| 1 |
On 2017-10-12 08:36, Alan McKinnon wrote: |
| 2 |
|
| 3 |
> > A more serious problem is how to find all the situations where |
| 4 |
> > /usr/local is baked in. It's not as simple as grep because when I |
| 5 |
> > could, I relied on the implicit PATH which would be configured |
| 6 |
> > somewhere else, or it might not even be configured - it might be |
| 7 |
> > compiled in (I think this is the case for some programs in the |
| 8 |
> > shadow package, and perhaps PAM modules). |
| 9 |
> |
| 10 |
> Not sure what the context is here. Are you talking about packages and |
| 11 |
> ports code, or your own stuff you compiled yourself? |
| 12 |
> |
| 13 |
> Either way, what problem were you asking about by writing that |
| 14 |
> paragraphs? |
| 15 |
|
| 16 |
GNU/Linux packages that set the PATH, on which my script stuff ends up |
| 17 |
relying. If I just move my scripts to, say, /opt/foobar, and edit the |
| 18 |
all the occurences of "/usr/local", something will still break because a |
| 19 |
script that is found now by virtue of the ambient PATH won't be found in |
| 20 |
the new world. |
| 21 |
|
| 22 |
-- |
| 23 |
Please don't Cc: me privately on mailing lists and Usenet, |
| 24 |
if you also post the followup to the list or newsgroup. |
| 25 |
Do obvious transformation on domain to reply privately _only_ on Usenet. |
Archives