fix use of alt code in 2015/endoh4
This commit is contained in:
1 parent
201516e67b
commit
0d9565dd4b
3 files changed
+191
-106
No files matched your search
+15
-6
@@ -4,11 +4,11 @@ Yusuke Endoh
|
||||
<https://github.com/mame/>
|
||||
Twitter: @mametter
|
||||
|
||||
## To build:
|
||||
|
||||
## Judges' comments:
|
||||
### To use:
|
||||
make all
|
||||
|
||||
make
|
||||
### To run
|
||||
|
||||
./prog <a number of arguments>
|
||||
|
||||
@@ -22,6 +22,16 @@ Twitter: @mametter
|
||||
|
||||
./prog $(< Makefile)
|
||||
|
||||
### Alternate code
|
||||
|
||||
An alternate version of this entry, prog.alt.c, is provided. This alternate code does not contain improvements suggested by Gil Dogon.
|
||||
|
||||
To compile this alternate version:
|
||||
|
||||
make alt
|
||||
|
||||
Use `prog.alt` as you would `prog` above.
|
||||
|
||||
### Selected Judges Remarks:
|
||||
|
||||
For this one liner it is best to avoid prime numbers. The output groups
|
||||
@@ -29,8 +39,8 @@ factors in the x-y dimensions. How many factors can be represented before you
|
||||
run out of command line argument space? How could you modify the program to
|
||||
experiment with numbers that have more or larger factors?
|
||||
|
||||
|
||||
## Author's comments:
|
||||
|
||||
Visual factorization 1-liner. Try:
|
||||
|
||||
gcc -o prog prog.c
|
||||
@@ -44,8 +54,7 @@ Visual factorization 1-liner. Try:
|
||||
|
||||
Inspired by [Animated Factorization Diagrams](http://www.datapointed.net/visualizations/math/factorization/animated-diagrams/).
|
||||
|
||||
|
||||
EDIT: `prog.orig.c` is the original version that I submitted. Gil Dogon kindly improved it during the review. Thank you!
|
||||
EDIT: `prog.alt.c` is the original version that I submitted. Gil Dogon kindly improved it during the review. Thank you!
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
<!--
|
||||
|
||||
Reference in new issue
Block a user