1 |
My god, I am a horrible self-editor. Sorry. Please ignore the magnum |
2 |
opus above and allow me to try again. |
3 |
|
4 |
In dev-lang/python*, we use |
5 |
|
6 |
append-ldflags '-L.' |
7 |
|
8 |
to ensure linking is performed against the built libpython.so in-tree, |
9 |
rather than than in the one in $(libdir). But, this doesn't work if |
10 |
LDFLAGS contains "-L$(libdir)". |
11 |
|
12 |
We could try to fix this like: |
13 |
|
14 |
export LDFLAGS="-L. ${LDFLAGS}" |
15 |
|
16 |
or so. That would cover 99.9% of the cases out there. But very rarely, |
17 |
indiscriminately placing our '-L.' before every other clause in LDFLAGS |
18 |
might cause an unanticipated side-effect. |
19 |
|
20 |
The flag-o-matic patch in my previous post analyses LDPATH and finds the |
21 |
first statement that might possibly cause the ld command-line |
22 |
library-search-path to be impacted, and inserts the "-L." just before |
23 |
that statement, hopefully achieving the desired result with the least |
24 |
possible side-effects. |
25 |
|
26 |
I wonder: do people feel this distinction is meaningful enough to |
27 |
justify actually making it? Or am I just being an anal-retentive nut-job? |
28 |
|
29 |
If you tried to read my first post, thanks for not offing yourself like |
30 |
the guys who get stuck sitting next to Ted Striker in _Airplane!_ |
31 |
|
32 |
-gmt |