Re: [Factor-talk] Some beginner questions with a focus on stack effects

[Alexander Ilin]

Hello, Luca!

 

 Such a humble email, I could not leave without an answer. Questions like this are completely appropriate on this mailing list.

 

 Here are a few changes I made to make your code compilable:

 

WAS:

: gather-input ( mapping -- seq\f )

    readln ;

 

CORRECT:

: gather-input ( -- seq\f )
   readln ;
 

 After that fix, the error you mentioned comes up, the one with the incorrect stack-effects in the `while` call.

 

WAS:

: solve ( -- )
   quote-mapping [ gather-input ] [ prepare-input process-input ] while drop ;
 

CORRECT:

: solve ( -- )
   quote-mapping [ gather-input dup ] [ prepare-input process-input ] while 2drop ;

 

 The `while` word consumes one output of the first quotation (the `?` parameter), so you need to `dup` the output of `gather-input` to keep a copy on the stack for the next quotation.

 

 The general approach you took is fine in Factor, as far as I can see. Here is my solution, similar to yours, but with a slightly different choice of the stream reading method. It uses recursion instead of the `while` loop:

 

 

USING: kernel io ;

IN: 00272_TEX_Quotes

: other-quote ( quote -- quote' )
   "''" = [ "``" ] [ "''" ] if ;

: process-input ( quote -- )
   "\"" read-until swap [ write ] when* [
       dup write other-quote process-input
   ] [ drop ] if ;

: solve ( -- )
   "``" process-input ;

MAIN: solve

 

 This is how I tested it in the listener:

"272.txt" utf8 [ solve nl ] with-file-reader
 

07.10.2018, 13:21, "Luca Di Sera" <bloodtype.si...@gmail.com>:

Hello to all,

 

I'm a beginner factor programming.

I learned about factor in the "Seven More Languages in Seven Weeks" book and could not avoid falling in love with it.

I'm trying to learn it at work in my lunch-breaks and one of the projects I'm following as a didactical exercise is to use it as a secondary language ( to C++ ) to solve competitive programming exercises.

 

Now, It is a few days that I'm stuck on some non-working code that I can't seem to solve on my own, for how embarrassing that is.

First of all this is the code:

 

USING: syntax sequences kernel io splitting ;

IN: UVa.00272_TEX_Quotes.Factor.00272_TEX_Quotes

 

CONSTANT: quote-mapping { "``" "''" } 

 

: switch-quote ( mapping x -- mapping x )

    [ reverse ] dip ;

 

: print-quote ( mapping x -- mapping x )

   [ dup first write ] dip ;

 

: gather-input ( mapping -- seq\f )

    readln ;

 

: prepare-input ( str -- seq )

    "\"" split ;

 

: process-input ( mapping seq -- mapping )

    [ print-quote switch-quote write ] each ;

 

: solve ( -- )

    quote-mapping [ gather-input ] [ prepare-input process-input ] while drop ;

 

MAIN: solve

 

This code is a, currently in testing, solution to UVa 272.

What I was trying to do is the following :

 

1.  Read all lines of input one by one
2. For each line that is read split it in a sequence where the " are
3. Print the sequence back with quotes between each element ( changing between closing and opening quotes )  

Now, the main problems I'm getting with this codes are related to stack-effects. In particular after solving some of the errors the one I can't currently solve is the following:

 

The word solve cannot be executed because it failed to compile

 

The input quotations to “while” don't match their expected effects

Input                           Expected         Got

[ gather-input ]                ( ..a -- ..b ? ) ( x -- x )

[ prepare-input process-input ] ( ..b -- ..a )   ( x x -- x )

 

From my understanding, this is an unbalanced-branches-error.

By reading the documentation I can see what this error is about, yet the reasons it is appearing and the method to correct the code are still flying over my head.

I supposed it may have something to do with the way I'm passing the values around on the stack.

 

Hoping this is an accepted topic on this mailing list, I was hoping someone could help me by explaining, or by providing resources I can deepen my understanding with,  how this code should actually be written to work and if there is some essential-error I'm doing in using factor in this code ( And I would be most thankful for suggestion on how to actually write something like this in a Factor way ).

 

 

---=====---

Александр

 

_______________________________________________
Factor-talk mailing list
Factor-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/factor-talk