Hello all,
I am trying to build UncommonSQL (along with its Postgres backend) for
CMUCL 18e. All are installed as the lastest unstable packages.
I can't compile the system. When I do:
(mk:oos :uncommonsql :compile)
This is the error:
Parse error in namestring: Expecting a file name, got #\..
CL-LIBRARY:UNCOMMONSQL;SQL;.ERR
^
[Condition of type COMMON-LISP::NAMESTRING-PARSE-ERROR]
I have looked at the uncommonsql.system file and at the sources, and
there is no sql/.err dir in there which makes me think this may be the
result of some previous build error which, instead of being handled,
caused ".err" to be appended to the pathname. But I have no idea how
to ascertain the problem point or what to do about it. I tried the CVS
version of UncommonSQL, too, but to no avail.
I am surprised that this is broken wrt rest of Debian's Lisp stuff - I
would've thought that these packages were used heavily enough together
to be maintained and in a working-out-of-box state.
Any fixes, pointers about where/how to look, or other suggestions are
much appreciated. Although I know my way around pretty well I am still
a somewhat newbile lisper, so the less esoteric the better. :)
Thanks in advance,
Jon Steinhart
PS - FWIW, here's a stack trace:
Backtrace:
0: (COMMON-LISP::EXPECTING "a file name" ((#\. . 27) ("ERR" . 28)))
1: (COMMON-LISP::PARSE-DIRECTORY ((#\. . 27) ("ERR" . 28)))
2: (COMMON-LISP::PARSE-LOGICAL-NAMESTRING "CL-LIBRARY:UNCOMMONSQL;SQL;.ERR" 0 31)
3: (COMMON-LISP::%PARSE-NAMESTRING "CL-LIBRARY:UNCOMMONSQL;SQL;.ERR" NIL #p"" 0 ...)
4: (COMMON-LISP::%PARSE-NAMESTRING 6 "CL-LIBRARY:UNCOMMONSQL;SQL;.ERR" NIL #p"" ...)[:EXTERNAL]
5: (PATHNAME-DIRECTORY "CL-LIBRARY:UNCOMMONSQL;SQL;.ERR" :CASE :LOCAL)
6: (MAKE::COMPONENT-FULL-PATHNAME-I #<FILE: package> :ERROR NIL)
Locals:
MAKE::COMPONENT = #<FILE: package>
MAKE::DEV = #:NOT-AVAILABLE
PATHNAME = "CL-LIBRARY:UNCOMMONSQL;SQL;.ERR"
TYPE = :ERROR
MAKE::VERSION = NIL
MAKE::VERSION-DIR = #:NOT-AVAILABLE
MAKE::VERSION-REPLACE = #:NOT-AVAILABLE
[No catch-tags]
7: (MAKE::COMPILE-FILE-OPERATION #<FILE: package> :NEW-SOURCE-AND-DEPENDENTS)
Locals:
MAKE::COMPONENT = #<FILE: package>
MAKE::FORCE = :NEW-SOURCE-AND-DEPENDENTS
MAKE::MUST-COMPILE = T
MAKE::OUTPUT-FILE = #:NOT-AVAILABLE
MAKE::SOURCE-PNAME = #:NOT-AVAILABLE
[No catch-tags]
8: (MAKE::COMPILE-AND-LOAD-OPERATION #<FILE: package> :NEW-SOURCE-AND-DEPENDENTS)
On Thu, 08 Apr 2004 00:48:29 GMT, Jonathan Steinhart <···@pennbitworks.spammityspam.com> wrote:
> I am trying to build UncommonSQL (along with its Postgres backend)
> for CMUCL 18e. All are installed as the lastest unstable packages.
UncommonSQL is no longer maintained. It has been merged into
CLSQL. See
<http://www.google.com/groups?selm=slrnc78vi5.99b.kevin%40tiger.med-info.com>
<http://clsql.b9.com/>
Edi.
> On Thu, 08 Apr 2004 00:48:29 GMT, Jonathan Steinhart <···@pennbitworks.spammityspam.com> wrote:
>
>> I am trying to build UncommonSQL (along with its Postgres backend)
>> for CMUCL 18e. All are installed as the lastest unstable packages.
The cl-uncommonsql package was removed from the Debian archive at the
request of its maintainer. I've uploaded a new version of CLSQL to
Debian which should get accepted in the next week which includes
a port of the UncommonSQL code.
CLSQL's objected-oriented interface supports CMUCL, SBCL, OpenMCL,
Allegro, and Lispworks.
--
Kevin Rosenberg
·····@rosenberg.net
"Kevin M. Rosenberg" <·····@news.swcp.com> writes:
> The cl-uncommonsql package was removed from the Debian archive at the
> request of its maintainer. I've uploaded a new version of CLSQL to
> Debian which should get accepted in the next week which includes
> a port of the UncommonSQL code.
Beauteous. A great package gets even better :)
Thanks for both replies. If only I had caught up on cll that day
before posting, well... then I wouldn't have had to post. And all
those precious bits would not have been squandered.
Jonathan Steinhart <···@pennbitworks.spammityspam.com> writes:
> "Kevin M. Rosenberg" <·····@news.swcp.com> writes:
> > The cl-uncommonsql package was removed from the Debian archive at the
> > request of its maintainer. I've uploaded a new version of CLSQL to
> > Debian which should get accepted in the next week which includes
> > a port of the UncommonSQL code.
>
> Beauteous. A great package gets even better :)
>
> Thanks for both replies. If only I had caught up on cll that day
> before posting, well... then I wouldn't have had to post. And all
> those precious bits would not have been squandered.
Think of it this way. Some piece of spam took an extra half second to
get to where it was going. Isn't that a good thing?
Where in PA are you? I'm in Delaware County, badly served by SEPTA,
not far from Philadelphia.
--
It would not be too unfair to any language to refer to Java as a
stripped down Lisp or Smalltalk with a C syntax.
--- Ken Anderson
http://openmap.bbn.com/~kanderso/performance/java/index.html
David Steuber <·····@david-steuber.com> writes:
> Jonathan Steinhart <···@pennbitworks.spammityspam.com> writes:
>>
>> Thanks for both replies. If only I had caught up on cll that day
>> before posting, well... then I wouldn't have had to post. And all
>> those precious bits would not have been squandered.
>
> Think of it this way. Some piece of spam took an extra half second to
> get to where it was going. Isn't that a good thing?
This isn't funny - some poor soul out there *needed* that generic
viagra ASAP, in order to refinance 1000's of XXX sites at the lowest
rates ever. All from the privacy of his own home, of course. Not that
I'm that guy or anything. (But if you have an extra copy of that one,
please forward it to me... for my records).
> Where in PA are you? I'm in Delaware County, badly served by SEPTA,
> not far from Philadelphia.
I'm in Pittsburgh - that glorious land of former industrial might;
present-day home of the nation's highest parking tax as well as the
nation's furthest-travelling autonomous vehicle - coincidence? Was the
Red Team was just looking for a free ride? Methinks they should just
admit they were only in it for the commute, and rename themselves The
Abelian Group.
I've noticed a few other PA people on here - great to know that the
light has reached even these dark places. Jealous of the lispNYC tales
I've read here, I tried (and failed utterly) to (re?)vitalize the
Pittsburgh Scheme/Lisp meetup last year (http://scheme.meetup.com/ -
if you're reading this from da Burgh, you should sign up!). Is there
anything like this in Philly? Or are you close enough to make it to
NYC if you'd ever care to drop in on their monthly fete?
We'll have to hold a PA lispers summit in the dead zone directly
between us (otherwise known as Harrisburg). I'm sure the local Amish
folk would share in the joy of sexprs.
Jonathan Steinhart <···@pennbitworks.spammityspam.com> writes:
> David Steuber <·····@david-steuber.com> writes:
>
> > Jonathan Steinhart <···@pennbitworks.spammityspam.com> writes:
> >>
> >> Thanks for both replies. If only I had caught up on cll that day
> >> before posting, well... then I wouldn't have had to post. And all
> >> those precious bits would not have been squandered.
> >
> > Think of it this way. Some piece of spam took an extra half second to
> > get to where it was going. Isn't that a good thing?
>
> This isn't funny - some poor soul out there *needed* that generic
> viagra ASAP, in order to refinance 1000's of XXX sites at the lowest
> rates ever. All from the privacy of his own home, of course. Not that
> I'm that guy or anything. (But if you have an extra copy of that one,
> please forward it to me... for my records).
I used to auto-archive my spam to my webserver. I had a nice hall of
shame page filled with spam along with the trace information and
whatever contact info I could get with automated methods. That became
too much to maintain.
> > Where in PA are you? I'm in Delaware County, badly served by SEPTA,
> > not far from Philadelphia.
>
> I'm in Pittsburgh - that glorious land of former industrial might;
> present-day home of the nation's highest parking tax as well as the
> nation's furthest-travelling autonomous vehicle - coincidence? Was the
> Red Team was just looking for a free ride? Methinks they should just
> admit they were only in it for the commute, and rename themselves The
> Abelian Group.
CMU is in Pittsburgh. Do you have any affiliation with that place? I
would think that CMU and UPENN would have a nice rivalry going if
nerds were just a little more popular.
> I've noticed a few other PA people on here - great to know that the
> light has reached even these dark places. Jealous of the lispNYC tales
> I've read here, I tried (and failed utterly) to (re?)vitalize the
> Pittsburgh Scheme/Lisp meetup last year (http://scheme.meetup.com/ -
> if you're reading this from da Burgh, you should sign up!). Is there
> anything like this in Philly? Or are you close enough to make it to
> NYC if you'd ever care to drop in on their monthly fete?
There is a Perl mongers group. I haven't looked for or found a Lisp
group. I really hate to trek into that city anyway. As for going to
NYC, that means going through New Jersey. With no due respect to the
people who live in that city and state, it ain't worth it! And having
to pass through Newark doesn't make things any better.
> We'll have to hold a PA lispers summit in the dead zone directly
> between us (otherwise known as Harrisburg). I'm sure the local Amish
> folk would share in the joy of sexprs.
I hear the Amish are really into Scheme. They fire up their Honda
generators to work their bare bones PCs with Linux to run MIT Scheme.
Naturally they forgo X Windows.
My sister whent to college in Elizebeth town. Sometimes the wind
would blow in from over the Hershey factory, over the cow pasturs,
mixing the fine smell of chocolate and cow dung. Now there is some
good air.
--
I wouldn't mind the rat race so much if it wasn't for all the damn cats.
David Steuber <·····@david-steuber.com> writes:
> CMU is in Pittsburgh. Do you have any affiliation with that place? I
> would think that CMU and UPENN would have a nice rivalry going if
> nerds were just a little more popular.
Hehe... yes, my affiliation is that I tried to get a job there as a
research programmer last year and failed. I think they may have scared
of my lack of paper credentials. Nevermind the fact that I've been
programming since I was seven (well, if you cound Basic and Logo - and
I know everyone here counts Logo :-P) and had my first AI programming
job as a junior in high school (10 years ago)... oh well. So, since
there's absolutely nothing else I want to be doing with my life, I
have decided school is the shortest path to an enjoyable & sustainable
future for me. Now I just have to figure out what I'm gonna do until
Fall '05... I think I might rather work as a stockboy than continue
with the paltry contracts I've been able to get of late (if I write
one more web app, my skull will implode).
> I hear the Amish are really into Scheme. They fire up their Honda
> generators to work their bare bones PCs with Linux to run MIT Scheme.
> Naturally they forgo X Windows.
The really fundamentalist ones use Analytical Engines.
Jonathan Steinhart <···@pennbitworks.spammityspam.com> writes:
> I think I might rather work as a stockboy than continue with the
> paltry contracts I've been able to get of late (if I write one more
> web app, my skull will implode).
Ah web apps! I blame them for the atrophy of my brain. So what am I
doing now? I'm looking to assemble some asdf installable packages to
build a Lisp based framework like Perl::CGI + Perl::DBI + misc for
delivering applications over the web. Very weird.
And of course everyone else has already done it better than I ever
will. I shall steal from them. Mwahhahahaha!
In the meantime, I have gotten the feeling that I am basicly striking
keys on my keyboard at random. Hell, that might be more productive.
--
I wouldn't mind the rat race so much if it wasn't for all the damn cats.
Apart from the problems with the actual maintainance status of
UncommonSQL, it looks like this may be a problem with your version of
MK:DEFSYSTEM. The latest CVS version from the CLOCC contains one patch
over the latest release that seem relevant to the error you observe.
I will make a new file release ASAIC.
Cheers
Marco
Jonathan Steinhart wrote:
> Hello all,
>
> I am trying to build UncommonSQL (along with its Postgres backend) for
> CMUCL 18e. All are installed as the lastest unstable packages.
>
> I can't compile the system. When I do:
>
> (mk:oos :uncommonsql :compile)
>
>
> This is the error:
>
> Parse error in namestring: Expecting a file name, got #\..
> CL-LIBRARY:UNCOMMONSQL;SQL;.ERR
> ^
> [Condition of type COMMON-LISP::NAMESTRING-PARSE-ERROR]
>
>
> I have looked at the uncommonsql.system file and at the sources, and
> there is no sql/.err dir in there which makes me think this may be the
> result of some previous build error which, instead of being handled,
> caused ".err" to be appended to the pathname. But I have no idea how
> to ascertain the problem point or what to do about it. I tried the CVS
> version of UncommonSQL, too, but to no avail.
>
> I am surprised that this is broken wrt rest of Debian's Lisp stuff - I
> would've thought that these packages were used heavily enough together
> to be maintained and in a working-out-of-box state.
>
> Any fixes, pointers about where/how to look, or other suggestions are
> much appreciated. Although I know my way around pretty well I am still
> a somewhat newbile lisper, so the less esoteric the better. :)
>
> Thanks in advance,
>
> Jon Steinhart
>
>
> PS - FWIW, here's a stack trace:
>
> Backtrace:
> 0: (COMMON-LISP::EXPECTING "a file name" ((#\. . 27) ("ERR" . 28)))
> 1: (COMMON-LISP::PARSE-DIRECTORY ((#\. . 27) ("ERR" . 28)))
> 2: (COMMON-LISP::PARSE-LOGICAL-NAMESTRING "CL-LIBRARY:UNCOMMONSQL;SQL;.ERR" 0 31)
> 3: (COMMON-LISP::%PARSE-NAMESTRING "CL-LIBRARY:UNCOMMONSQL;SQL;.ERR" NIL #p"" 0 ...)
> 4: (COMMON-LISP::%PARSE-NAMESTRING 6 "CL-LIBRARY:UNCOMMONSQL;SQL;.ERR" NIL #p"" ...)[:EXTERNAL]
> 5: (PATHNAME-DIRECTORY "CL-LIBRARY:UNCOMMONSQL;SQL;.ERR" :CASE :LOCAL)
> 6: (MAKE::COMPONENT-FULL-PATHNAME-I #<FILE: package> :ERROR NIL)
> Locals:
> MAKE::COMPONENT = #<FILE: package>
> MAKE::DEV = #:NOT-AVAILABLE
> PATHNAME = "CL-LIBRARY:UNCOMMONSQL;SQL;.ERR"
> TYPE = :ERROR
> MAKE::VERSION = NIL
> MAKE::VERSION-DIR = #:NOT-AVAILABLE
> MAKE::VERSION-REPLACE = #:NOT-AVAILABLE
> [No catch-tags]
> 7: (MAKE::COMPILE-FILE-OPERATION #<FILE: package> :NEW-SOURCE-AND-DEPENDENTS)
> Locals:
> MAKE::COMPONENT = #<FILE: package>
> MAKE::FORCE = :NEW-SOURCE-AND-DEPENDENTS
> MAKE::MUST-COMPILE = T
> MAKE::OUTPUT-FILE = #:NOT-AVAILABLE
> MAKE::SOURCE-PNAME = #:NOT-AVAILABLE
> [No catch-tags]
> 8: (MAKE::COMPILE-AND-LOAD-OPERATION #<FILE: package> :NEW-SOURCE-AND-DEPENDENTS)
>
>
Marco Antoniotti <·······@cs.nyu.edu> writes:
> Apart from the problems with the actual maintainance status of
> UncommonSQL, it looks like this may be a problem with your version of
> MK:DEFSYSTEM. The latest CVS version from the CLOCC contains one
> patch over the latest release that seem relevant to the error you
> observe.
For kicks (mostly to test my own meager understanding of things) I
grabbed today's CVS snapshot of defsystem from CLOCC and gave this a
try. I fired up my vanilla CMUCL image in SLIME, went into the MAKE
package, uninterned everything, loaded the CVS defsystem.lisp without
any warning (save an "Old style in-package" message, which sounds
harmless), reloaded uncommonsql.system and odcl.system, and tried to
compile UncommonSQL. No go - same error.
I *only* mention this in providing for the improbable case where what
I did actually should have worked, given that all bugs are presumed
fixed (and this all wasn't just a brain-dead rookie bumble). In said
case, it would seem that there's still something awry in there.
I would remove the cl-defsystem3 package from my system to be
absolutely sure everything's current, but the gazillions of
dependencies on it make that not worth it for this little
experiment. If you should suspect that there is still a problem
though, I will do what I can to help you track it down. Otherwise,
feel free to ignore me. :)