OFFSET:LENGTH vs. OFFSET:OFFSET may be purely semantics. The relationship that I showed in the table is exactly the way I coded it in my script. Can you come up with any new tests which aren't successful?

> OFFSET:LENGTH vs. OFFSET:OFFSET may be purely semantics.

doesn't seem so.

https://docs.python.org/3/glossary.html#term-slice

slice
An object usually containing a portion of a sequence. A slice is creat
+ed using the subscript notation, [] with 
colons between numbers when several are given, such as in variable_nam
+e[1:3:5]. The bracket (subscript) notation
uses slice objects internally.
[download]

https://docs.python.org/3/library/functions.html#slice

class slice(stop)
class slice(start, stop, step=1)
Return a slice object representing the set of indices specified by ran
+ge(start, stop, step). The start and step 
arguments default to None. 

...
[download]

something is off (-set ;) with this test suite. Unfortunately I have no Python at hand to test ...

updates

references and citations updated

Cheers Rolf
_{(addicted to the 𐍀𐌴𐍂𐌻 Programming Language :)
Wikisyntax for the Monastery}

In the OP, a notation that looked like [x:y], where x or y could be blank, was given. I noted this equivalence:

Python	Perl
[x:y]	splice ARRAY, x, y
[:y]	splice ARRAY, 0, y
[x:]	splice ARRAY, x
[:]	splice ARRAY, 0

I coded that and it passed all tests. I took all given tests, and their results, on face value. One exception is [:], which I added myself and was extrapolated from information already provided.

If the original data was wrong in some way, then I'm working from a false premise. If the tests are insufficient, further tests could show that the seen equivalence doesn't hold for other values.

I'm not trying to sell something here. I don't care what the OP uses for his "poetry" project — as already stated: "What you choose is entirely up to you.".

If you think that there's something wrong with the source, please take it up with the OP, not me.

— Ken

Here's one:

    [\@test_array, '[3:4]', 'd'],
[download]

In my OP I mentioned needing help from "bilingual" (i.e. Perl and Python) people; otherwise I'd have included a description of Python's slice notation. I'm sorry if this was negligence on my part, but I assumed that anyone answering would be familiar with Python's oddball slice semantics. [a:b] means "the characters starting at offset a and ending at offset b-1". Unless a or b is negative in which case .. hell, I don't remember ;-)

The code I posted correctly slices arrays using Pythonic parameters, although it doesn't attempt to handle the optional third argument, a step value. And I'm happy to say that the translation effort has succeeded and Perl is now churning out gibberish sonnets at a much faster rate than its Python equivalent.

Update:

It correctly slices arrays if the arguments aren't nonsensical -- like being out of range, which Python forgives. It needs a few more sanity tests to handle those cases.

> Unless a or b is negative in which case .. hell, I don't remember ;-)

AFAIR from a previous discussion is Perl using closed intervals while Python using half open ones.

That means the end point is excluded in Python.

One advantage is that [:X] and [X:] are splitting an interval including X exactly into two dijoint parts, without needing to decrement the first end boundary.

I.O.W always $end-- when translating to Perl.

Cheers Rolf
_{(addicted to the 𐍀𐌴𐍂𐌻 Programming Language :)
Wikisyntax for the Monastery}