Humour (137)

3 décembre 2018


Les bons conseils

20 avril 2018

Les meilleurs conseils sont souvent gratuits…  Comme ceux de Rob Pike, Rules Of programming.


Humour (128)

1 avril 2018


Serpents et échelles

20 décembre 2017

Aucun enfant n’y a échappé, nous avons tous joué au jeu de Serpents et échelles (Snakes and Ladders) plus jeunes! Là où ce jeu devient intéressant pour un adulte, c’est lorsqu’il s’agit d’en étudier les aspects mathématiques!

Combien de coups sont en moyenne nécessaires pour une partie? De quoi a l’air la distribution des coups nécessaires? À quoi ressemble la fonction de répartition?  Plutôt que d’utiliser une chaîne de Markov pour trouver ces réponses, quoi de plus simple et rapide que de simuler des parties?

C’est exactement ce qu’a fait Jake VanderPlas tel que décrit dans cet article.


Mu, sigma & friends

1 mai 2017

You’re a programmer and sometimes you need some statistical tools and knowledge?  Here’s a short list of resources that could be helpful to you.


The patch

28 février 2017

Have you ever asked yourself why program fixes are called « patches »?  I never did.  I’ve never known why a code fix was called a patch either.  Now I know why!

the_patch


There’s a T-shirt for you

31 janvier 2017

Every developer can relate…

Those shirts are hilarious!  A few examples…

testinproduction

requirements

itworkedonmymachine

projectmanager

weeksofcoding

 


Naughty strings

19 janvier 2017

Testing user-input can be tedious.  And complex.  And long.  Fortunately, whether you want to test your software against SQL injection, basic (or complex) user-input validation or reserved keywords that could make your software balk, there is now a big list of all the crap and garbage you could ever imagine. A nice and easy way to test the GIGO concept!

It’s all here (and free!), the Big List Of Naughty Strings!


Sorting

7 décembre 2016

Remember sorting algorithms, big O notation and the countless hours you spent implementing all kinds of crazy sorts?  Well, my university days are far behind me and it was kind of fun to be reminded of all the time I once wasted coding that stuff over and over, before libraries, frameworks and code reuse would be the norm!

Here’s a nice & short video showing 15 different sorting algorithms in action.

Here’s another short video showing 24 different sorting algorithms perform at the same time.

Save


How do you parse fractions with GREL and OpenRefine?

13 août 2016


I just came across a question related to parsing fractions on StackOverflow and I thought about how easy that kind of stuff could be done Smalltalk (because we have real fractions!).  And then, I wondered how hard that challenge would be with something as limited as GREL (the language used in OpenRefine, namely Google Refine Expression Language).  So I tried to do it in GREL!

Here’s the challenge : how do you go from a textual representation of a fraction like -10 2/3 to the number -10.66666 with GREL?

Here’s my test data:

fraction, target
10,10
-10,-10
10 2/3, 10.666666666666666
-10 2/3, -10.666666666666666
2/3, 0.666666666666666
-2/3, -0.666666666666666
+10,10
+10 2/3, 10.666666666666666
+2/3, 0.666666666666666
2,2
-2,-2
2 2/3, 2.666666666666666
-2 2/3, -2.666666666666666
+2,2
+2 2/3, 2.666666666666666
+2/3, 0.666666666666666
10 11/13, 10.8461538461538461
-10 11/13, -10.8461538461538461
11/13, 0.8461538461538461
-11/13, -0.8461538461538461
+10 11/13, 10.8461538461538461
+11/13, 0.8461538461538461
17/4, 4.25
+17/4, 4.25
-17/4, -4.25

The trick is to start small and focus on one case at a time! Just make sure your logic isn’t flawed first and then you can work on each specific case.  Here’s how I did it.

Start by making sure your code logic is right:

if(contains(value, « / »), if(contains(value,  » « ), « integer with fraction », « fraction »), « integer »)

Then, for each case, dig deeper and make sure you identify each case precisely. In the expression above, replace :

« integer with fraction » with:
if(startsWith(value, « -« ), « negative integer with fraction », « positive integer with fraction »)

« fraction » with:
if(startsWith(value, « -« ), « negative fraction », « positive fraction »)

« integer » with:
if(startsWith(value, « -« ), « negative integer », « positive integer »)

You should end up with something like this:

if(contains(value, « / »), if(contains(value,  » « ), if(startsWith(value, « -« ), « negative integer with fraction », « positive integer with fraction »), if(startsWith(value, « -« ), « negative fraction », « positive fraction »)), if(startsWith(value, « -« ), « negative integer », « positive integer »))

Now test!

fractions logic

Once every case is covered, now you’ll have to transform every of the 6 cases in code to extract and convert those numbers. It’s just a matter of splitting every part of the number with GREL’s usual friendly methods, namely indexOf, substring and startsWith.  And since the division operator (/) in GREL behaves like the floor function, you need to multiply your results (the strings that have been converted to numbers using toNumber) by 1.0 to force a conversion to a decimal number.

It’s really painful to code but, honestly, all you need is a bit of patience and make sure you deal with one part of the string at a time (and one case at a time)!

Wanna test your code?  Just select the « fraction » column and apply the code below in a transform expression : the target value and the transformed value should be the same in the preview pane!

strings to fractions

So here’s the GREL code to parse fractions (and convert them to decimal) :

if(contains(value, "/"), if(contains(value, " "), 
if(startsWith(value, "-"), -1.0*(toNumber(substring(value, 1, indexOf(value, " ")))+
(1.0*toNumber(substring(value, indexOf(value, " "), indexOf(value, "/")))/
toNumber(substring(value, indexOf(value, "/")+1)))), 
(toNumber(substring(value, 0,indexOf(value, " ")))+
(1.0*toNumber(substring(value, indexOf(value, " "), 
indexOf(value, "/")))/toNumber(substring(value, indexOf(value, "/")+1))))), 
if(startsWith(value, "-"), 1.0*toNumber(substring(value, 0, indexOf(value, "/")))/
toNumber(substring(value, indexOf(value, "/")+1)), 
1.0*toNumber(substring(value, 0, indexOf(value, "/")))/
toNumber(substring(value, indexOf(value, "/")+1)))), 
if(startsWith(value, "-"), toNumber(value), toNumber(value)))

Enjoy!
NOTE: for some strange reason, WordPress keeps changing my double quotes 
with the French quotes.  Don't worry, it's all regular English quotes you 
should see!

Save

Save

Save

Save

Save

Save

Save

Save