make 'error-prone' usage consistent

This commit is contained in:
Thibault Kruse 2017-06-25 12:34:09 +09:00
parent 9aa9e8fc9f
commit d14067a5b4

View File

@ -822,7 +822,7 @@ Avoid errors leading to (possibly unrecognized) wrong results.
##### Example ##### Example
void increment1(int* p, int n) // bad: error prone void increment1(int* p, int n) // bad: error-prone
{ {
for (int i = 0; i < n; ++i) ++p[i]; for (int i = 0; i < n; ++i) ++p[i];
} }
@ -5593,7 +5593,7 @@ The common action gets tedious to write and may accidentally not be common.
##### Reason ##### Reason
If you need those constructors for a derived class, re-implementing them is tedious and error prone. If you need those constructors for a derived class, re-implementing them is tedious and error-prone.
##### Example ##### Example
@ -9898,7 +9898,7 @@ Flag variable and constant declarations with multiple declarators (e.g., `int* p
##### Reason ##### Reason
* Simple repetition is tedious and error prone. * Simple repetition is tedious and error-prone.
* When you use `auto`, the name of the declared entity is in a fixed position in the declaration, increasing readability. * When you use `auto`, the name of the declared entity is in a fixed position in the declaration, increasing readability.
* In a template function declaration the return type can be a member type. * In a template function declaration the return type can be a member type.
@ -12890,7 +12890,7 @@ When did you even want to sort memory?
Really, we sort sequences of elements, typically stored in containers. Really, we sort sequences of elements, typically stored in containers.
A call to `qsort` throws away much useful information (e.g., the element type), forces the user to repeat information A call to `qsort` throws away much useful information (e.g., the element type), forces the user to repeat information
already known (e.g., the element size), and forces the user to write extra code (e.g., a function to compare `double`s). already known (e.g., the element size), and forces the user to write extra code (e.g., a function to compare `double`s).
This implies added work for the programmer, is error prone, and deprives the compiler of information needed for optimization. This implies added work for the programmer, is error-prone, and deprives the compiler of information needed for optimization.
double data[100]; double data[100];
// ... fill a ... // ... fill a ...