1 |
On Sun, 20 Dec 2015 00:56:39 +0100 |
2 |
Ulrich Mueller <ulm@g.o> wrote: |
3 |
|
4 |
> >>>>> On Sat, 19 Dec 2015, Michał Górny wrote: |
5 |
> |
6 |
> >> set -f |
7 |
> >> echo -e ${DOC_CONTENTS} | ... |
8 |
> >> |
9 |
> >> versus: |
10 |
> >> |
11 |
> >> read -d '' -r -a DOC_CONTENTS <<<"${DOC_CONTENTS}" |
12 |
> >> echo -e "${DOC_CONTENTS[*]}" | ... |
13 |
> >> |
14 |
> >> The second one is (IMHO) harder to understand, less efficient, and |
15 |
> >> relies on undocumented behaviour. |
16 |
> |
17 |
> > On WHAT?! |
18 |
> |
19 |
> Documentation of the read command doesn't specify behaviour for an |
20 |
> empty argument of the -d option: |
21 |
> |
22 |
> `-d DELIM' |
23 |
> The first character of DELIM is used to terminate the input |
24 |
> line, rather than newline. |
25 |
|
26 |
Feel free to use $'\0' if that makes it more explicit for you. |
27 |
|
28 |
-- |
29 |
Best regards, |
30 |
Michał Górny |
31 |
<http://dev.gentoo.org/~mgorny/> |