Talk:Rosetta Code/Count examples/Full list/Tier 4

From Rosetta Code

Entry 123 claims that F# does not implement Matrix Multiplication. has Matrix Multiplication under Pages in category "F Sharp".--Nigel Galloway (talk) 15:10, 31 January 2022 (UTC)

The report generating logic doesn't use any heuristics to try to determine if a particular piece of code is written in any particular language. It only looks at the presence (or absence) of the Rosettacode supported language header to tell if that language has an example or not. The supported header for F Sharp is =={{header|F_Sharp|F#}}==; the Matrix Multiplication F Sharp entry header is =={{header|F#}}==, so it is not recognized as belonging to F_Sharp (the languages name on Rosettacode).
Is it ideal? No, it's not. But the algorithm is pretty darn accurate as long as the task header markup is correct. Am I particularly inclined to modify the parsing logic to find unsupported markup? No, not really. --Thundergnat (talk) 15:55, 31 January 2022 (UTC)
So it is rosettacode that is supporting unsupported markup when it generates Where is supported markup defined? --Nigel Galloway (talk) 16:08, 31 January 2022 (UTC)
Ok. "supported" may be overstating it a bit. "currently recognized" is probably more accurate. If "support" is lacking, it is by my report generator. The problem is not with Rosettacode (at least, not specifically) but with the mediawiki software it runs (possibly ran,) on. Apparently, language names that have a octothorpe in them caused issues with internal links. Someone, (I don't know who,) at some point (I don't know when) made the decision to use the written out word 'sharp' to stand in for '#', and made a note of it on the language category pages for C# and F#; the very first lines at the top of the pages. My report generator just uses the language name as used in the Category:{whatever} category page as the language name. If the task example markup doesn't use the same "language name", it won't be recognized as belonging to it.
I am not defending that decision or advocating for or against it, just stating what the situation is as I understand it. And who knows? I could be completely full of crap. Wouldn't be the first time. --Thundergnat (talk) 17:06, 31 January 2022 (UTC)
(And by the way, that 123 is the number of languages that it recognizes as having an entry for that task, not not an entry number.) --Thundergnat (talk) 16:02, 31 January 2022 (UTC)

For what it is worth, here are some other F Sharp tasks that aren't being counted due to faulty header markup (in no particular order): --Thundergnat (talk) 22:55, 31 January 2022 (UTC)

Irregular markup

see White space inside XML/HTML tags </lang > is not irregular.--Nigel Galloway (talk) 15:10, 2 February 2022 (UTC)

White space inside XML/HTML tags is not incorrect but, as there were only seven instances I could find on the whole of Rosettacode, it was indeed irregular. --Thundergnat (talk) 15:40, 2 February 2022 (UTC)

Meaning of O

Normally when a task is implemented the symbol ㉞ is used. For entry 128 "Include a file" the symbol 'O' is used . What does this mean?--Nigel Galloway (talk) 14:10, 21 February 2022 (UTC)

There is no entry 128. There is a (several) task(s) for which there are 128 entries. The symbol ㉞ is used for every task where the language ranked 34th by the number of tasks with an entry, has a task entry. That ranking can, and often does change. In this case, for this week, ㉞ corresponds to F_Sharp. The meaning of the symbol O? Let me quote the main Full list page:

Languages that omit a certain task have an uppercase O filled in for that task.

Somebody, somewhere, at some time, decided that the task was not {pick one: easy, stylistically allowed, possible, whatever } to be implemented in F_Sharp and marked the task page as such. --Thundergnat (talk) 15:47, 21 February 2022 (UTC)