I think it's commendable to vertically align code. (Just not with tabs.) It makes it look neater.
post
the fact that these are strings instead of an object that is broken up by country code, area code, and number makes me irrationally angry.
Could be tests for a parser to convert it from string to object.
Not like your end users are going to type each piece into a separate field.
i think you think that telephone numbers are well-structured. they are not. they are messy. they do not fit a certain schematic.
I recommend also the following topic: "people have names". https://www.kalzumeus.com/2010/06/17/falsehoods-programmers-believe-about-names/
Names do not in general fit into the schematic "first name, last name"
Is there not a "falsehoods programmers think about phone numbers" yet?
Edit: And once again, I'm still confused about some of these. Do we need to expand unicode for names? It's supposed to be universal. WTF is up with 40?
WTF is up with 40?
People have names.
I suppose that a counterexample to this might be Tibetan children, who get named at puberty, IIRC. Before that, they have no names. They are just referred to as "child" or "somebody's child".
People’s names are all mapped in Unicode code points.
I suppose a counterexample to that might be cultures which do not use script in general. Then, obviously, there's no Unicode characters for these non-existant glyphs.
Ah, so it dovetails with the whole "children get a name reasonably fast" thing. I was interpreting that as "ever, in a natural lifespan". My bad, haha.
I suppose a counterexample to that might be cultures which do not use script in general. Then, obviously, there’s no Unicode characters for these non-existant glyphs.
True, but there's little risk of a name being entered into a form without some kind of transcription.
You absolute buffoon. How do you figure this code isn't testing how to parse a string into such an object??
Found the Java programmer...
"The number" is itself two parts hence the dash. The first section being the prefix and the last part being the line number.
Not gonna lie, that vertical alignment is making my brain purr like a kitten
Junior devs don’t know Jenny.
It's true. Who's Jenny, for the zoomers in the chat?
It was a very popular song in the 80s. Jenny’s phone number was 867-5309.
https://en.m.wikipedia.org/wiki/867-5309/Jenny
If you’re ever at a store and need a frequent shopper card, this number with your local area code is usually registered.
Interesting how Wikipedia handles slashes in the title.
Good to know! Although I have to wonder if that's true for programs more recently set up as well. That was >40 years ago.
For test data I like to use different movie or comic book names for different epics.
So when we start a new epic with multiple user stories I decide this will be all star wars, or matrix or donald duck characters, etc etc. The nice thing about it is when certain test data gives unexpected errors in other parts of application its easy to track where the test data is from.
Life Pro Tip - Jenny's number also (usually) works at the grocery checkout line to get the advertised store member discount on items. Choose whichever area code you like, someone has likely set it up in the past.
Jim Bo Bob 555-867-5309 123 test Street, Testville Ca, 55555
Is my goto
From someone who does this for a living... vary your names and addresses. Less chance of collisions if your suite teardown fails to clean up properly. Depending on your needs, having a hard-coded unique name/address per test can be fine, or if you're using Python, there's a library called Faker that will generate ISO-valid test data. It's also a bit easier to see where a teardown failed if maybe an exception got swallowed.
I once made test users who were all named John Smith in different languages. I learned that basically every name is a variant of John.
all 27 comments