Since [strippathname] segfaults for me, does anybody have another idea to convert the paths sent by [openpanel] to relative paths, or to strip everything but the filename?
thx! d.
derek holzer wrote:
Since [strippathname] segfaults for me, does anybody have another idea to convert the paths sent by [openpanel] to relative paths, or to strip everything but the filename?
use [s2l] (zexy) with "/" as delimiter, and getthe last element of the list (with something like [packel -1])
mfg.asa.drx IOhannes
On Tue, 4 May 2004, derek holzer wrote:
Since [strippathname] segfaults for me, does anybody have another idea to convert the paths sent by [openpanel] to relative paths, or to strip everything but the filename?
As these externals are probably "influenced" by the old stripdir that was in my library, you might use the original one. Of course if Tom just copied the code and changed the name it might suffer the same problem :)
Joking aside. I had an interesting discussion with Victor Lazzarini at the ZKM this weekend. We more or less agreed that one problem that causes the proliferation of externals is that for any mediocre programmer it is easier to write such an external from scratch than to actually search for an existing one.
Another problem is that general symbol manipulations are missing. (e.g something like a "sed" external, which would solve this and all related problems)
I tried to solve this problem with the CVS and single externals, the idea was to make sure that not two objects with the same name and only slightly different behaviour exist, or, the other way round, with different names and doing the same ...
Anyhow, meeting some of the developers at ZKM too, we found a solution that we all could agree on. More on that later on the devel list ..
Guenter
thx! d.
-- derek holzer ::: http://www.umatic.nl ---Oblique Strategy # 57: "Do the words need changing?"
PD-list mailing list PD-list@iem.at to manage your subscription (including un-subscription) see http://iem.at/cgi-bin/mailman/listinfo/pd-list
hi Guenter,
I do not think anybody skillful enough for sed scripting would have any trouble using scripting externals that exist already (a simple example: [+tot file tail $in] -> [+tot +out])
Btw, where is strippathname' from? I found astrippath'
(apparently unfinished) in cvs, which conflicts with the
cyclone's one -- currently a dummy, but it is going to be
implemented, cross-platformly, some day.
Krzysztof
guenter geiger wrote: ...
Another problem is that general symbol manipulations are missing. (e.g something like a "sed" external, which would solve this and all related problems)
On Thu, 6 May 2004, Krzysztof Czaja wrote:
hi Guenter,
I do not think anybody skillful enough for sed scripting would have any trouble using scripting externals that exist already (a simple example: [+tot file tail $in] -> [+tot +out])
Yes, right. I didn't mean it that literally. But how could you know. I thought more about the simple functionally bash variables have, but translated to pd talk
[strip */] [strip /*]
and similar things.
Btw, where is
strippathname' from? I found astrippath' (apparently unfinished) in cvs, which conflicts with the cyclone's one -- currently a dummy, but it is going to be implemented, cross-platformly, some day.
As always, a wonderful addition. Finally something that works.
Guenter
guenter geiger wrote: ...
Another problem is that general symbol manipulations are missing. (e.g something like a "sed" external, which would solve this and all related problems)
Krzysztof Czaja wrote:
Btw, where is `strippathname' from?
The one that was segfaulting my box came from IEMlib2.
d.
-
derek holzer -
guenter geiger -
IOhannes m zmoelnig -
Krzysztof Czaja