5% of 666 Python repos had comma typo bugs (inc V8, TensorFlow and PyTorch)

Post by jstx1 » Fri, Jan 07, 2022, 9:59 PM UTC

dylan604 wrote: post #222

jstx1 wrote: post #133

Earlier quoted context omitted.

Yes, it is crazy. I guess this isn't really the place for it but ... From the official docs: The Figure is the final image that may contain 1 or more Axes. The Axes represent an individual plot (don't confuse this with the word "axis", which refers to the x/y axis of a plot). This is infuriatingly bad and I firmly believe that it makes sense only to people who already know how it works. There's an image, axes (this w…

>axes (this word alone is a crime), why so? you prefer something like axiis?

Post by onphonenow » Fri, Jan 07, 2022, 10:06 PM UTC

dragonwriter wrote: post #165

onphonenow wrote: post #160

Earlier quoted context omitted.

I'd totally agree - there's been a burst of sort of the perl style stuff (:= ?) to gain relatively small wins. ie, instead of for line in lines: print(line) we are supposed to be using while line := f.readline(): print(line) I've not been super impressed with this type of thing. That said, string formatting is better with f strings. They also rolled back some the forced breakage from trying to force unicode with 3 wh…

> ie, instead of > for line in lines: print(line) > we are supposed to be using > while line := f.readline(): print(line) No, we’re not. Walrus, in loops, IME, is more for replacing this pattern: while True: myvar = get_it() if not ok(myvar): break # code that uses myvar with this pattern: while ok(myvar := get-it()): # code that uses myvar

  if any(len(longline := line) >= 100 for line in lines):
     print("Extremely long line:", longline)

    for line in lines:
        if len(line) >= 100:
            print("Extremely long line:", line)
            break

Post by trulyme » Fri, Jan 07, 2022, 10:11 PM UTC

pfisherman wrote: post #129

micimize wrote: post #62

For those looking to avoid this specific problem, there is a flake8 rule: https://pypi.org/project/flake8-no-implicit-concat . More broadly, the https://codereview.doctors makers are making the point that their tool caught an easy-to-miss issue that most wouldn't think to add a rule for. A bit of an open question to me how many of those there really are at the language level, but still seems like a neat project.

Ime, Black will add parenthesis to clearly and explicitly indicate a tuple where there is trailing comma. Figured this out when I made the trailing comma mistake and wondered why Black kept reformatting my code.

Post by hsbauauvhabzb » Fri, Jan 07, 2022, 10:13 PM UTC

aeturnum wrote: post #87

The high-level goals of python end up creating these little syntactic landmines that can get even experienced coders. My personal nomination for the worst one of these is that having a comma after a single value often (depending on the surrounding syntax) creates a tuple. It's easy to miss and creates maddening errors where nothing works how you expect. I've moved away from working in Python in general, but I think t…

Post by trulyme » Fri, Jan 07, 2022, 10:18 PM UTC

aeturnum wrote: post #138

aylmao wrote: post #99

Earlier quoted context omitted.

The lack of a static type-system is IMO what makes these one-character mistakes very annoying. The compiler can't tell you something is wrong, so you're just left to figure out why things are broken, just to realize it was the smallest of typos.

I love how simple and forgiving Python is for small projects. The "trailing comma creates a tuple" situation comes out of, as far as I can tell, a desire to create maximally convenient syntax in the scenarios where tuples are intended. I think that's great for small code! I just wish that the core team would take that same zeal for a "pythonic" experience with small code and use it to develop more scaled-up systems f…