Re: [Fink-devel] cctools virtual package

[Benjamin Reed]

On Saturday, February 15, 2003, at 03:18 PM, David R. Morrison wrote:


Ben Reed will use this in qt3 and kde, to detect whether folks have 
upgraded
to the December 2002 tools or not.

Yup, just put together some test qt3 and kde packages and haven't had 
any problems yet.  The cctools dep works like it's supposed to.



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] index behaviour change (post 0.11.x)

[Carsten Klapp]

I can confirm this, the same thing happened to me once within the last 
two or three days--due to a patch file which was no longer 
present/pertinent. No insight to offer towards a solution, sorry :( 
'sudo fink index' bypassed the problem.

On Saturday, February 15, 2003, at 10:03  pm, Ben Hines wrote:

Speaking of indexing breakage, there is also a bug in the current code 
again, recently i have selfupdate-cvsed, then update-alled a few times 
recently, and its tried to build and old version, dying due to lack of 
patch file (since its not really there anymore..). Only once or twice 
has this happened.



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] new package manager release

[David R. Morrison]

Sure, Max, I realize you are busy, and I'm not trying to make more work for
you!

I've revised my patch for the cctools virtual package, so I hope that this
can also be included in the next release.

  -- Dave


---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] index behaviour change (post 0.11.x)

[Ben Hines]

On Saturday, February 15, 2003, at 05:43  PM, Max Horn wrote:



To ease the developer's pain somewhat, we could add a "fink index 
foo.info" command, which would add the specified .info file(s) to the 
index. This way, you could just (re)index the files you are working 
on, avoiding having to do full indexing too often.


Actually, I was thinking of something like:
"fink install foo.info" to perform the chosen operation on that info 
file.

We would have to handle the cases where the info file is outside the 
tree (and where to put the deb if so - local? cwd?). And we would have 
to figure out where in the tree it is, if it is in the tree, we could 
put it in the proper location.

Speaking of indexing breakage, there is also a bug in the current code 
again, recently i have selfupdate-cvsed, then update-alled a few times 
recently, and its tried to build and old version, dying due to lack of 
patch file (since its not really there anymore..). Only once or twice 
has this happened.

-Ben



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] index behaviour change (post 0.11.x)

[Benjamin Reed]

On Saturday, February 15, 2003, at 08:43 PM, Max Horn wrote:


After suffering from extremely bad fink performance on my iBook in the 
past couple months (despite the nice improvements recently done), I 
begin to wonder if maybe we should reconsider our current package DB 
indexing behavior.

I think it makes sense -- 99% of the users don't need to index any more 
often than when they selfupdate-cvs...



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

[Fink-devel] index behaviour change (post 0.11.x)

[Max Horn]

After suffering from extremely bad fink performance on my iBook in 
the past couple months (despite the nice improvements recently done), 
I begin to wonder if maybe we should reconsider our current package 
DB indexing behavior.

The current approach is restricted in its performance by the fact 
that it traverses the full /sw/fink/dists hierarchy to determine if 
the DB is up-to-date. It takes longest if the DB is actually clean 
(since then it can't cutoff the search early). This alone takes a lot 
of time. On my iBook, with a clean pkg DB, and the HD cache empty, it 
takes a simple "fink info foo" ~8 secs to just load the DB; a second 
call still takes 4 secs; and from our past profiling, a noticeable 
amount is spent on dir traversing.

The motivation for this approach (which causes an automatic reindex 
when a .info file was changed) was/is that we wanted to be fail safe: 
if the user (or developer in our case) messes with an .info file, if 
the index is not updated, this causes unexpected results (namely the 
.info file changes are not honored).

While it's "fail safe", it's *slow* and there is not much room to 
speed it up much as long as we keep traversing the directories (the 
speed of that is determined by the HD, the FS driver in the OS, and 
the OS HD cache, none of which we can control). And with new package 
it gets worse, as we have to stat more and more files.


Hence, I am thinking about changing to this behavior:

1) Stop the auto-indexing-when-info-files change (i.e. don't traverse 
the /sw/fink/dists hierarchy, thus cutting the speed hog)
2) Of course keep the reindexing in selfupdate(-cvs).
3) track the value of the "Trees" setting in fink.conf in the DB, if 
it changes, reindex

This way for the average joe user, nothing should change compared to 
the current setup (please correct me if I am wrong). The people for 
which something changes are the developers. Now we have to remember 
to reindex manually whenever we change a package. In my situation, 
that's often a boon, I hate it when I corrected a typo in a .info 
file, then need to quickly check some other package, type "fink info 
foo", and have to wait ages till it finishes indexing.

To ease the developer's pain somewhat, we could add a "fink index 
foo.info" command, which would add the specified .info file(s) to the 
index. This way, you could just (re)index the files you are working 
on, avoiding having to do full indexing too often.


I know this is quite a reversal from my previous stance on this; and 
I am still not sure if I really like it, but I guess we should at 
least consider it. Comments? Thoughts? What did I miss?


Cheers,

Max


---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] new package manager release

[Max Horn]

At 15:29 Uhr -0500 15.02.2003, David R. Morrison wrote:

There have been no updates to the Fink package manager for two weeks,
other then Ben Hines' adding 10.2.4 as a recognized OS X version this morning.

Max said he wanted to do more things before the next package manager release,
but I would prefer not to wait forever.


Understandable, but I am very low on time. I just did check in Ben 
Hines epoch support. This has been running on his system for a long 
time, and on mine for 2 weeks or so. I performed some tests and it 
worked well in all of them. So there is no need why it should delay 
any release. [Note: there is one important thing missing in this 
regard, the new Epoch field has to be documented!]

I would like to make the release myself. Does Wednesday sound like a 
good day for this?

It would be nice to have Ben's MD5 patch in there as well. I am going 
to look over Justin's CVS-repair patch now, but it's fairly big and 
IMHO should not be rushed out, so I doubt it should be included in 
the upcoming release.


I propose that a few people test my "cctools" patch (which we need to get
in to fink very soon, so that we can update KDE one of these days), that
we add the cctools patch to fink and then release fink-0.11.3 within a
few days.


I have added some comments on the tracker item as to what needs to be 
added to the patch before its commited.

One more thing, if somebody finds the time to verify our mirror 
lists, to see if they are up-to-date, that would be useful. Maybe I 
can finish that half done which script which automates, then I can do 
it myself.



Cheers,

Max


---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Fink Documentation revision -- switch to .cshrc?

[Benjamin Reed]

On Saturday, February 15, 2003, at 07:30 PM, Max Horn wrote:


I do not mind that much if we change to .tcshrc from a techincal point 
of view. What I fear, though, is that we also will have some troubles 
in the transition period. Basically not so advanced users with a 
.cshrc based setup will start reading about .tcshrc. In the worst 
case, they may start using .tcshrc w/o taking their existing stuff in 
.cshrc - suffering!

Hence, if we do this, an absolutly required minimum thing is that we 
add a page or a FAQ item which clearly explains the difference (or 
lack of, except for the reading order) between the two files. It 
should be possible to even explain to a newbie in a few sentences what 
the situation is. FAQ 3.18 is not sufficient for this; rather, add a 
seperate FAQ entry "What is the difference between .tcshrc and .cshrc" 
or so, and maybe link to it from the appropriate places (like 3.18).

Perhaps we should continue to tell them to add it to ~/.cshrc, and then 
say that if ~/.tcshrc exists, add "source ~/.cshrc" to ~/.tcshrc.



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] I want Gimp in English!

[Max Horn]

At 21:15 Uhr -0700 14.02.2003, Ludwin wrote:

My Computer "speaks" french, and the Gimp software I have just 
downloaded speaks mainly English. However, there is french as well. 
I imagine fink can somehow detect that my ibook is in French, so 
that Gimp shows some cammands in French. But this produces 
uncertainty: the documentation is in English, and this language mix 
is confusiong. Is there a way to instruct Gimp to use only English, 
despite the main languge on my ebook is French?


If you use tcsh, try this:

  unsetenv LANG
  gimp

That should launch gimp in english


Max


---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Fink Documentation revision -- switch to .cshrc?

[Max Horn]

I do not mind that much if we change to .tcshrc from a techincal 
point of view. What I fear, though, is that we also will have some 
troubles in the transition period. Basically not so advanced users 
with a .cshrc based setup will start reading about .tcshrc. In the 
worst case, they may start using .tcshrc w/o taking their existing 
stuff in .cshrc - suffering!

Hence, if we do this, an absolutly required minimum thing is that we 
add a page or a FAQ item which clearly explains the difference (or 
lack of, except for the reading order) between the two files. It 
should be possible to even explain to a newbie in a few sentences 
what the situation is. FAQ 3.18 is not sufficient for this; rather, 
add a seperate FAQ entry "What is the difference between .tcshrc and 
.cshrc" or so, and maybe link to it from the appropriate places (like 
3.18).


Cheers,

Max


---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Fink Documentation revision -- switch to .cshrc?

[Martin Costabel]

Neal Parikh wrote:
[]

I wanted to suggest either changing the install instructions to use 
~/.tcshrc or to at least make a note there that if you have a ~/.tcshrc 
file, ~/.cshrc will not be read at all and Fink won't work properly. If 

There is an FAQ entry #3.18 explaining this, added 2 months ago.

I think when this was discussed last time it was found that the ~/.cshrc 
manipulation is mentioned so often on the web pages and elsewhere that 
no one had the courage to change it everywhere.

you're not familiar with the problem, it's that ~/.cshrc is not read by 
tcsh when ~/.tcshrc is present. I haven't figured out any reason to use 
~/.cshrc over ~/.tcshrc, since no one is really using pure csh on Mac OS 
X or Darwin, but I think the issue outlined above is a very strong 
reason to switch to ~/.tcshrc. Over half the problems I help people with 
are related to this.

This is why this question has been discussed several times over the past 
18 months, without much of a result (there was a long discussion on 6/7 
December 2001, for example). I never understood the choice of ~/.cshrc 
either.
[]
The instructions I think most people are using are located at 
. One way to 
fix the problem would simply be to search for .cshrc and replace it with 
.tcshrc globally. 

There are a couple of other places besides the web pages, README, 
INSTALL and USAGE files in the various installers, for example, or 
bootstrap scripts.

Another way would be to add a disclaimer like the
following:

[]
This is far too complicated to be added at all relevant places.


so maybe it's more practical to just put a note like that in the FAQ. 

This exists already, see above, but it doesn't seem to be sufficient.

Another idea would be to offer a little shell script that checks whether 
~/.tcshrc or ~/.cshrc exist and whether "source /sw/bin/init.csh" is 
already there and adds it if and where necessary.

--
Martin




---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

[Fink-devel] Fink Documentation revision -- switch to .cshrc?

[Neal Parikh]

Hi,

I was talking to drm in #fink earlier, and he said I should email the 
list with my suggestion for Fink documentation. I end up helping a lot 
of people install, configure, or use Fink, and in my experience, when 
things go wrong, it is often related to a $PATH error and .cshrc. 
People get 'fink: command not found' errors and it's always due to one 
of two mistakes: (1) They simply didn't follow instructions and add 
'source /sw/bin/init.csh' to their ~/.cshrc file; (2) They did, but 
it's not being read because they have a ~/.tcshrc that's doing other 
things. The latter actually occurs more frequently than the former.

I wanted to suggest either changing the install instructions to use 
~/.tcshrc or to at least make a note there that if you have a ~/.tcshrc 
file, ~/.cshrc will not be read at all and Fink won't work properly. If 
you're not familiar with the problem, it's that ~/.cshrc is not read by 
tcsh when ~/.tcshrc is present. I haven't figured out any reason to use 
~/.cshrc over ~/.tcshrc, since no one is really using pure csh on Mac 
OS X or Darwin, but I think the issue outlined above is a very strong 
reason to switch to ~/.tcshrc. Over half the problems I help people 
with are related to this.

You might not think that a lot of people have ~/.tcshrc files to begin 
with, but a lot do. Sites like Mac OS X Hints will have how-tos in 
which the author will use .tcshrc simply because that's what they 
prefer, and if you follow the hint, hey presto! Fink stops working. 
This can also come up in support given in web forums (e.g. MacNN's Unix 
forum) or simply suggestions people get from friends (e.g. "Hey, this 
is a neat feature, put 'set complete=enhance' in your .tcshrc file"). 
Random other tutorials littered over the internet can use ~/.tcshrc as 
well.

The instructions I think most people are using are located at 
. One way to 
fix the problem would simply be to search for .cshrc and replace it 
with .tcshrc globally. Another way would be to add a disclaimer like 
the following:

NOTE: If you have a .tcshrc already in your home directory, you should 
follow the instructions below but add the lines to .tcshrc and not 
.cshrc. To check if you have a .tcshrc file, run the following > command:

% ls -l ~/.tcshrc

If you get output similar to the line below, you have an existing 
.tcshrc file and should use it for your Fink installation:

-rw-r--r--   1 neal staff4333 Jan 13 22:21 
/Users/neal/.tcshrc

If you do not get any output, proceed with the instructions below as 
written.

I think just telling people to use ~/.tcshrc would be better, but then 
people who are just using .cshrc would have to be informed of the 
issue, so maybe it's more practical to just put a note like that in the 
FAQ. I would be happy to contribute any text you need, if necessary.

Anyway, I've written more than enough for now. Thoughts?

- Neal

--
Neal Parikh <[EMAIL PROTECTED]>
Web - http://www.nparikh.org/
"Only do what only you can do." (Edsger W. Dijkstra)



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

[Fink-devel] new package manager release

[David R. Morrison]

There have been no updates to the Fink package manager for two weeks,
other then Ben Hines' adding 10.2.4 as a recognized OS X version this morning.

Max said he wanted to do more things before the next package manager release,
but I would prefer not to wait forever.

I propose that a few people test my "cctools" patch (which we need to get
in to fink very soon, so that we can update KDE one of these days), that
we add the cctools patch to fink and then release fink-0.11.3 within a
few days.

  -- Dave


---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

[Fink-devel] cctools virtual package

[David R. Morrison]

I've made a patch for Fink which supplies a cctools virtual package:

https://sourceforge.net/tracker/index.php?func=detail&aid=687178&group_id=17203&atid=317203

After some discussion on #fink, we concluded that this package should be
"BuildDependsOnly = True", and as such, we won't need to make dpkg aware
of it, only fink.

I'd like to have some other knowledgeable people test this hypothesis before
I commit the patch to fink, however.

Ben Reed will use this in qt3 and kde, to detect whether folks have upgraded
to the December 2002 tools or not.

  -- Dvae


---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

[Fink-devel] ncurses 5.3

[Chris Zubrzycki]

Ok. I have finally committed ncurses 5.3, and just wanted to let 
everyone know. I have verified that the previously missing symbols are 
in fact there, and tested the libs with old and new ncurses progs, no 
problems.

I also committed versions for 10.1, which i know at least a few people 
use. I have also tested these, (thanks inca  ;-)  )  and unless i made 
a last minute typo, they work also.


If other ppl need things backported, let me know and i'll help.

-chris zubrzycki
- --
PGP public key: http://homepage.mac.com/beren/publickey.txt
ID: 0xA2ABC070
Fingerprint: 26B0 BA6B A409 FA83 42B3  1688 FBF9 8232 A2AB C070


"Sadly, text alone cannot convey the depths of my sarcasm."



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel