode the URL, because some URLs are now causing problems. These modifications are minor, though.
It was written in script style.
<lang scala>import java.net.{URL, URLEncoder} import scala.io.Source.fromURL
val allTasksURL = "http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&cmtitle=Category:Programming_Tasks&cmlimit=500&format=xml" val allTasks = xml.parsing.XhtmlParser(fromURL(new URL(allTasksURL)))
val regexExpr = "(?i)==\\{\\{header\\|".r def oneTaskURL(title: String) = "http://www.rosettacode.org/w/index.php?title=%s&action=raw" format URLEncoder.encode(title.replace(' ', '_'), "UTF-8") def count(title: String) = regexExpr findAllIn fromURL(new URL(oneTaskURL(title)))(io.Codec.UTF8).mkString length
val counts = for (task <- allTasks \\ "cm" \\ "@title" map (_.text)) yield scala.actors.Futures.future((task, count(task)))
counts map (_.apply) map Function.tupled("%s: %d examples." format (_, _)) foreach println println("\nTotal: %d examples." format (counts map (_.apply._2) sum)) </lang>
Tcl
Using the json
package from
<lang tcl>package require Tcl 8.5 package require http package require json
fconfigure stdout -buffering none
proc get_tasks {category} {
set start [clock milliseconds] puts -nonewline "getting $category members..." set base_url http://www.rosettacode.org/w/api.php set query {action query list categorymembers cmtitle Category:%s format json cmlimit 500} set this_query [dict create {*}[split [format $query $category]]] set tasks [list]
while {1} { set url [join [list $base_url [http::formatQuery {*}$this_query]] ?] set response [http::geturl $url] if {[set s [http::status $response]] ne "ok" || [http::ncode $response] != 200} { error "Oops: url=$url\nstatus=$s\nhttp code=[http::code $response]" } set data [json::json2dict [http::data $response]] http::cleanup $response # add tasks to list foreach task [dict get $data query categorymembers] { lappend tasks [dict get [dict create {*}$task] title] } if {[catch {dict get $data query-continue categorymembers cmcontinue} continue_task] != 0} { # no more continuations, we're done break } dict set this_query cmcontinue $continue_task } puts " found [llength $tasks] tasks in [expr {[clock milliseconds] - $start}] milliseconds" return $tasks
}
- This proc can be replaced by a single regexp command:
- set count [regexp -all "***=$needle" $haystack]
- However this proc is more efficient -- we're dealing with plain strings only.
proc count_substrings {needle haystack} {
set count 0 set idx 0 while {[set idx [string first $needle $haystack $idx]] != -1} { incr count incr idx } return $count
}
set total 0 foreach task [get_tasks Programming_Tasks] {
set url [format "http://www.rosettacode.org/w/index.php?title=%s&action=raw" [string map {{ } _} $task]] set response [http::geturl $url] if {[set s [http::status $response]] ne "ok" || [http::ncode $response] != 200} { error "Oops: url=$url\nstatus=$s\nhttp code=[http::code $response]" } set count [count_substrings "\{\{header|" [http::data $response]] puts [format "%3d examples in %s" $count $task] http::cleanup $response incr total $count
}
puts "\nTotal: $total examples"</lang>