Talk:Rosetta Code/Rank languages by popularity: Difference between revisions

==More task clarification requested==

My example (zkl) has been marked as incorrect. It uses the members field as the count. As does UnixPipes example (if I read it correctly). I checked my answer against the UnixPipes one and got the exact same results. I also note other examples yield the same results. What am I missing?

 

: Other pitfalls are that some programming languages are in different (letter) cases (Maple, MAPLE;   Gdl, GDL;   NewLISP, NewLisp;   etc.);   several languages use unicode character(s), others use different names, and still others aren't "registered" in Rosetta Code properly (such that they aren't recognized as a programming language).   Another big stumbling block is that most examples don't properly handle the ranking of tied languages. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 09:49, 12 March 2014 (UTC)

::It really does seem like this task needs additional verbiage such that it can be solved/implemented against the description rather than other implementation and "hidden" knowledge.

 

==task clarification==

 

: Note:   this is done by REXX. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:03, 30 September 2013 (UTC)

 

==incorrect sample==

Regarding the sample output (as part of the task description) appears to be incorrect (as shown on March 24^th, 2014.

<br>It doesn't show the &nbsp; PARI/GP &nbsp; language entry (it's currently ranked 28th at 394 entries) and it was ranked approximately the same at that time by the REXX entry that was also executed at around the same time frame. &nbsp; Could it be that the solidus is goofing things up in the program? -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 16:14, 24 March 2014 (UTC)

 

==Incorrect examples==

ALL examples, except Python "Working Solution" are not working. Please fix. --[[User:Guga360|Guga360]] 03:49, 29 July 2009 (UTC)

:How, exactly, are all other solutions "not working"? Please elaborate. --[[User:Glennj|glennj]] 10:12, 2 August 2009 (UTC)

 

I noticed about a handful of programming examples that "only" grabbed 500 languages. &nbsp; Now, when they were first entered (for the most part), there weren't 500 languages, but now there are over 500 languages. &nbsp; At least two examples were added after there were 500 languages. &nbsp; Are these (or should these) examples be considered as incorrect? -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 21:16, 15 January 2014 (UTC)

 

==ALGOL 68==

::: Actually, this does not solve the problem. We're still reading Special:Categories, if there is a language starting with Z, probably it will not get in the list. --[[User:Guga360|Guga360]] 01:30, 16 June 2009 (UTC)

 

:::::::* Z80 Assembly

:::::::* Zonnon

:::::::* ZED

 

== Ruby example question ==

(ary.length/n + 1).times {|i| yield ary[i*n,n]}

end</lang>

 

== Redundant task? ==

: And the downside is that the current XPath implementation relies on [http://sax.sourceforge.net/ SAX], and maintenance has gone so bad there that you can basically only use this J implementation on 32 bit J6.02. The underlying problems of standards decay and implementation decay has a lot of causes, but the bottom line might be that any technical solution relies partially on popularity for support and thus will fail sooner or later. I guess that means I (or, ok, someone else - but if you want something done right you need to take responsibility for accomplishing it) should come up with another J implementation. --[[User:Rdm|Rdm]] ([[User talk:Rdm|talk]]) 00:11, 10 April 2016 (UTC)

 

 

 

 

<br>

<br>

 

 

 

 

Currently, the REXX example reports on all &nbsp; <strike>'''471'''</strike> &nbsp; (see below) &nbsp; programming languages, but there is code to support the skipping of languages that have less than a certain (specified) number of examples.

 

 

 

 

<strike>

</strike>

 

 

 

 

 

 

:::* &nbsp; <big> ''' µC++ ''' </big>

:::* &nbsp; <big> ''' ΜC++ ''' </big> &nbsp; &nbsp; (unicode version)

</strike>

 

 

<br>

 

REXX did find the '''MK-61/52 ObjectScript''' language, but it's translated to (spelled as) '''MK-61/52 ObjectScript''', and it's ranked 191^st (tied) with 31 entries. <br>

 

REXX did find the '''Forth''' language, but it's spelled '''FORTH''', and it's ranked

I took a look at the raw file (which the REXX program uses), the category file. &nbsp; There is no entry for '''AngelScript''' in that file, but there is an entry for that language in the (raw) language file. &nbsp; The REXX program uses the category file to find entries, and when it finds an entry, it verifies that the language exists (that is, it is indeed a language instead of just a special category), it only then recognizes it as a language. &nbsp; Languages which have no entries in the category file aren't counted. &nbsp; However, the category file does have an entry for '''AngelScript User''' (1 member), so noting that '''AngelScript User''' is not a language, it is just a special category.

 

 

::I took this Rosetta Code task's requirement quite literally:

And, yes, I would like to see (or receive) Tcl's complete (output) list with counts.

 

 

 

<br>

--[[User:Aspectcl|Aspectcl]] ([[User talk:Aspectcl|talk]]) 11:03, 23 July 2015 (UTC)

 

 

<br>

<br>

 

 

: The way identically ranked languages (identical in the sense that they have the same number of entries) is sorted in the order in which they appear first in the Rosetta Code list. &nbsp; Thus, some languages aren't ranked fairly because of a (weak) sorting artifact of having the same number (of entries). &nbsp; Strictly speaking, if the following were true:

with all 2^nd place names marked as ''tied'' for 2^nd, and nothing marked as 3^rd, 4^th or 5^th.

<br>These duplicates (tied for placement) would make a good addition to this task (to rank languages ''correctly'') -- or lacking that, a good Rosetta Code task that can stand by itself.

 

:According to [http://rosettacode.org/wiki/Category:A%2B its task page] A+ has no tasks implemented. It seems as if a language with no tasks implemented is treated as if it has three

<br> (from http://www.rosettacode.org/mw/index.php?title=Special:Categories&limit=5000)

<br><br>(The bold font was added by me.) &nbsp; Note that it didn't say &nbsp; '''implementations''', &nbsp; but &nbsp; '''members'''.

 

==case of names of programming languages==

<br><br>Also added the unicode translation of &nbsp; '''╨£╨Ü-61/52''' &nbsp; (Cyrillic &nbsp; '''МК-61/52)''' &nbsp; into &nbsp; '''MK-61/52'''. -- [[User:Gerard Schildberger|Gerard Schildberger]] 20:24, 15 February 2013 (UTC)

<br><br>REXX code for other unicode versions of programming languages have been added since then. -- [[User:Gerard Schildberger|Gerard Schildberger]] 19:45, 30 March 2013 (UTC)

 

== Task suggestion for update ==

 

:::Hi [[User:Paddy3118|Paddy3118]] - regarding Python I tried the current code but Rosetta Code returns "Access Denied" based on my "browser signature". It must see it as a bot. It's not my IP as Awk works using the same API call. Also had a "Segmentation Fault" when running the C code. Just posting in case you want to try if different results. -- [[User:3havj7t3nps8z8wij3g9|3havj7t3nps8z8wij3g9]] ([[User talk:3havj7t3nps8z8wij3g9|talk]]) 21:43, 27 May 2015 (UTC)

 

==limit=5000==

The jq example uses limit=5000:

<lang sh>'http://rosettacode.org/mw/index.php?title=Special:Categories&limit=5000'</lang>

[[User:craigd|CraigD]] Wed, 29 Jul 2015 06:51:52 GMT: I noticed a problem while looking at the Nim example; there is a request limit of 500, while there are over 500 languages. Should this be bumped?

:{"warnings":{"categorymembers":{"*":"cmlimit may not be over 500 (set to 5000) for users"}

 

==why task change==

 

When I try to run the Racket code locally I get:

 

<pre>

 

FWIW, the Tcl solution uses <tt>regsub {^Category:} $name {}</tt> to get rid of the prefix, which seems safer than dropping the first 9 chars. I suggest Racket either do the same, or skip entries it cannot parse.