June 2018 1HaskellADay Problems and Solutions

• June 29th, 2018: Friday's #haskell problem is partially parsing a very large file. For Friday's #haskell solution we parse a JSON map by hand. 
• June 27th, 2018: In Wednesday's #haskell problem, we parse JSON ... TWICE! to get to the underlying graph structure. Wednesday's #haskell solution is a little bit (a lot, actually) of JSON parsing, and then some graph analysis.
• June 26th, 2018: Tuesday #haskell problem converts a .properties file to JSON to make a REST call. Converting a .properties file to JSON in 23,952 easy steps for Tuesday's #haskell solution. 
• June 25th, 2018: Monday's #haskell problem is parsing a JSON error report and updating the JSON with new information. In Monday's #haskell solution, we get a little 'lens-y' to update JSON data. 
• June 22nd, 2018: Friday's #haskell problem sends articles-as-JSON to an AWS Lambda function for processing. 
• June 20th, 2018: Wednesday's #haskell problem is SymbolTable from a set of articles, then document vectors. 
• June 19th, 2018: We find out that 'materially' is the word for Tuesday's #haskell problem. For Tuesday's #haskell solution we find that parsing JSON is simpler than parsing TSV. This comes as a surprise to no one but me. 
• June 18th, 2018: Monday's #haskell problem asks: "When are unique IDs not unique IDs?" More dater-analyses. Monday's #haskell solution: we go from 2300+ duplicates to 30+ duplicates. A good day's work. 
• June 15th, 2018: Friday's #haskell problem is sorting and removing duplicatesfrom a data table by defining Ord. Friday's #haskell solution removes duplicate articles by ID... but are the articles duplicates? Hmmmm! 
• June 8th, 2018: "Dater-analytics" is the name of the game for Friday's #haskell problem. You get some real-world dater that's yucky, and you get to unyuckify it. You're welcome. 
• June 6th, 2018: Now we take that restructured JSON (see yesterday's exercise) and store that JSON in a PostgreSQL database for Wednesday's #haskell exercise. Wednesday's #haskell solution populates entities with wikipedia reference into a PostgreSQL database. 
  
• June 5th, 2018: Tuesday's #haskell problem is to transform 'pancake' JSON into ... '3-layer cake'-JSON? Sure. #nailedit Tuesday's #haskell solution we restructured JSON and saved result to ... JSON! YAY! 
• June 4th, 2018: Monday's #haskell problem: interlanguage communication via the function main. 3 + 4 = 7 ... now do that and have Python talking to Haskell. 

May 2018 1HaskellADay Problems and Solutions

• May 31st, 2018: In Thursday's #haskell problem we are analyzing JSON so we can produce ... MORE JSON! And thus grows our Big Data. Thursday's #haskell solution shows us the results of Big Data analytics? More Big Data. YES! 
• May 30th, 2018: Wednesday's #haskell problem we process 'big-ish' data. Wednesday's #haskell solution: You take your big(-ish) data, and you chart it. THIS WE CALL DATA SCIENCE! 
  
• May 25th, 2018: Friday's #haskell problem is load testing a web application with a database pull. Friday's #haskell solution: a load-tester in Haskell! AHA! 😎
• May 24th, 2018: Thursday's #haskell exercise is adding articles to be cleaned up, post-ETL, to a dirty table in PostgreSQL. Thursday's #haskell solution stages articles loaded into the database to be cleaned up later. 
  
• May 21st, 2018: Monday's #haskell problem is to insert NEW (triaged) articles into a PostgreSQL database. 
• May 18th, 2018: Friday's #haskell exercise: there seems to be a lot of tagged terms for articles. Download these terms-as-JSON from a REST endpoint and count them is today's exercise. Friday's #haskell solution has got me singing "Havana-na-na-na!" ... no, wait: I meant: "uploading tags from a REST endpoint to a PostgreSQL database." That's what I meant. 
• May 17th, 2018: Thursday's #haskell exercise is to add database update functionality to existing code and in a Writer/IO monad; yikes! Thursday's #haskell solution: from the triaged articles, the update SQL statements naturally fall out. 
• May 11th, 2018: Friday's #haskell exercise: bridging Python and Haskell to deliver a polyglot system. 
• May 8th, 2018: Tuesday's #haskell problem uses articles fetched from a REST endpoint and article metadata to triage articles for daily upload to a PostgreSQL database. Tuesday's #haskell solution... triage: get! 
• May 7th, 2018: Monday's #haskell problem is extracting metadata for articles stored in a PostgreSQL database. Monday's #haskell solution repurposes a library to fetch article metadata from a new article database. 
  
• May 4th, 2018: For Friday's #haskell problem we look at a daily upload process from a REST endpoint to a SQL data store and start to implement it. Friday's #haskell solution gets a week's worth of data from the REST endpoint. 

April 2018 1HaskellADay Problems and Solutions

• April 25th, 2018: For Wednesday's #haskell problem we prove, definitively, @fermatslibrary n^5 is n ... for very small n. For Wednesday's #haskell solution we learn that number theory is so cool! 
  
• April 24th, 2018: For Tuesday's #haskell problem we look for a fix-point of simplifying HTML. Good thing it's not NP-hard! ... no ... wait. Tuesday's #haskell solution is the fix-point for HTML. Just like the fix-point for factorial. Structurally, maybe, but otherwise, kinda not. 
• April 23rd, 2018: What happens when things go wrong? We start to look at debugging for Monday's #haskell problem. For Monday's #haskell solution: the error is not here; the error is in another castle. 
  
• April 20th, 2018: Friday's #haskell problem is to decode HTML entities from titles of articles. Friday's #haskell solution: deHTMLification: get 
• April 18th, 2018: Hey! Let's store the raw JSON as a fall-back should our ETL process fail for Wednesday's #haskell problem. Wednesday's #Haskell solution: storing raw JSON in a SQL data store BECAUSE WE CAN! 
  
• April 17th, 2018: For Tuesday's #haskell problem we upload articles from a different publisher of a compressed JSON archive with a different format. We find adding a differently-structured set of articles to IxArt store is simple for Tuesday's #haskell solution. 
• April 16th, 2018: For Monday's #haskell problem, we take articles-as-json from various sources and put them into a common SQL data store. For Monday's #haskell solution we upload articles from different publications to a common SQL data store, both compressed and uncompressed archives. 
  
• April 13th, 2018: Friday's #haskell problem is to grab packets of articles from a REST endpoint for later processing. Friday's #haskell solution: Those are some rather large packets downloaded from the REST endpoint! 
• April 12th, 2018: Thursday's #haskell problem: auditing and logs. Thursday's #haskell solution is to wrap packet extraction and insertion in an ETL process. 
  
• April 11th, 2018: Wednesday's #haskell problem: when we download packets of articles, let's record that event. Wednesday's #haskell solution: we're inserting packets into our new database. 
  
• April 10th, 2018: Tuesday's #haskell problem: 🎵 "Does anybody know what time it is?" 🎶 (bonus: name that tune) Tuesday's #haskell solution: "Hey, Mister! You got the time?" 
• April 9th, 2018: Monday's #haskell problem is to extract article tags from JSON then store them in a PostgreSQL lookup table. Monday's #haskell solution: Inserting tags for articles into a PostgreSQL database is a breeze! 
  
• April 6th, 2018: Friday's #haskell problem is to parse categories-as-JSON, strip it down to its essentials, and store in a PostgreSQL data table. Friday's #haskell solution uploaded categories to a PostgreSQL databse after 'removing the stupid.' That's a technical term. 
  
  
• April 5th, 2018: Thursday's #haskell problem is revising the database connector now that we're managing multiple databasen. Thursday #haskell solution is making connections to SQL databases configurable. Whoa. Edgy, tweeps! I'm really pressing the bleeding edge of technological advancement here with configurations and database connections. 
• April 4th, 2018: Wednesday's #haskell problem is storing authors in a PostgreSQL table. Wednesday's #haskell solution shows that transferring authors from articles to a SQL database is as easy as 1, 2, 3! 
  
  
• April 3rd, 2018: Tuesday's #haskell exercise: AUTHOR! AUTHOR! Is there an author in the house? Tuesday's #haskell solution: extracting author information from very unstructured data. 
• April 2nd, 2018: Monday's #haskell problem is to explore JSON to find the structure in the data. Monday's #haskell solution: oooh! Pritteh JSON! But what's this weirdness with the author-identifier? 
  

February 2018 1 Liner 1HaskellADay Problems and Solutions

• February 8th, 2018: We have maybe :: b -> (a -> b) -> Maybe a -> b. But we don't have list? Or do we? Define list:
  list :: b -> ([a] -> b) -> [a] -> b
  list nullanswer flist lst = undefined
• BONUS: Both Maybe a and [a] are binary types, ... so is MonadPlus:
  
  Maybe a = Nothing | Just a
  List a = Cons a (List a) | Nil
  MonadPlus m => mzero | m a `mplus` m a
  
  Is there some generalization that maybe and list are functions of? What is that generalization?
• February 6th, 2018:
  You have f :: a -> [b] -> [c]
  But instead of just one a you have [a]
  Define g :: [a] -> [b] -> [c]
  in terms of f
• Daniel @leptonyu g as bs = foldl (\xs a -> f a bs ++ xs) [] as
• ptdr_bot @m0rth0n g as bs = flip f bs =<< as
• Victoria C @ToriconPrime g as bs = concat $ fmap ($ bs) (fmap f as)
• matt @themattchan g = flip $ concatMap . flip f
• Nicoλas @BeRewt g = flip (flip (>>=) . flip f) Or: g as bs = as >>= flip f bs
• Sangeet Kar @sangeet_kar g = foldMap f

March 2018 1HaskellADay problems and solutions

• March 14th, 2018: Wednesday #haskell problem I am thinking about sumaSumasCuadradosDivisores from the Haskell community via @Jose_A_Alonso 
• March 12th, 2018: Monday's #haskell problem is Improving Tarski's logic language with curried functions. 
• March 8th, 2018: For Thursday's #haskell problem, thanks to @wtfunctional, we'll play with Other People's Code! ... and Mandelbrot sets. 
  
• March 6th, 2018: Tuesday's #haskell exercise asks the eternal question: "Are you my mom?"... and also looks at curried functions in a logic framework... which is nice. Last Tuesday's #haskell solution ended in pathos: I know 'momOf', but I cannot answer, 'are you my mom?' Can you not feel the sads? 
• March 5th, 2018: Monday's #haskell problem is a little "Introduction to Logic" by Tarski to start our week of right. I went a little 'forall' in the #haskell solution today. Sorry. #notsorry 
  

February 2018 1HaskellADay problems and solutions

• February 28th, 2018: Wednesday's #haskell problem: (sung:) Ya know that I love Casey's Mom! (guitar wails) Wednesday's #haskell solution we solve with LOGIC PROGRAMMING! ... IN HASKELL! YES! 
• February 27th, 2018: Tuesday's #haskell problem: A job scheduling problem, posed by Mensa, answered by Haskell. Tuesday's #haskell solution is this lcm/scan-like function, because "... the work is never done." 
• February 22nd, 2018: Thursday's #haskell problem is counting in Haskell IN YOUR NATIVE LANGUAGE! 
• February 21st, 2018: Wednesday's #haskell problem is sorting lists and making list-o-lists from lists, via P99 and random.org because my love-affair with #Prolog. *ahem* (pure) Prolog. Wednesday's #haskell solution: lsort and lfsort, provided to you by sortOn and by @bazzargh and @xgrommx 
• February 20th, 2018: Tuesday's #haskell problem: More fun with elements in a list! Whee! Tuesday's #haskell solution: the monad-y approach, with alternatives by @xgrommx @bazzargh @johannesweiss 
• February 19th, 2018: Monday's #haskell exercise is from @fermatslibrary: permutable primes. Monday's #haskell solution: Primes. Permutable. ... #PWND! 

• February 16th, 2018: For Friday's #haskell problem P99 Prolog problem 33, coprime numbers with an assist by random.org. Friday's #haskell solution is coprimes: as easy as you please! 
  
• February 15th, 2018: Thursday's #haskell problem is the greatest common denominator via P99 Prolog problem set. For today's #haskell solution I'VE JUST (re)INVENTED GCD! KNEEL BEFORE ZOD! 
• February 14th, 2018: For Wednesday's #haskell problem: "I'm thinking of a word that has the letter 'V' ..." Happy Valentine's Day ... as an anagram! Today's #haskell solution was defined with ALACRITY! YUS! 
• February 13th, 2018: Tuesday's #haskell problem: we have unicode, and we could parse that as Text, but why do that when they want plain, old ASCII? ... why, indeed! Removing non-ASCII characters from documents is one approach to parsing them... 
• February 12th, 2018: A little CSV to JSON work via #haskell today. Monday's #haskell solution: 🎵 You say CSV, I say JSON. Let's call the whole thing off! 🎵 
• February 9th, 2018: Logs get stale, quickly! Let's archive old log entries for Friday's #haskell problem... so quickly, in fact, that the Friday's #haskell solution archived 270k+ rows of log entries! 
• February 7th, 2018: So, we will update articles in our PostgreSQL data store, but what happens when your codec is LATIN1 and you need to store the article as UTF8? Wednesday's #haskell problem. Wednesday's #haskell solution: TFW the exercise is an update function, but the solution goes whole-hog! smh 🙄 
• February 6th, 2018: Tuesday's #haskell problem is timestamping packets and associating articles with the packets that downloaded them. Tuesday's #haskell solution: associating the articles we uploaded TODAY with the packet we uploaded TODAY! 
• February 5th, 2018: On #SuperBowlSunday I release Monday's #haskell problem which is to timestamp log entries, because that's what we live for. Today's #haskell solution was a stunning victory by the Philadelphia Eagles ... no, ... wait: it was log entries, timestamped. There. #nailedit 
  
  
  
  
• February 2nd, 2018: We've downloaded a the most recent set of articles from a REST endpoint; now we have to exclude Associated Press articles from our local paper's set. Friday's #haskell problem. Today's #haskell solution filters out Associated Press articles from out downloaded article-set. 
• February 1st, 2018: Thursday's #haskell problem is from a picture posted from @fermatslibrary: anagrams as prime-multiples. Teacher: Class the following words are all English words: "Elaps lapse Lepas Pales salep saple sepal slape spale speal" Class: ... <<- a="" haskell="" href="https://3.bp.blogspot.com/-XHCUGvp7QhE/WnM0upUr8xI/AAAAAAAACCU/wUk6_uprpi4kqC2VZbHzlgs7XJHC3mP1gCLcBGAs/s1600/anagram-by-primes.jpg" imageanchor="1" no="" s="" solution.="" style="color: #7d181e; margin-left: 1em; margin-right: 1em; text-decoration: none;" words.="" yesterday="">

January 2018 1Liner 1HaskellADay problems and solutions

• January 8th, 2018: from Nicoλas‏ @BeRewt
  A small @1HaskellADay, old-school. Define foo:
  
  > foo 3 [1..5]
  [([1,2,3], 4), ([2,3,4], 5)]
  
  > foo 2 [1..4]
  [([1,2], 3), ([2,3], 4)]
  
  > foo 2 [1..20]
  [([1,2],3), ([2,3],4), ..., ([18,19],20)]
  
  > foo 20 [1..2]
  []
• Demiurge With a Teletype @mrkgrnao
  foo n
    = tails
    # filter (length # (> n))
    # map (splitAt n # second head)
  
  (#) = flip (.)
• Andreas Källberg @Anka213
  I haven't tested it, but this should work:
  foo n xs = [ (hd,x) | (hd , x:_) <- n="" splitat=""> tails xs ]
• <- n="" splitat="">Nicoλas @BeRewt foo n = zip <$> fmap (take n) . tails <*> drop n
• January 5th, 2018: You have the following DAG-paths:
  
  a -> b -> c -> e
  a -> b -> d -> e
  q -> r -> s
  w -> x
  y -> z
  
  and many more.
  
  From a path, provide a bi-directional encoding* given maximum graph depth is, say, 7, max number of roots is, say, 10, and max number of nodes is, say, 1000.
• *bi-directional encoding of a graph path:
  
  DAG path -> enc is unique for an unique DAG path
  enc -> DAG path yields the same DAG path that created the unique enc.
  
  *DAG: "Directed, acyclic graph."
• January 5th, 2018: given s :: Ord k => a -> (k,[v])
  
  define f using s
  
  f :: Ord k => [a] -> Map k [v]
  
  with no duplicate k in [a]
• Christian Bay @the_greenbourne f = foldr (\e acc -> uncurry M.insert (s e) acc) M.empty
• me: you can curry away the acc variable easily
• Christian Bay @the_greenbourne You're right :)
  f = foldr (uncurry M.insert . s) M.empty
• Bazzargh @bazzargh fromList.(map s) ?
• me: Yuppers