What Programming Languages do you use?

[whartung]

Python, not so much.

Well, in theory you could compile the entire Python runtime and download that to run scripts.

There's been recent chatter of someone who has compiled and is embedding the entire SQLite engine in WebAssembly so that his web page can read (read only) static SQL databases. It's pretty clever.

It's a 400K artifact, but it's still clever.

 

[ShawnDriscoll]

You can run Python in pretty much anything these days.

 

[aramis]

Agreed. But I just wonder if there's anything out there besides LLVM. I avoid JavaScript when I can. C++ and Go don't appeal to me when I'm just noodling around. It's silly to hope for support of non-statically-typed languages, e.g. Python.

auses: :consults the googles:

There's even a COBOL compiler for wasm. Pascal, too.

Python, not so much.

there is a method for conversion of Py to C/C++...

https://github.com/pradyun/Py2C

 

[ShawnDriscoll]

there is a method for conversion of Py to C/C++...

Most scientific libraries for Python are written in Cython. You can run C++ code in your Python. And run Python code in your C++. This is for the MSVC++ 14.0 build of Python mostly. The end result is to create Python versions of DLLs (PYDs) that can be called by Python. I've got Python code to run 100s times faster when cross-compiled into C/C++. And C/C++ code can make calls to or embed Python code (and run it faster) when variables are non-dynamic.

There is a Linux version as well. And there is a gcc for it that is compatible with MSVC++ 14.0 (Visual Studio 2015).

Since Python 3.7, data can be strongly-typed if wanted.

 

[willtron3030]

there is a method for conversion of Py to C/C++...

https://github.com/pradyun/Py2C

It's pre-alpha, and focuses on statically typed data. Still, it's nice to see I'm not the only one.

There are projects to run Python in the browser. They work, -ish, but it's a big payload.

 

[ShawnDriscoll]

There are projects to run Python in the browser. They work, -ish, but it's a big payload.

Brython isn't big at all. Unless you are still using NN 4 or IE 7.

 

[willtron3030]

:Checks file size:

lays tetris whilst watchig process monitor:

pens TOC of for browser interface module docs:

Uhm... I'll be at my desk, thank you.

 

[ShawnDriscoll]

If you have Chrome installed, https://brython.info/ shows Python running inside of HTML. View the webpage source to see how easy it is.

 

[robject]

I got talking last night with a few friends and we were reminiscing on our days with C64 BASIC. [...] ++laugh++ I still used a C64 way up into early 2000s up until the point instructors stopped accepting papers typed and printed from my c64.

LOL

Just last year, I brushed off my Commodore BASIC skills, and wrote a Traveller program. The emulator runs in the browser:

https://www.commanderx16.com/emulator/x16emu.html?manifest=/emulator/100-traveller-trader-mini-/

It contains over 500 worlds in and around the Spinward Marches... and it's primitive.

 

[Leitz]

On the off chance that anyone else uses Perl, I've been working on some code to collate text files into a text based "book".

Book::Collate is also on CPAN, and can be installed as a regular module. In the GitHub repo are some example programs; one to use Book::Collate and one to scrub out various exteranous lines if you copy a forum web page into a pure text document.

 

[willtron3030]

If I weren't already so invested in Markdown and RST toolchains, this would really scratch an itch for me.

Side question: Is this a common benchmark for estimating grade level?

sub grade_level {
  my $self = shift;
  my $sentence_average = $self->word_count / $self->sentence_count;
  my $word_average      = $self->syllable_count / $self->word_count;
  my $grade = 0.39 * $sentence_average ;
  $grade    += 11.8 * ( $word_average );
  $grade    -= 15.59;
  return sprintf("%.2f", $grade); 
}

 

[Leitz]

If I weren't already so invested in Markdown and RST toolchains, this would really scratch an itch for me.

Side question: Is this a common benchmark for estimating grade level?

sub grade_level {
  my $self = shift;
  my $sentence_average = $self->word_count / $self->sentence_count;
  my $word_average      = $self->syllable_count / $self->word_count;
  my $grade = 0.39 * $sentence_average ;
  $grade    += 11.8 * ( $word_average );
  $grade    -= 15.59;
  return sprintf("%.2f", $grade); 
}

Straight from Flesch-Kincaid, scroll down to "Flesch–Kincaid grade level"

 

[robject]

Hey, I remember using this algorithm when rating marketing material on our corporate internal website!

 

[aramis]

Straight from Flesch-Kincaid, scroll down to "Flesch–Kincaid grade level"

The average rubric used also includes the specific vocabulary and whether or not the definitions are included...

 

[Leitz]

The average rubric used also includes the specific vocabulary and whether or not the definitions are included...

I'm using the Perl code for my fiction, and there's a glossary and "Library Data" section being built.

I write so the books are easy to read for non-native English speakers. Identifying the higher grade level sections tells me where to focus when revising.

 

[aramis]

I'm using the Perl code for my fiction, and there's a glossary and "Library Data" section being built.

I write so the books are easy to read for non-native English speakers. Identifying the higher grade level sections tells me where to focus when revising.

You should probably look for the standard wordlists by grade, too.
Most adults with adolescent acquisition of ESL get to an 8th to 12th grade reading level, according to research presented at an in-service day lecture. Which is why the parent materials are supposed to be written at 8th grade level.

The Dolch lists are the top 200-300 (depending upon specific revision) most frequent sight-words in American English.
The Fry lists are grade level sorted common word lists, without the sight word requirement.
Integrating those into your algorithm would also be useful.

Also note: pre-adolescent acquisition of ESL gets pretty close to age peers, but many adults in the US have a reading level below 12th grade...

 

[Leitz]

The Dolch lists are the top 200-300 (depending upon specific revision) most frequent sight-words in American English.
The Fry lists are grade level sorted common word lists, without the sight word requirement.
Integrating those into your algorithm would also be useful.

Excellent! I have both lists now, and am mentally pottering about how to integrate them. Thanks!

 

[willtron3030]

... but many adults in the US have a reading level below 12th grade...

The energy sector company I worked for in the 90s had a standard for written business communications, where we were to write at an eight grade level.

We were also told to assume a maximum memory span or span of control of five. Any procedural task list had to be one page and at most five items long, breaking into a hierarchy of sub-lists of necessary.

 

[Leitz]

The Fry lists are grade level sorted common word lists, without the sight word requirement.

It's been a rough few weeks but I finally got back to this. The code currently checks each unique word usage against the Fry list. I have the Fry list words sorted by grade level but the first push was just to integrate the list into the book reporting.

Fry Word Usage: 
  Used   800
  Custom 0
  Miss   3629

The plan is to allow a user to set up a custom word list so that the "Miss" rate can go down and the Fry ratio improve. I'm also using my idea of "custom word list" as "stuff I should write a library data entry on".

 

[robject]

I see this thread is coming close to five years old now.

The trend in programming language usage here might be interesting.