September 2017 1HaskellADay problems and solutions

• September 29th, 2017: Friday's #haskell problem uses name-parsing to post parsed names into PostgreSQL database and join them to articles. Today's #haskell solution extracts rows of raw names associated with articles, parses then stores them in PostgreSQL.
  
• September 28th, 2017: Thursday's #haskell problem inserts reified articles into PostgreSQL database, including names, and throws confetti! Today's #haskell solution inserts the articles from our compressed archive. YAY! Throw confetti! 
  
• September 27th, 2017: Wednesday's #haskell problem looks at parsing names in various formats. Today's #haskell solution solves the name-parsing problem by parsing each name-form in turn. 
• September 26th, 2017: Tuesday's #haskell problem looks at staging data on PostgreSQL for deferred processing using Haskell for ETL. Today's #haskell solution transforms RawNames to JSON and uploads it to a staging table on the PostgreSQL database.
   
• September 25th, 2017: Friday we took a compressed archive and broke that into individual articles. Today we divide articles into sections. Today's #haskell solution uncovers structure bit by bit as we reify an article structure from a block of text. 
• September 22nd, 2017: Today's #haskell problem we're going to scan into an 'unstructured' set of documents and start to look for structure. Welp, today's #haskell solution is a 'start': from one blob to 11 blocks of raw text. We'll dig in deeper next week. 
• September 21st, 2017: TOMORROW's #haskell problem (announced a wee bit early) queries PostgreSQL database to fetch keyword dictionary state. Today's #haskell solution picks up where we left off yesterday then computes the top keywords across articles. 
  
• September 20th, 2017: Today's #haskell problem asks to push keywords extracted from article onto a PostgreSQL data table. Today's #haskell solution pushes keywords – PUSHES KEYWORDS REAL GOOD – to a PostgreSQL database. 
  
  
  
  
• September 19th, 2017: So we've encoded keywords into Ints, today's #haskell problem decodes those ints back to the keywords. Today's #haskell solution 'reverses the arrows' of the Map, decoding a keyword from an index 
• September 18th, 2017: Today's #haskell problem computes the relative word 'strength' of each word in a document. Today's #haskell solution breaks the word-strength problem into small pieces then foldM's over the small pieces. 
• September 15th, 2017: Today's #haskell problem takes a look at KEA/Keyword Extraction Algorithm. Ooh, doggies! State IO Map KeyWord something-or-otherfor today's #haskell solution.
• September 14th, 2017: Today's #haskell problem involves using the simple interface to PostgreSQL to insert rows of data. Ooh! Exciting! Today's #haskell solution uses ToRow and ToField to insert Haskell values into #PostgreSQL #DaaS instance! lolneat! 
  
• September 13th, 2017: Today's #haskell problem asks the article names have encoded Julian dates ... OR DO THEY? So, in solving today's #haskell problem, I realized THOSE AREN'T JULIAN DATES! THOSE ARE DATE DATES! 
• September 8th, 2017: Continuing our ETL exploration with a load/compress/storefor today's #haskell exercise. Today's #haskell solution shows load/compress/store ... that's a word now. 
• September 5th, 2017: Today's #haskell exercise is to build a directory web serviceusing the Snap framework [or a framework you prefer] 
• September 4th, 2017: Today's #haskell problem reads in source documents and does some word frequency analyses. Today's #haskell solution shows, surprisingly, 'the' is the most popular English word. Or, not so surprisingly. 
• September 1st, 2017: Today's #haskell problem #SPARQL-queries wikidata.org to fetch countries' populations for FB analysis. Yesterday, India won FB by population, today we have a surprise winner in Thailand for FB by percentage user base. 

August 2017 1HaskellADay 1Liners Problems and Solutions

• August 1st, 2017:
f :: (Maybe a, b) -> Maybe (a, b) Define points-free.
• August 1st, 2017:
  Given f above and f a and f b are mutually exclusive in Maybe monad, define
  g :: Maybe (a, b) -> Maybe (a, b) -> (Maybe a, b)
  points free
• August 1st, 2017:
  Now, let's define the dual of f
  f' :: Maybe (a, b) -> (Maybe a, b)
  points free

August 2017 1HaskellADay problems and solutions

• August 31st, 2017: Today's #haskell exercise is by way of @ahnqir. What are the percentage of facebook users by country? Today's #haskell solution charts FB users by country. GO INDIA! 
  
• August 4th, 2017: For today's #haskell problem we are looking at the caesar cipher as a better rot13. And today's #haskell solution with rot, rot, roslein rot everywhere.  
• August 1st, 2017: For today's #haskell problem we are parsing a string into word-tokens and munging/demunging them with rot13! WOOT! Munging plaintext and demunging cyphertext, oh, my! 

July 2017 1HaskellADay 1Liner

• July 7th, 2017:
  In LU-decomposition of matrices you have square P-matrix:
  [[1,0..],
   [0,2,0..],
   [0,0,3,0..],
  ...]
  For matrices of n² size
  Code that
  
  • ∃! David Turner @DaveCTurner
  • matrix n = let td = take n . drop 1 in td [td $ replicate i 0 ++ [i] ++ repeat 0 | i <- [0..]]

July 2017 1HaskellADay Problems and Solutions

• July 26th, 2017: We look at unifying fresh variables to ground terms for today's #haskell problem. And we have unification working for free variables and ground terms for today's #haskell solution. 
• July 25th, 2017: We look at the beginnings of unification with ground terms for today's #haskell problem. Today's #haskell solution shows unification of ground terms in a monadic domain. 
• July 24th, 2017: Commuters today occupy my mind for today's #haskell problem. A little reasoning, a little harsh reality, and we have today's #haskell solution. 
• July 20th, 2017: I didn't go for my morning jog, so today's #haskell problem looks at running ... expressions. Eheh. For today's #haskell solution, some people run on empty, but I run on LOGIC! ... same thing. 
• July 19th, 2017: Today's #haskell problem is our first step in our scheme to reasoning! MWA-HAHA! via the Reasoned Schemer. Today's #haskell solution:some days are #s; some days are #u 
  
  
  
  
• July 7th, 2017: It's FRIDAY, and you know what that means? SCRABBLE DAY! for today's #haskell problem 
• July 6th, 2017: For today's #haskell problem we invert a matrix to solve a system of equations. 
• July 5th, 2017: For today's #haskell problem we have a * b = c ... HOW HARD CAN THAT BE? #famouslastwords The #haskell solution took like a whole second! THAT WAS HARD! 
• July 4th, 2017: Boys and girls! Gather 'round as I spin you a story of Jemima and Roland for today's #haskell problem. 
• July 3rd, 2017: Today's #haskell problem is all 'bout the Ord, 'bout the Ord, 'bout the Ord. No Functor. Today's #haskell solution uses a little sorting logic (not sordid logic; that's different) to show Jim is the eldest.
  

June 2017 1HaskellADay 1Liners

• June 17th, 2017:
  f :: (a, [a]) -> [a] -> [a]
  f (c, w1) w2 = c:w1 ++ w2
  
  Define f points-free
• bazzargh @bazzargh (++).uncurry(:)
• Felt there must be a nicer way to exploit symmetry of mappend.uncurry(mappend.pure) but can't find it

June 2017 1HaskellADay Problems and Solutions

• June 21st, 2017: Today's #haskell problem connects to and queries a SQL database of your choosing. 
• June 16th, 2017: Today's #haskell problem has us interacting with the OS environment in anticipation of SQL database queries. Today's #haskell solution uses Applicative Functors to string together values we collect from the OS environment.
• June 15th, 2017: Today's Haskell problem comes by way of @aisamanra, Mr. Lists-are-Mu-functions-in-the-function-of-Mus. Today's #haskell solution shows Lists are Mu-functors! ... now, making them instances of type-classes, however ... 
  
• June 14th, 2017: Today's #haskell problem is brought by way of @matthieubulte: approximating solutions with Runge-Kutta. The #haskell solution shows I can (ψ, ε, k) with the best of them! AHA! ... but is it the CORRECT solution? 🙄 
  
• June 6th, 2017: In today's #haskell problem, we go clockwise around a binary tree, because ... we've had one too many?