Rework lessons on strings #628

encukou · 2020-06-08T13:55:35Z

In in-person courses, I tended to explain strings in more depth than what the materials provided.
There's enough depth in quoting and characters (esp. special characters) to make it a separate lesson; same goes for the new concepts of indexing/slicing and methods.

This splits the material into:

String literals (how to write strings; what is a character)
String indexing and slicing
String methods
F-strings (split out earlier)

String comparisons might come later :)

Please review the changes as a whole, don't look at individual commits. But, to make cherry-picking easier, don't squash when merging.

In in-person courses, I tended to explain strings in more depth than what the materials provided. There's enough depth in quoting and characters (esp. special characters) to make it a separate lesson; same goes for the new concepts of indexing/slicing and methods. This splits the matrial into: - String literals (how to write strings; what is a character) - String indexing and slicing - String methods - F-strings (split out earlier) String comparisons might come later.

Kobzol · 2020-06-08T20:10:07Z

It looks good to me! One thing that I think could be improved, but it's not a big deal:
Maybe we should attempt to justify zero-based indexing without resorting to statements such as "it's black magic"? Pointer-based reasoning is too complex and also not really applicable to Python, but maybe we can find better justifications.

There are multiple reasons why 0-based indexing is useful (open-closed intervals, multi-dimensional indexing, arrays represented with pointers to their first element, using modulo to calculate indices etc.). Some of them are too difficult for this part of the course, but maybe we can pick some examples that they can understand? Or if we can't find any, say that we will show some examples later and add specific references to zero-based indexing later. For example:

Jak sis možná už všiml{{a}}, programátoři počítají od nuly.
...
Proč je tomu tak? Zjednodušeně řečeno je to z mnoha důvodů praktičtější.
Některé důvody jsou zatím moc komplikované a odkážeme se na ně později, ale prozatím si například můžeš všimnout, že:

<< intuitivní a jednoduchý příklad >>
<< např.: K indexování používáme v Pythonu celá čísla. Kladné indexy nám vrací prvky od začátku řetězce, záporné indexy prvky od konce řetězce, ale pokud bychom začali počítat od jedničky, jaký prvek by poté vrátil index 0? Bylo by nepraktické mít hodnotu indexu, která by nikdy nebyla validní >>

Počítání od nuly se používá ve většině programovacích jazyků, a i když ti prozatím může přijít zvláštní, časem si na něj určitě zvykneš. Můžeš si to představovat tak, že (kladný) index N udává počet prvků od začátku řetězce/pole, o které se musíš posunout, aby ses dostal(a) k žádanému prvku.

Later, for example in slicing/substrings (s[0:n]), we can refer to this with something like "Zde je příklad toho, proč je indexování od 0 přirozenější" and talk about open-closed intervals.

encukou · 2020-06-08T20:57:34Z

Thanks! I opened #631 with this idea.

I would like to have this PR merged if it is an improvement, not if it is perfect. The PR doesn't change the explanation of zero-based indexing, so IMO this review is not a place to improve it.
Also, as with any idea for the materials, I'd like to test yours in a live class before it's merged to the self-study course.

Kobzol · 2020-06-08T21:32:06Z

You're right, it's a separate discussion.

fivaldi · 2020-06-11T06:37:40Z