Rosetta Code/Rank languages by popularity: Difference between revisions

Content added Content deleted

Inline

Revision as of 00:24, 21 November 2009

Sort most popular programming languages based in number of members in Rosetta Code categories (from http://www.rosettacode.org/w/index.php?title=Special:Categories&limit=800)

Sample output on 30 October 2009:

  1. 351 - Tcl
  2. 314 - Python
  3. 307 - Ruby
  4. 265 - Common Lisp
  5. 265 - OCaml
  6. 263 - C
  7. 260 - J
  8. 258 - Ada
  9. 250 - Perl
 10. 242 - Java
 11. 237 - E
 12. 236 - Haskell
 13. 227 - AutoHotkey
 14. 226 - D
 15. 213 - R
 16. 195 - ALGOL 68
 17. 194 - C++
 18. 185 - Forth
 19. 171 - JavaScript
 20. 161 - Fortran
 ...

Filtering wrong results is optional. You can check against http://www.rosettacode.org/wiki/Special:MostLinkedCategories

ALGOL 68

Works with: ALGOL 68G version mk8+ for Unix and Linux - tested with release mk15-0.8b.fc9.i386 - uses non-standard library routines http content and grep in string.

Note: the routine http content is currently not available on Win32 systems.

This example is incorrect. It does not accomplish the given task. Please fix the code and remove this message.

<lang algol>PROC good page = (REF STRING page) BOOL:

    IF grep in string("^HTTP/[0-9.]* 200", page, NIL, NIL) = 0
    THEN TRUE
    ELSE IF INT start, end;
             grep in string("^HTTP/[0-9.]* [0-9]+ [a-zA-Z ]*", page,
                            start, end) = 0
         THEN print (page[start : end])
         ELSE print ("unknown error retrieving page")
         FI;
         FALSE
    FI;

MODE LISTOFSTRING = STRUCT(REF LINK first, last, INT upb); MODE LINK = STRUCT(STRING value, REF LINK next);

PRIO LISTINIT = 1; OP LISTINIT = (REF LISTOFSTRING new, REF LINK first)REF LISTOFSTRING: (

 new := (first, first, (first IS REF LINK(NIL) | 0 | 1 ));
 new

);

OP +:= = (REF LISTOFSTRING list, []CHAR item)VOID: (

 HEAP LINK new := (STRING(item), REF LINK(NIL));
 IF first OF list IS REF LINK(NIL) THEN
   first OF list := new
 ELSE
   next OF last OF list := new
 FI;
 last OF list := new;
 upb OF list +:= 1

);

OP UPB = (LISTOFSTRING list)INT: upb OF list;

OP ARRAYOFSTRING = (LISTOFSTRING list)[]STRING:(

 [UPB list]STRING out;
 REF LINK this := first OF list;
 FOR i TO UPB list DO out[i] := value OF this; this := next OF this OD;
 out

);

INT match=0, no match=1, out of memory error=2, other error=3;

PROC re split = (STRING re split, REF STRING beetles)[]STRING:(

   LISTOFSTRING out := (NIL, NIL, 0); # LISTINIT REF LINK NIL; #
   INT start := 1, pos, end;
   WHILE grep in string(re split, beetles[start:], pos, end) = match DO
     out +:= beetles[start:start+pos-2];
     out +:= beetles[start+pos-1:start+end-1];
     start +:= end
   OD;
   IF start > UPB beetles THEN
     out +:= beetles[start:]
   FI;
   ARRAYOFSTRING(out)
 );

IF STRING reply;

  INT rc =
     http content (reply, "www.rosettacode.org", "http://www.rosettacode.org/w/index.php?title=Special:Categories&limit=500", 0);
  rc /= 0 OR NOT good page (reply)

THEN print (("Error:",strerror (rc))) ELSE

 STRING  # hack: HTML should be parsed by an official HTML parsing library #
   re html tag = "<[^>]*>",
   re a href category = "^<a href=""/wiki/Category:.*"" title=",
   re members = "([1-9][0-9]* members)";

 MODE STATISTIC = STRUCT(INT members, STRING category);
 FLEX[0]STATISTIC stats;

 OP +:=  = (REF FLEX[]STATISTIC in out, STATISTIC item)VOID:(
     [LWB in out: UPB in out+1]STATISTIC new;
     new[LWB in out: UPB in out]:=in out;
     new[UPB new]:=item;
     in out := new
   );

hack: needs to be manually maintained #

 STRING re ignore ="Programming Tasks|WikiStubs|Maintenance/OmitCategoriesCreated|"+
                   "Unimplemented tasks by language|Programming Languages|"+
                   "Solutions by Programming Language|Implementations|"+
                   "Solutions by Library|Encyclopedia|Language users|"+
                   "Solutions by Programming Task|Basic language learning|"+
                   "RCTemplates|Language Implementations";

 FORMAT category fmt = $"<a href=""/wiki/Category:"g""" title=""Category:"g""""$;
 STRING encoded category, category;
 FORMAT members fmt = $" ("g" members)"$;
 INT members;

 FLEX[0]STRING tokens := re split(re html tag, reply);
 FOR token index TO UPB tokens DO
   STRING token := tokens[token index];
   FILE file;
   IF grep in string(re a href category, token, NIL, NIL) = match THEN
     associate(file, token);
     make term(file,"""");
     getf(file, (category fmt, encoded category, category));
     close(file)
   ELIF grep in string(re members, token, NIL, NIL) = match THEN
     IF grep in string(re ignore, category, NIL, NIL) /= match THEN
       associate(file, token);
       getf(file, (members fmt, members));
       stats +:= STATISTIC(members, category);
       close(file)
     FI
   FI
 OD;

 OP < = (STATISTIC a,b)BOOL:
   members OF a < members OF b;

 MODE SORTSTRUCT = STATISTIC;
 PR READ "prelude/sort.a68" PR;

 stats := in place shell sort reverse(stats);

 INT max = 10;
 FOR i TO (UPB stats > max | max | UPB stats) DO
   printf(($g(-0)". "g(-0)" - "gl$,i,stats[i]))
 OD

FI</lang> Output:

1. 233 - Python
2. 222 - Ada
3. 203 - OCaml
4. 203 - C
5. 201 - Perl
6. 193 - Haskell
7. 182 - Java
8. 179 - D
9. 178 - ALGOL 68
10. 160 - Ruby

AutoHotkey

This example is incorrect. It does not accomplish the given task. Please fix the code and remove this message.

<lang AutoHotkey> filedelete, url.txt urldownloadtofile, http://www.rosettacode.org/w/index.php?title=Special:Categories&limit=500, url.txt loop, read, url.txt {

if instr(A_LoopReadLine, "

") { reg = title=\"Category:(.+?)" regexmatch(A_LoopReadLine, reg, name) regexmatch(A_LoopReadLine, "(\d*)\smembers", count) print = %count1% `- %name1% `n %print% } } sort, print, RN msgbox %print% </lang>

AWK

Works with: gawk

This example is incorrect. It does not accomplish the given task. Please fix the code and remove this message.

This solution needs help from external tools to fetch the HTML from rosettacode.org, and also to do a numeric sort <lang sh>printf "GET %s HTTP/1.0\n\n" 'http://www.rosettacode.org/w/index.php?title=Special:Categories&limit=500' | nc www.rosettacode.org 80 | gawk '

   # ignore categories that are not languages
   /Basic language learning/            {next}
   /Encyclopedia/                       {next}
   /Implementations/                    {next}
   /Language Implementations/           {next}
   /Language users/                     {next}
   /Maintenance\/OmitCategoriesCreated/ {next}
   /Programming Languages/              {next}
   /Programming Tasks/                  {next}
   /RCTemplates/                        {next}
   /Solutions by Library/               {next}
   /Solutions by Programming Language/  {next}
   /Solutions by Programming Task/      {next}
   /Unimplemented tasks by language/    {next}
   /WikiStubs/                          {next}

   match($0, /<li.*>([^<]*)<\/a> \((digit:+) members?/, m) {
       lang[++i] = m[2] " - " m[1]
   }

   END {
       len = 0; for (i in lang) len++
       sorter = "sort -rn"
       for (i = 1; i <= len; i++) {
           print lang[i] |& sorter
       }
       close(sorter, "to")
       i = 1
       while((sorter |& getline line) > 0) {
           print i++ ". " line
       }
       close(sorter)
   }

'</lang>

C#

Sorting only programming languages.

This example is incorrect. It does not accomplish the given task. Please fix the code and remove this message.

<lang csharp>using System; using System.Net; using System.Linq; using System.Text; using System.Text.RegularExpressions; using System.Collections; using System.Collections.Generic;

class Program {

   static void Main(string[] args)
   {
       string get1 = new WebClient().DownloadString("http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&cmtitle=Category:Programming_Languages&cmlimit=500&format=json");
       string get2 = new WebClient().DownloadString("http://www.rosettacode.org/w/index.php?title=Special:Categories&limit=500");

       ArrayList langs = new ArrayList();
       Dictionary<string, int> qtdmbr = new Dictionary<string, int>();

       MatchCollection match1 = new Regex("\"title\":\"Category:(.+?)\"").Matches(get1);
       MatchCollection match2 = new Regex("title=\"Category:(.+?)\">.+?</a> \\((\\d+) members\\)").Matches(get2);
       
       foreach (Match lang in match1) langs.Add(lang.Groups[1].Value);

       foreach (Match match in match2) {
           if (langs.Contains(match.Groups[1].Value)) {
               qtdmbr.Add(match.Groups[1].Value, Int32.Parse(match.Groups[2].Value));                 
           }
       }
       
       string[] test = qtdmbr.OrderByDescending(x => x.Value).Select(x => String.Format("{0} - {1}", x.Key, x.Value)).ToArray();

       int count = 1;

       foreach (string i in test)
       {
           Console.WriteLine("{0}. {1}",count,i);
           count++;
       }
   }

}</lang>

Object-orinted solution

This example is incorrect. It does not accomplish the given task. Please fix the code and remove this message.

<lang csharp>using System; using System.Net; using System.Linq; using System.Text; using System.Text.RegularExpressions; using System.Collections.Generic;

class Category {

   private string _title;
   private int _members;

   public Category(string title, int members) {
       _title = title;
       _members = members;
   }

   public string Title {
       get {
           return _title;
       }
   }

   public int Members {
       get {
           return _members;
       }
   }

}

class Program {

   static void Main(string[] args) {
       string get1 = new WebClient().DownloadString("http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&cmtitle=Category:Programming_Languages&cmlimit=500&format=json");
       string get2 = new WebClient().DownloadString("http://www.rosettacode.org/w/index.php?title=Special:Categories&limit=500");

       MatchCollection match1 = new Regex("\"title\":\"Category:(.+?)\"").Matches(get1);
       MatchCollection match2 = new Regex("title=\"Category:(.+?)\">.+?</a> \\((\\d+) members\\)").Matches(get2);

       string[] valids = match1.Cast<Match>().Select(x => x.Groups[1].Value).ToArray();
       List<Category> langs = new List<Category>();

       foreach (Match match in match2) {
           string category = match.Groups[1].Value;
           int members = Int32.Parse(match.Groups[2].Value);

           if (valids.Contains(category)) langs.Add(new Category(category, members));
       }

       langs = langs.OrderByDescending(x  => x.Members).ToList();
       int count = 1;

       foreach (Category i in langs) {
           Console.WriteLine("{0}. {1} - {2}", count, i.Title, i.Members);
           count++;
       }
   }

}</lang>

C++

using g++ under Linux with g++ -lboost_thread -lboost_system -lboost_regex <lang C++>#include <string>

include <boost/regex.hpp>
include <boost/asio.hpp>
include <vector>
include <utility>
include <iostream>
include <sstream>
include <cstdlib>
include <algorithm>
include <iomanip>

struct Sort { //sorting programming languages according to frequency

  bool operator( ) ( const std::pair<std::string,int> & a , const std::pair<std::string,int> & b ) 
     const {

return a.second > b.second ;

} ;

int main( ) {

  try {
     //setting up an io service , with templated subelements for resolver and query
     boost::asio::io_service io_service ; 
     boost::asio::ip::tcp::resolver resolver ( io_service ) ;
     boost::asio::ip::tcp::resolver::query query ( "rosettacode.org" , "http" ) ;
     boost::asio::ip::tcp::resolver::iterator endpoint_iterator = resolver.resolve( query ) ;
     boost::asio::ip::tcp::resolver::iterator end ;
     boost::asio::ip::tcp::socket socket( io_service ) ;
     boost::system::error_code error = boost::asio::error::host_not_found ;
     //looking for an endpoint the socket will be able to connect to
     while ( error && endpoint_iterator != end ) {

socket.close( ) ; socket.connect( *endpoint_iterator++ , error ) ;

     }
     if ( error )

throw boost::system::system_error ( error ) ;

     //we send a request
     boost::asio::streambuf request ;
     std::ostream request_stream( &request ) ;
     request_stream << "GET " << "/mw/index.php?title=Special:Categories&limit=800" << " HTTP/1.0\r\n" ;
     request_stream << "Host: " << "rosettacode.org" << "\r\n" ;
     request_stream << "Accept: */*\r\n" ;
     request_stream << "Connection: close\r\n\r\n" ;
     //send the request
     boost::asio::write( socket , request ) ;
     //we receive the response analyzing every line and storing the programming language 
     boost::asio::streambuf response ;
     std::istream response_stream ( &response ) ;
     boost::asio::read_until( socket , response , "\r\n\r\n" ) ;

boost::regex e( "

<a href=\".+\">([a-zA-Z\\+#1-9]+?)</a>\\s?\$(\\d+) members\$

" ) ; std::ostringstream line ; std::vector<std::pair<std::string , int> > languages ; //holds language and number of examples boost::smatch matches ; while ( boost::asio::read( socket , response , boost::asio::transfer_at_least( 1 ) , error ) ) { line << &response ; if ( boost::regex_search( line.str( ) , matches , e ) ) { std::string lang( matches[2].first , matches[2].second ) ; int zahl = atoi ( lang.c_str( ) ) ; languages.push_back( std::make_pair( matches[ 1 ] , zahl ) ) ; } line.str( "") ;//we have to erase the string buffer for the next read } if ( error != boost::asio::error::eof ) throw boost::system::system_error( error ) ; //we sort the vector entries , see the struct above std::sort( languages.begin( ) , languages.end( ) , Sort( ) ) ; int n = 1 ; for ( std::vector<std::pair<std::string , int> >::const_iterator spi = languages.begin( ) ; spi != languages.end( ) ; ++spi ) { std::cout << std::setw( 3 ) << std::right << n << '.' << std::setw( 4 ) << std::right << spi->second << " - " << spi->first << '\n' ; n++ ; } } catch ( std::exception &ex ) { std::cout << "Exception: " << ex.what( ) << '\n' ; } return 0 ; } </lang> Sample output ( just the "top ten" ): 1. 321 - Python 2. 309 - Ruby 3. 267 - OCaml 4. 261 - Ada 5. 255 - Perl 6. 252 - Haskell 7. 243 - Java 8. 230 - AutoHotkey 9. 226 - D 10. 201 - C++

Perl

Sorting only programming languages.

<lang perl>use MediaWiki::API; my $api = new MediaWiki::API({api_url => 'http://rosettacode.org/mw/api.php'});

my @pairs =

   sort {$b->[1] <=> $a->[1] or $a->[0] cmp $b->[0]}
   map {$_->{title} =~ s/\ACategory://;
        [$_->{title}, $_->{categoryinfo}{size} || 0];}
   values %{$api->api
     ({action => 'query',
       generator => 'categorymembers',
       gcmtitle => 'Category:Programming Languages',
       gcmlimit => 'max',
       prop => 'categoryinfo'})->{query}{pages}};

for (my $n = 1 ; @pairs ; ++$n)

  {my ($lang, $tasks) = @{shift @pairs};
   printf "%3d. %3d - %s\n", $n, $tasks, $lang;}</lang>

Python

This example is incorrect. It does not accomplish the given task. Please fix the code and remove this message.

Works with: Python version 2.6

<lang python>import urllib import re

cont = urllib.urlopen("http://www.rosettacode.org/w/index.php?title=Special:Categories&limit=5000").read() qtdmbr = re.findall("title=\"Category:(.+?)\">.+?</a> $(\d+) members$",cont)

result = [(x,int(y)) for x,y in qtdmbr if not re.search(invalid,x,re.IGNORECASE)]

for n, i in enumerate(sorted(result,key=lambda x: x[1],reverse=True),start=1):

   print "%3d. %3d - %s" % (n, i[1], i[0])</lang>

Sorting only programming languages.

This example is incorrect. It does not accomplish the given task. Please fix the code and remove this message.

<lang python>import urllib,re

key1 = lambda x: int(x[1])

get1 = urllib.urlopen("http://www.rosettacode.org/w/api.php?action=query&list=categorymembers&cmtitle=Category:Programming_Languages&cmlimit=500&format=json").read() get2 = urllib.urlopen("http://www.rosettacode.org/w/index.php?title=Special:Categories&limit=500").read()

langs = re.findall("\"title\":\"Category:(.+?)\"",get1) qtdmbr = re.findall("title=\"Category:(.+?)\">.+?</a> $(\d+) members$",get2)

result = [(x,int(y)) for x,y in qtdmbr if x in langs]

for n, i in enumerate(sorted(result,key=key1,reverse=True)):

   print "%3d. %3d - %s" % (n+1, i[1], i[0])</lang>

Ruby

Works with: Ruby version 1.8.7

Now that there are more than 500 categories, the URL given in the task description is insufficient. I use the RC API to grab the categories, and then count the members of each category.

Uses the RosettaCode module from Count programming examples#Ruby <lang ruby>require 'rosettacode' $stdout.sync=true

langs = [] RosettaCode.rc_tasks("Solutions_by_Programming_Language") {|lang| langs << lang}

langcount = {} langs.each_slice(50) do |sublist|

 newquery = {
   "action" => "query",
   "prop" => "categoryinfo",
   "format" => "xml",
   "titles" => sublist.join("|"),
 }
 newurl = RosettaCode.rc_url("api.php", newquery)

 doc = REXML::Document.new open(newurl)
 REXML::XPath.each(doc, "//page") do |page|
   lang = page.attribute("title").value
   info = REXML::XPath.first(page, "categoryinfo")
   langcount[lang] = info.nil? ? 0 : info.attribute("pages").value.to_i
 end

end

puts "There are #{langcount.length} languages" puts "the top 15:" langcount.sort_by {|key,val| val}.reverse[0,15].each_with_index do |(lang, count), i|

 puts "#{i+1}. #{count} - #{lang.sub(/Category:/, )}"

end</lang> Results

There are 151 languages
the top 15:
1. 326 - Tcl
2. 289 - Python
3. 268 - Ruby
4. 248 - C
5. 243 - Ada
6. 238 - Perl
7. 221 - E
8. 219 - Java
9. 218 - AutoHotkey
10. 217 - OCaml
11. 208 - Haskell
12. 195 - ALGOL 68
13. 186 - D
14. 176 - C++
15. 173 - Forth

Tcl

By web scraping

<lang tcl>package require Tcl 8.5 package require http

set response [http::geturl http://rosettacode.org/mw/index.php?title=Special:Categories&limit=1000]

array set ignore {

   "Basic language learning"           1
   "Encyclopedia"                      1
   "Implementations"                   1
   "Language Implementations"          1
   "Language users"                    1
   "Maintenance/OmitCategoriesCreated" 1
   "Programming Languages"             1
   "Programming Tasks"                 1
   "RCTemplates"                       1
   "Solutions by Library"              1
   "Solutions by Programming Language" 1
   "Solutions by Programming Task"     1
   "Unimplemented tasks by language"   1
   "WikiStubs"                         1

}

foreach line [split [http::data $response] \n] {

   if {[regexp {>([^<]+)</a> \((\d+) member} $line -> lang num]} {
       if {![info exists ignore($lang)]} {
           lappend langs [list $num $lang]
       }
   }

}

foreach entry [lsort -integer -index 0 -decreasing $langs] {

   lassign $entry num lang
   puts [format "%d. %d - %s" [incr i] $num $lang]

}</lang> Produces this output on 31 July 2009 (top 15 entries only):

1. 329 - Tcl
2. 292 - Python
3. 270 - Ruby
4. 250 - C
5. 247 - Ada
6. 238 - Perl
7. 223 - E
8. 221 - Java
9. 220 - AutoHotkey
10. 219 - OCaml
11. 210 - Haskell
12. 197 - ALGOL 68
13. 188 - D
14. 179 - C++
15. 175 - Forth
……

By using the API

Inspired by the Ruby version...

Works with: Tcl version 8.5

Library: tDOM

<lang tcl>package require Tcl 8.5 package require http package require tdom

namespace eval rc {

   ### Utility function that handles the low-level querying ###
   proc rcq {q xp vn b} {

upvar 1 $vn v dict set q action query

       # Loop to pick up all results out of a category query

while 1 { set url http://rosettacode.org/mw/api.php?[http::formatQuery {*}$q] puts -nonewline stderr . ;# Indicate query progress... set token [http::geturl $url] set doc [dom parse [http::data $token]] http::cleanup $token

# Spoon out the DOM nodes that the caller wanted foreach v [$doc selectNodes $xp] { uplevel 1 $b }

# See if we want to go round the loop again set next [$doc selectNodes //query-continue/categorymembers] if {![llength $next]} break dict set q cmcontinue [[lindex $next 0] getAttribute cmcontinue] }

   ### API function: Iterate over the members of a category ###
   proc members {page varName script} {

upvar 1 $varName var set query [dict create cmtitle Category:$page {*}{ list categorymembers format xml cmlimit 500 }] rcq $query //cm item { # Tell the caller's script about the item set var [$item getAttribute title] uplevel 1 $script }

   ### API function: Count the members of a list of categories ###
   proc count {cats catVar countVar script} {

upvar 1 $catVar cat $countVar count set query [dict create prop categoryinfo format xml] for {set n 0} {$n<[llength $cats]} {incr n 40} { dict set query titles [join [lrange $cats $n $n+39] |] rcq $query //page item { # Get title and count set cat [$item getAttribute title] set info [$item getElementsByTagName categoryinfo] if {[llength $info]} { set count [[lindex $info 0] getAttribute pages] } else { set count 0 } # Let the caller's script figure out what to do with them uplevel 1 $script } }

   ### Assemble the bits into a whole API ###
   namespace export members count
   namespace ensemble create

}

Get the list of programming languages

rc members "Solutions by Programming Language" lang {

   lappend langs $lang

}

Get the count of solutions for each, stripping "Category:" prefix

rc count $langs l c {

   lappend count [list [regsub {^Category:} $l {}] $c]

} puts stderr "" ;# Because of the progress dots...

Print the output

puts "There are [llength $count] languages" puts "Here are the top fifteen:" set count [lsort -index 1 -integer -decreasing $count] foreach item [lrange $count 0 14] {

   puts [format "%1\$3d. %3\$3d - %2\$s" [incr n] {*}$item]

}</lang>

UnixPipes

This example is incorrect. It does not accomplish the given task. Please fix the code and remove this message.

<lang bash> echo "GET http://www.rosettacode.org/w/index.php?title=Special:Categories&limit=500 HTTP/1.0\n\n"

 | nc www.rosettacode.org 80 
 | sed -n -e 's,<[^>]*>,,g' -e's,^\([^(]*\)(\([^)]*\) members*) *,\2 - \1,g' -e'/^[0-9]\+./p'
 | sort -rn</lang>