Timeline for Build a Compiler Bomb
Current License: CC BY-SA 3.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 17, 2021 at 19:58 | comment | added | Aleksi Torhamo |
@AnshumanKumar: In Python interactive shell? If you put it into a variable (so eg. v = (1<<19**8,)*4**7), and you have enough free memory (a bit over 2GB), nothing much - it'll take a second or two to complete, and the process will use that much more memory. If you don't put it into a variable, Python will try to display the number, which means it needs to convert it to decimal first - which will take some time (and more memory) with a number that big. How long? Depends on the computer, but on the order of a year. But after that, it should finally print the multi-gigabyte output for you!
|
|
| Feb 17, 2021 at 19:56 | comment | added | aivarsk | Just a 2021 update: constant folding has been fixed in current versions of Python3. I managed to reproduce the compiler bomb with python3.5 from deadsnakes PPA. | |
| Feb 17, 2021 at 15:24 | comment | added | Anshuman Kumar | Honest question : If I run this in my terminal what will happen? | |
| Jun 17, 2020 at 9:04 | history | edited | Community Bot |
Commonmark migration
|
|
| Oct 6, 2016 at 23:34 | comment | added | user16488 | IIRC, python uses 30 bits per 32-bit word in its integer representation | |
| Jan 17, 2016 at 21:17 | comment | added | Aleksi Torhamo | @Dave: Ahhh, I totally misunderstood what you meant. :P | |
| Jan 17, 2016 at 21:04 | comment | added | Dave |
@AleksiTorhamo it's small because I had to change it to 1<<18**8 instead of 1<<19**8. The full version failed with an invalid IO operation (so I'll check it on another OS & filesystem)
|
|
| Jan 17, 2016 at 18:17 | comment | added | Aleksi Torhamo |
@Dave: The exact size might depend on the version (1.5GB sounds weird no matter what, though); I was using Python 3.3.5, and used python -m py_compile asd.py to generate the .pyc-file.
|
|
| Jan 17, 2016 at 17:22 | comment | added | Dave |
Nice. Since this is only 13 bytes, we finally have a challenger to the first-posted answer! I was only able to confirm 1<<18 on my machine (1.5GB) but I'll test it on linux later, where I expect it will work with the full 8GB (not going to try the 32TB version!)
|
|
| Jan 15, 2016 at 8:29 | history | edited | Aleksi Torhamo | CC BY-SA 3.0 |
Use smaller code that still generates >4GB output for a better "score"
|
| Jan 14, 2016 at 10:46 | comment | added | Aleksi Torhamo | @ChristianIrwan: Yeah, I'd forgotten that rule, only realized it a few minutes ago and haven't figured out what kind of edit I should make yet. :-) | |
| Jan 14, 2016 at 8:58 | history | answered | Aleksi Torhamo | CC BY-SA 3.0 |