Rosetta Code/Run examples
This task is based on an idea hatched from this C1R Implementation.
Write a program that will accept as input the name of a task from Rosetta Code and the name of a language. The program should then download the solution for the specified task in the specified language, present the source to the user and prompt the user to confirm running the example.
The program should verify that the tools needed to compile or run the solution are present before running it. If the solution can not be run, a graceful exit should happen. (i.e. the program should not crash)
Besides its own language, the program should support at least two other languages. (Ideally it would support most of the languages available, but that is too much to ask. Though some languages are simple, e.g. python, pike, perl, bash and several more only need the solution saved in a file and given to the language as argument to run, so it should be easy to support many languages like that).
If you know what is needed to support a particular language, please help to add support for that language to implementations in other languages.
Extra credit: add a function to get a list of all solutions of a given language and run them to create a report on which solutions failed to run.
More credit: also test if the output of a solution compares to a given result. The expected output should be loaded from a file with the name of the task. (This way all implementations can share the same set of files, and anyone can add more files. In the future the content of these files could be stored directly in a section of the task on Rosetta Code itself.)
Liberty BASIC
<lang lb> ' ******************************************************************** ' ** ** ' ** parseAndRun.bas v26b tenochtitlanuk November 2012 ** ' ** ** ' ** select a LB solution from RC site & run it locally ** ' ** ** ' ******************************************************************** 'retrieve proper temporary path and filename to save downloaded HTML: Source$ = GetTempFileName$("htm") 'nomainwin
' Download main RC LB page which has current tasks on it. Save as 'source.html' ' run "C:\Program Files\Mozilla Firefox\firefox.exe http://rosettacode.org/wiki/Category:Liberty_BASIC" 'testing routine print " Fetching current RC page of completed Liberty BASIC RC solutions." 'result = DownloadToFile( "http://rosettacode.org/wiki/Category:Liberty_BASIC", "E:\source.html") result = DownloadToFile( "http://rosettacode.org/wiki/Category:Liberty_BASIC", Source$)
if result <>0 then print "Error downloading LB solved tasks.": end else print: print " Displaying solved tasks.": print
' Load source into a string. Go through and save in a 2D array all topic titles ' and the appropriate web addresses to find them. 'open "E:\source.html" for input as #f open Source$ for input as #f
html$ = input$( #f, lof( #f))
close #f kill Source$ 'remove temp file
dim solutions$( 500, 2)
global count count =1 first =0 last =0 reading =0
' The first topic is the '100 doors' so skip all html jump ref's earlier than this. do
r$ =getHtmlSection$( html$, first, last) if instr( r$, "/rosettacode.org/mw/index.php") then exit do ' We've read all LB solved tasks. if r$ ="wiki/100_doors" then reading =1
if reading =1 then ' we can start recording path & name solutions$( count, 1) ="http://rosettacode.org/" +r$ +"#Liberty_BASIC"
special =instr( r$, "%2B"): if special <>0 then r$ =left$( r$, special -1) +"+" +mid$( r$, special +3) special =instr( r$, "%27"): if special <>0 then r$ =left$( r$, special -1) +"'" +mid$( r$, special +3) special =instr( r$, "%C3%A8"): if special <>0 then r$ =left$( r$, special -1) +chr$( 232) +mid$( r$, special +6) solutions$( count, 0) =mid$( r$, 6) ' we want the bit beyond '/wiki/' if instr( solutions$( count, 0), "/") then newName$ ="" for ii =1 to len( solutions$( count, 0) ) n$ =mid$( solutions$( count, 0), ii, 1) if n$ ="/" then n$ ="_" newName$ =newName$ +n$ next ii solutions$( count, 0) =newName$ end if print count, solutions$( count, 0)'; tab( 60); solutions$( count, 1) count =count +1 end if
loop until 0 print: print count -1; " tasks solved in LB."
'input " Choose task # "; R ' Choose a page to try. for R =1 to 283 print print " Choosing a task at random viz #"; R; " out of "; count -1; " completed in LB." print " Task is "; chr$( 34); solutions$( R, 0); chr$( 34) print
'********************run "C:\Program Files\Mozilla Firefox\firefox.exe " +solutions$( R, 1)
' Fetch the RC task page with all the sol'ns including LB one. print " Downloading the page for this task." result = DownloadToFile( solutions$( R, 1), "rx.html")
if result <>0 then print "Error downloading.": end
print " Now finding the LB section of the html code." ' Now finding the appropriate LB section on this topic.
open "rx.html" for input as #r
L =lof( #r) print " Length of source html of this topic's page is "; L t$ =input$( #r, L)
close #r
preamble$ =">Liberty BASIC</a>" +chr$( 10)
lP =len( preamble$) print " Finding the preamble string at "; beg =instr( t$, preamble$)' +len( preamble$) print beg
lookFor$ ="source" +chr$( 34) +">" beg =instr( t$, lookFor$, beg) ' get to start of BASIC code. beg =beg +len( lookFor$)
print " Found LB section at "; beg;
fin =instr( t$, "", beg)
print " and ending at "; fin
print " Chopping off unwanted earlier & later sections of html source." t$ =mid$( t$, beg, fin -beg) ' discard earlier & later parts of html code.
open solutions$( R, 0) +".txt" for output as #LbText
#LbText t$;
close #LbText
L =len( t$)
print " Relevant html code LB section being parsed for LB BASIC code."
' Read the rest of the LB code section to section ..
LB$ ="" j =1
print " Dropping html tags & translating html entities." print print " LB code follows." print
do
nxtChr$ =mid$( t$, j, 1) select case ' _______________________________________________________________________________________________________ case ( nxtChr$ =chr$( 10)) or ( nxtChr$ =chr$( 13)) j =L print "End reached- CRLF"
case nxtChr$ ="<" ' we've found a html tag. Omit. 'print " Starting a tag with a <"; item$ ="<" do ' keep looking until find a '>' or finish... j =j +1 nxtChr$ =mid$( t$, j, 1) item$ =item$ +nxtChr$ loop until nxtChr$ =">" 'print " Closing a tag with a >."
if item$ ="" then j =L ' end reached
if item$ ="
" then LB$ =LB$ +chr$( 10) ' code for CRLF if item$ ="
" then LB$ =LB$ +chr$( 10) ' code for CRLF, now if j <>L then j =j +1
case nxtChr$ ="&" ' we've found an html entity. ' replace with plain-text equivalents. 'print " html entity starting with & "; select case ' .............................................................................. case mid$( t$, j+1, 5) ="quot;" LB$ =LB$ +chr$( 34): j =j +6 ' &guot; " case mid$( t$, j+1, 3) ="gt;" LB$ =LB$ +">": j =j +4 ' > > case mid$( t$, j+1, 3) ="lt;" LB$ =LB$ +"<": j =j +4 ' < < case right$( mid$( t$, j, 5), 1) =";" v =val( mid$( t$, j +2, 2)): j =j +5 ' 2-digit character-code if v =39 then LB$ =LB$ +chr$( 39) else LB$ =LB$ +chr$( v) ' eg ' ( 40,41) '() case right$( mid$( t$, j, 6), 1) =";" ' 3-digit character-code v =val( mid$( t$, j +2, 3)) if v =160 then v =32 'print "Hard space!" ' convert hard- to soft-space. j =j +6: LB$ =LB$ +chr$( v) end select ' .............................................................................. 'print " and finishing with ;" case else ' not an html entity nor a tag. Use as-is unless it's the final hard-space plus semi-colon..
if mid$( t$, j +1, 5) ="#160;" and mid$( t$, j +5, 6) ="" then j =L else LB$ =LB$ +nxtChr$: j =j +1
end select ' _________________________________________________________________________________________________________ scan
loop until j >= fin -beg -4
print: print LB$
open solutions$( R, 0) +".bas" for output as #LB
#LB LB$;
close #LB
print print " Done"
timer 5000, [on2] wait [on2] timer 0
' Run with LB. ' *************************************run chr$( 34) +"C:\Program Files\Liberty BASIC v4.04\liberty.exe" +chr$( 34) +" -R E:\" +solutions$( R, 0) +".bas" next R end
' **************************************************************
Function DownloadToFile( urlfile$, localfile$)
open "URLmon" for dll as #url calldll #url, "URLDownloadToFileA",_ 0 as long,_ 'null urlfile$ as ptr,_ 'url to download localfile$ as ptr,_ 'save file name 0 as long,_ 'reserved, must be 0 0 as long,_ 'callback address, can be 0 DownloadToFile as ulong '0=success close #url
end function end
function getHtmlSection$( string$, byref first, last)
a =instr( string$, "<a href=" +chr$( 34), first) if a =0 then getHtmlSection$ =" Sorry! html link not found": exit function b =instr( string$, chr$( 34), a +9) getHtmlSection$ =mid$( string$, a +10, b -a -10) first =b +1 ' Reset value of "first" so that in the next call to ' getHtmlSection$( the next html link can be found
end function
function GetTempFileName$(prefix$)
TempPath$=GetTempPath$() TempFile$ = space$(256)+chr$(0)
calldll #kernel32, "GetTempFileNameA",_ TempPath$ as ptr,_ 'directory for temp file prefix$ as ptr,_ 'desired prefix for temp filename 0 as ulong,_ '0=file created,nonzero=you must create file TempFile$ as ptr,_ 'string buffer to hold qualified path and filename result as ulong 'nonzero=success
'TempFile$ holds complete path and filename info GetTempFileName$ = TempFile$ end function
Function GetTempPath$()
CallDLL #kernel32, "GetTempPathA",_ 0 as long,_ _NULL as long,_ length as long
buf$ = space$(length)
CallDLL #kernel32, "GetTempPathA",_ length as long,_ buf$ as ptr,_ ret as long
GetTempPath$ = buf$
End Function </lang>
Perl 6
This is a fairly comprehensive task code runner. It is set up to work for Perl 6 only at this point, but could be easily tweaked to work with other languages. There is so much variation to the task requirements and calling conventions that it would be problematic to make a general purpose, language agnostic code runner. (Heck, a single language one is hard enough.)
By default, this will download the Perl 6 section of any (every) task that has a Perl 6 example, extract the code blocks and attempt to run them. Many tasks require files or user interaction to proceed, others are not complete runnable code blocks (example code fragments), some tasks run forever. To try to deal with and compensate for this, this implementation can load a %resource hash that will: supply input files where necessary, skip unrunnable code fragments, limit long and/or infinite running blocks, supply user interaction code where possible, and skip blocks where user interaction is unavoidable.
The complete implementation is too large and cumbersome to post in it's entirety here, only the main task retrieval and execution code is included.
For the whole ball of wax see the rc-run github repository.
Run with no parameters to run every implemented task on Rosetta Code. Feed it a task name to only download / run that task.
Note: This is set up to run under Linux. It could be adapted for Windows (or OSX I suppose) fairly easily but I don't have access to those OSs, nor do I care to seek it.
<lang perl6>use HTTP::UserAgent; use URI::Escape; use JSON::Fast; use MONKEY-SEE-NO-EVAL;
my $client = HTTP::UserAgent.new; my $url = 'http://rosettacode.org/mw';
- Language specific variables
my $language = 'Perl_6'; # language my $header = 'Perl 6'; # header text my $exe = 'perl6'; # executable to run perl6 in a shell my $ext = '.p6'; # file extension for perl6 code
- weird break in lang tag to placate terminally confused wiki formatter and syntax highlighter
my $tag = rx/<?after '<lang perl6>'> .*? <?before '</' 'lang>'>/; # language entry regex
my $view = 'xdg-open'; # image viewer, this will open default under Linux my %resource = load-resources(); my $download = True;
my @tasks;
run('clear'); note 'Retrieving tasks';
if @*ARGS {
@tasks = |@*ARGS; $download = False;
}
if $download {
@tasks = mediawiki-query( $url, 'pages', :generator<categorymembers>, :gcmtitle("Category:$language"), :gcmlimit<350>, :rawcontinue(), :prop<title> )»<title>.grep( * !~~ /^'Category:'/ );
}
for @tasks -> $title {
# If you want to resume partially into automatic # downloads, adjust $skip to skip that many tasks my $skip = 0; next if $++ < $skip; note "Skipping first $skip tasks..." if $skip; next unless $title ~~ /\S/; # filter blank lines say $skip + ++$, " $title";
my $page = $client.get("{ $url }/index.php?title={ uri-escape $title }&action=raw").content; say "Whoops, can't find page: $url/$title :check spelling." and next if $page.elems == 0; say "Getting code from: http://rosettacode.org/wiki/{ $title.subst(' ', '_', :g) }#$language";
my $entry = $page.comb(/'==' $header '==' .+? [<?before \n'=='<-[={]>*'{{header'> || $] /).Str // whoops;
if $entry ~~ /^^ 'See [[' (.+?) '/' $language / { $entry = $client.get("{ $url }/index.php?title={ uri-escape $/[0].Str ~ '/' ~ $language }&action=raw").content; }
my $name = $title.subst(/<-[-0..9A..Za..z]>/, '_', :g);
my $dir = mkdir "./rc/tasks/$name";
spurt( "./rc/tasks/$name/$name.txt", $entry );
my @blocks = $entry.comb: $tag;
unless @blocks { if %resource{"$name"}<skip> { whoops; say "Skipping $name: ", %resource{"$name"}<skip>, "\n"; } }
for @blocks.kv -> $k, $v { my $n = +@blocks == 1 ?? !! $k; spurt( "./rc/tasks/$name/$name$n$ext", $v ); say "Skipping $name$n: ", %resource{"$name$n"}<skip>, "\n" and next if %resource{"$name$n"}<skip>; say "\nTesting $name$n"; run-it($name, "$name$n"); } say '=' x 79;
}
sub mediawiki-query ($site, $type, *%query) {
my $url = "$site/api.php?" ~ uri-query-string( :action<query>, :format<json>, :formatversion<2>, |%query); my $continue = ;
gather loop { my $response = $client.get("$url&$continue"); my $data = from-json($response.content); take $_ for $data.<query>.{$type}.values; $continue = uri-query-string |($data.<query-continue>{*}».hash.hash or last); }
}
sub run-it ($dir, $code) {
my $current = $*CWD; chdir "./rc/tasks/$dir/"; if %resource{$code}<file> -> $fn { copy "./../../resources/{$fn}", "./{$fn}" } my @cmd = %resource{$code}<cmd> ?? |%resource{$code}<cmd> !! "$exe $code$ext"; for @cmd -> $cmd { say "\nCommand line: $cmd"; try EVAL(shell $cmd); } chdir $current; say "\nDone $code";
}
sub uri-query-string (*%fields) { %fields.map({ "{.key}={uri-escape .value}" }).join('&') }
sub clear { "\r" ~ ' ' x 100 ~ "\r" }
sub whoops { note "{'#' x 79}\n\nNo code found\nMay be bad markup\n\n{'#' x 79}"; }
sub load-resources { () } # load resources for finer control </lang>
- Output:
with command line
Retrieving tasks 1 Determine if a string is numeric Getting code from: http://rosettacode.org/wiki/Determine_if_a_string_is_numeric#Perl_6 Testing Determine_if_a_string_is_numeric Command line: perl6 Determine_if_a_string_is_numeric.p6 Coerce Don't coerce String whitespace whitespace <1> True True <1.2> True True <1.2.3> False False <-6> True True <1/2> True True <12e> False False <B17> False False <1.3e+12> True True <1.3e12> True True <-2.6e-3> True True <zero> False False <0x> False False <0xA10> True True <0b1001> True True <0o16> True True <0o18> False False <2+5i> True True <1 1 1> False False <> True False < > True False Done Determine_if_a_string_is_numeric ===============================================================================
Or, if the full %resource hash is loaded it will automatically feed input parameters to tasks that require them:
perl6 RC-run.p6 Lucky_and_even_lucky_numbers
Retrieving tasks 1 Lucky_and_even_lucky_numbers Getting code from: http://rosettacode.org/wiki/Lucky_and_even_lucky_numbers#Perl_6 Testing Lucky_and_even_lucky_numbers Command line: perl6 Lucky_and_even_lucky_numbers.p6 20 , lucky 79 Command line: perl6 Lucky_and_even_lucky_numbers.p6 1 20 (1 3 7 9 13 15 21 25 31 33 37 43 49 51 63 67 69 73 75 79) Command line: perl6 Lucky_and_even_lucky_numbers.p6 1 20 evenlucky (2 4 6 10 12 18 20 22 26 34 36 42 44 50 52 54 58 68 70 76) Command line: perl6 Lucky_and_even_lucky_numbers.p6 6000 -6100 (6009 6019 6031 6049 6055 6061 6079 6093) Command line: perl6 Lucky_and_even_lucky_numbers.p6 6000 -6100 evenlucky (6018 6020 6022 6026 6036 6038 6050 6058 6074 6090 6092) Done Lucky_and_even_lucky_numbers ===============================================================================
Run BASIC
<lang runbasic>bf$ = "" a$ = httpGet$("http://rosettacode.org/wiki/Category:Run_BASIC") ' get RB tasks from [RC] a1$ = word$(a$,2,"Pages in category ""Run BASIC")
a1$ = word$(a1$,1,"")
i = 2
b$ = word$(a1$,i,"
Tasks | |
Task | "
html "<select size=10 id='runProg' name='runProg'>" while b$ <> "" b$ = left$(b$,instr(b$,"""")-1) b$ = strRep$(b$,"%2B","+") b$ = strRep$(b$,"%27","'") html "<option>"+b$+"</option>" i = i + 1b$ = word$(a1$,i," |
"
' BUTTON options to Run It or Exit button #run, "Run It", [runProg] button #ex, "Exit", [quit]html " |
wait
[runProg] progName$ = #request get$("runProg") print progName$ a$ = httpGet$("http://rosettacode.org/wiki/"+progName$)
i = instr(a$,"<a href=""#Run_BASIC"">")
a$ = mid$(a$,i-6,6) a$ = word$(a$,2,"-") a$ = word$(a$,1,"""") cls ' clear screen 'print a$ ' this is the program number used in the [RC] editor
a$ = httpGet$("http://rosettacode.org/mw/index.php?title="+progName$+"&action=edit§ion="+a$)
a$ = word$(a$,2,"{header|Run BASIC}") i = instr(a$,">") a$ = mid$(a$,i+1) i = instr(a$,"/lang>") a$ = left$(a$,i-5) a$ = strRep$(a$,"<","<") ' this is the good program code ' place the code in the rb$ file rb$ = DefaultDir$ + "\projects\a_project\rcCode.bas" ' RC program open rb$ for output as #f print #f,a$ close #f
print "================== Run Basic Solution ===========================" run rb$,#handle ' point RunBasic to the file with the program render #handle ' render the runned code [quit] ' that's it folks end
' -------------------------------- ' string replace rep str with ' -------------------------------- FUNCTION strRep$(str$,rep$,with$) ln = len(rep$) ln1 = ln - 1 i = 1 while i <= len(str$)
if mid$(str$,i,ln) = rep$ then strRep$ = strRep$ + with$ i = i + ln1 else strRep$ = strRep$ + mid$(str$,i,1) end if
i = i + 1 WEND END FUNCTION</lang>
Tcl
This code only includes support for running Tcl task solutions, but it can download any language's; it assumes that the first <lang…> is sufficient when it comes to task extraction (definitely not true universally, but mostly good enough).
<lang tcl># Code to download task contents from find-bare-lang-tags task package require Tcl 8.5 package require http package require uri
proc getUrlWithRedirect {base args} {
set url $base?[http::formatQuery {*}$args] while 1 {
set t [http::geturl $url] if {[http::status $t] ne "ok"} { error "Oops: url=$url\nstatus=$s\nhttp code=[http::code $token]" } if {[string match 2?? [http::ncode $t]]} { return $t } # OK, but not 200? Must be a redirect... set url [uri::resolve $url [dict get [http::meta $t] Location]] http::cleanup $t
}
} proc getTaskContent {task} {
set token [getUrlWithRedirect http://rosettacode.org/mw/index.php \
title $task action raw]
set content [http::data $token] http::cleanup $token return $content
}
- Code to extract the first <lang> section for a language
proc getTaskCodeForLanguage {task language} {
set content [getTaskContent $task] set startRE {==\s*\{\{header\|@LANG@(?:\|[^{}]+)?\}\}\s*==} set startRE [string map [list @LANG@ $language] $startRE] if {![regexp -indices $startRE $content start]} {
error "$language does not implement task \"$task\""
} if {![regexp -indices -start [lindex $start end] \
"==\\s*\\\{\\\{header" $content end]} { set end {end end}
} set content [string range $content [lindex $start 1] [lindex $end 0]] # Extended format RE used to allow embedding within _this_ task's <lang>! if {![regexp {(?x)<lang .*?>(.*?)</ lang>} $content -> solution]} {
error "$language solution of task \"$task\" has no useful code"
} return "$solution\n"
}
- How to download and run a Tcl task
proc runTclTaskForLanguage {task} {
puts "Fetching task solution..." set solution [getTaskCodeForLanguage $task Tcl] set filename rcsoln_[string map {/ _ " " _} $task].tcl set f [open $filename w] puts $f $solution close $f puts "Executing task solution with: tclsh $filename" exec [info nameofexecutable] $filename <@stdin >@stdout 2>@stderr
} runTclTaskForLanguage {*}$argv</lang>
UNIX Shell
See C1R Implementation for an incomplete implementation. (only supports C)