Sunday, March 13, 2005, 18:19

Dear [EMAIL PROTECTED],

Regarding your message dated 03/03/05 15:21,
subject "plucker-list digest, Vol 1 #971 - 13 msgs",

plrro>
plrro> How about adding an option to plucker-build for filtering each
plrro> downloaded file through tidy?

Just for the record, simple Windows users can also use pdaConverter
and/or Proxomitron to filter unwanted HTML code.

pdaConverter (www.jakewalk.de)
Proxomitron (http://www.proxomitron.info/)

Regards,

Michael A. Lees
[EMAIL PROTECTED]

-------------------
Random quote:
Religious truth is captive in a small number of little manuscripts
which guard the common treasures, instead of expanding them. Let us
break the seal which binds these holy things; let us give wings to
truth that it may fly with the Word, no longer prepared at vast expense,
but multitudes everlastingly by a machine which never wearies to every
soul which enters life.
-Johannes Gutenberg


---8<---Your original message starts here (or part of it)--------
plrro> Send plucker-list mailing list submissions to
plrro>  [email protected]

plrro> To subscribe or unsubscribe via the World Wide Web, visit
plrro>  http://lists.rubberchicken.org/mailman/listinfo/plucker-list
plrro> or, via email, send a message with subject or body 'help' to
plrro>  [EMAIL PROTECTED]

plrro> You can reach the person managing the list at
plrro>  [EMAIL PROTECTED]

plrro> When replying, please edit your Subject line so it is more specific
plrro> than "Re: Contents of plucker-list digest..."


plrro> Today's Topics:

plrro>    1. Re: dataMgr.c, Line:11231, Index out of range (t5) (Kornelis 
Sietsma)
plrro>    2. Re: dataMgr.c, Line:11231, Index out of range (t5) (David A. 
Desrosiers)
plrro>    3. What is /required/ for an install (Faux_Pseudo)
plrro>    4. Re: dataMgr.c, Line:11231, Index out of range (t5) (Kornelis 
Sietsma)
plrro>    5. Re: dataMgr.c, Line:11231, Index out of range (t5) (David A. 
Desrosiers)
plrro>    6. plucker-build and tidy (Justus Piater)
plrro>    7. Re: plucker-build and tidy (David A. Desrosiers)
plrro>    8. Re: plucker-build and tidy (Justus Piater)
plrro>    9. Re: plucker-build and tidy (David A. Desrosiers)
plrro>   10. Re: dataMgr.c, Line:11231, Index out of range (t5) (Michael 
Nordstrom)
plrro>   11. Re: dataMgr.c, Line:11231, Index out of range (t5) (Jim Anderson)
plrro>   12. IR Beaming a plucker document? (Dan Stromberg)
plrro>   13. Re: What is /required/ for an install (Michael Nordstrom)

plrro> --__--__--

plrro> Message: 1
plrro> Date: Thu, 03 Mar 2005 13:11:11 +1100
plrro> From: Kornelis Sietsma <[EMAIL PROTECTED]>
plrro> To: [email protected]
plrro> Subject: Re: dataMgr.c, Line:11231, Index out of range (t5)
plrro> Reply-To: [email protected]

plrro> David A. Desrosiers wrote:
>>>Someone else reported this problem to the list in October last year,
>>>but there doesn't seem to have been a response - has anyone else 
>>>seen this?  Any workarounds/fixes?
>> 
>>      Does running CleanPlkr help? You can find it on the "Snapshot" 
>> page of the main Plucker website. 

plrro> Yes and no.  Yes, it repairs Plucker so I can load other documents*, but
plrro> no, it doesn't allow me to view the actual document in question.

plrro> If I get time I'll clean up the file that is causing problems and post
plrro> it to a web site somewhere, but I'm swamped with work at the moment so
plrro> this may take time.  I was mainly hoping someone else had solved this
plrro> already :)

plrro> - Korny
plrro> * I actually already found a workaround here - rename the directory
plrro> containing my docs to "/palm/ebooksx", run plucker (which resets it's
plrro> file list), then rename the directory back.  Some things are much easier
plrro> with the t5 and usb-drive-mode...

plrro> --
plrro> Kornelis Sietsma  [EMAIL PROTECTED]

plrro> --__--__--

plrro> Message: 2
plrro> Date: Wed, 2 Mar 2005 21:24:23 -0500 (EST)
plrro> From: "David A. Desrosiers" <[EMAIL PROTECTED]>
plrro> To: Plucker General List <[email protected]>
plrro> Subject: Re: dataMgr.c, Line:11231, Index out of range (t5)
plrro> Reply-To: [email protected]


>> Yes and no.  Yes, it repairs Plucker so I can load other documents*,
>> but no, it doesn't allow me to view the actual document in question.

plrro>  Are you sure the document does not contain corrupted records? 


plrro> David A. Desrosiers
plrro> [EMAIL PROTECTED]
plrro> http://gnu-designs.com

plrro> --__--__--

plrro> Message: 3
plrro> Date: Thu, 3 Mar 2005 00:23:17 -0500
plrro> From: Faux_Pseudo <[EMAIL PROTECTED]>
plrro> To: plucker-list <[email protected]>
plrro> Subject: What is /required/ for an install
plrro> Reply-To: [email protected]

plrro> There was a comment here a while ago mentioning about copying some of
plrro> the *.py files into your path and calling it a day.  But what files
plrro> does one copy?  Do any of them need to be renamed?  I don't need the
plrro> desktop.  Just the spider and converter and the resulting pdb file(s).
plrro> Command line only is just fine. 

plrro> I posted a C++ bug here a few months ago.  No joy.
plrro> I have lurked hoping to find a solution.  No joy.
plrro> I have googled to no end.  I post this here in an attempt to get some
plrro> issues fixed and to make a record of what had to be done in order to
plrro> get it working so that it may help others.

plrro> cd plucker-1.8

plrro> sudo ./configure --disable-desktopbuild
plrro> --disable-gtkviewer --disable-palmosbuild

plrro> Version one:

plrro> make ; sudo make install

plrro> make install bombs because 
plrro> cp AUTHORS BUGREPORT COPYING CREDITS ChangeLog FAQ NEWS
plrro> README REQUIREMENTS TODO /usr/local/share/plucker/doc
plrro> cp: cannot stat `TODO': No such file or directory
plrro> make: *** [install-data] Error 1

plrro> touch TODO
plrro> and sudo make install again, no errors.  
plrro> The installer should not choke on a missing file that is not manditory
plrro> for the running of the app.


plrro> ls /usr/local/bin/pluck*
plrro> lrwxr-xr-x  1 root root     36 2005-03-02 23:23
plrro> /usr/local/bin/pluck-comics ->
plrro> /usr/local/PyPlucker/pluck-comics.py
plrro> -rwxr-xr-x  1 root root 116093 2004-11-08 07:49 /usr/local/bin/plucker*
plrro> lrwxr-xr-x  1 root root     30 2005-03-02 23:23
plrro> /usr/local/bin/plucker-build -> /usr/local/PyPlucker/Spider.py
plrro> lrwxr-xr-x  1 root root     35 2005-03-02 23:23
plrro> /usr/local/bin/plucker-decode ->
plrro> /usr/local/PyPlucker/PluckerDocs.py
plrro> lrwxr-xr-x  1 root root     30 2005-03-02 23:23
plrro> /usr/local/bin/plucker-dump -> /usr/local/PyPlucker/Decode.py
plrro> -rwxr-xr-x  1 root root   5223 2005-03-02 23:56
plrro> /usr/local/bin/plucker-setup*

plrro> Each of those links is a dead link. The actual locations should point
plrro> to files in /usr/lib/python2.3/site-packages/PyPlucker/ 
plrro> So I correct that.  

plrro> Then I can run
plrro> plucker-build -c  -H ~/.plucker/home.html
plrro> Which results in good news and then bad:
plrro> Wrote 374 <= plucker:/~special~/links1
plrro> Wrote 375 <= plucker:/~special~/links201
plrro> Done!
plrro> Traceback (most recent call last):
plrro>   File "/usr/local/bin/plucker-build", line 1768, in ?
plrro>     sys.exit(realmain(None))
plrro>   File "/usr/local/bin/plucker-build", line 1762, in realmain
plrro>     os.remove(os.path.join(tempfile.tempdir, file))
plrro> OSError: [Errno 1] Operation not permitted: '/tmp/tmp.xpi'

plrro> So then we try and continue 
plrro> plucker-build -f file

plrro> Which results in:
plrro> exactly the same error as above

plrro> But this does work.  It results in a readable file.  But should anyone
plrro> have to go through this much manual work in the make process?  What
plrro> are these errors for if they do not prevent it from working?

plrro> Version two:  this one is mostly a rant because using the 
plrro> install-plucker in the unix dir shouldn't have so many issues.

plrro> cd unix
plrro> running install-plucker as a non-root user is not an option because
plrro> some files like
plrro> /home/faux/plucker-1.8/viewer-GTK+2-POSIX will get
plrro> permissioned as root/root and you will get permission errors.  Same
plrro> reason the configure command above is done sudo

plrro> running it as sudo results in:

plrro> gmake: *** [palmos_client] Error 2
plrro> Traceback (most recent call last):
plrro>   File "./install-plucker", line 140, in ?
plrro>     os.symlink(os.path.join(MOD_DIR, "PyPlucker",
plrro> "Spider.py"), os.path.join(BIN_DIR, "plucker-build"))
plrro> OSError: [Errno 17] File exists

plrro> commenting out the following lines in install-plucker helps remove
plrro> that and other errors that result if one doesn't comment them out:

plrro>     if os.path.exists(os.path.join(BIN_DIR, "plucker-build")):
plrro>         os.remove(os.path.join(BIN_DIR, "plucker-build"))
plrro>     os.symlink(os.path.join(MOD_DIR, "PyPlucker",
plrro> "Spider.py"), os.path.join(BIN_DIR, "plucker-build"))

plrro>     if os.path.exists(os.path.join(BIN_DIR, "plucker-decode")):
plrro>         os.remove(os.path.join(BIN_DIR, "plucker-decode"))
plrro>     os.symlink(os.path.join(MOD_DIR, "PyPlucker",
plrro> "PluckerDocs.py"), os.path.join(BIN_DIR, "plucker-decode"))

plrro>     if os.path.exists(os.path.join(BIN_DIR, "plucker-dump")):
plrro>         os.remove(os.path.join(BIN_DIR, "plucker-dump"))
plrro>     os.symlink(os.path.join(MOD_DIR, "PyPlucker",
plrro> "Decode.py"), os.path.join(BIN_DIR, "plucker-dump"))

plrro> So we run the install command again: 
plrro> gmake[1]: Entering directory
plrro> `/home/faux/plucker-1.8/viewer-GTK+2-POSIX'
plrro> gmake[1]: Nothing to be done for `all'.
plrro> gmake[1]: Leaving directory
plrro> `/home/faux/plucker-1.8/viewer-GTK+2-POSIX'
plrro> gmake: *** No rule to make target `install-prc-files'.  Stop.

plrro> Why is it going into viewer-GTK+2-POSIX ?  Didn't I tell it to not
plrro> build gtkviewer with the --disable-gtkviewer option?  

plrro> So I add a "install-prc-files: : " section in
plrro> /home/faux/plucker-1.8/viewer-GTK+2-POSIX/Makefile just to get past
plrro> it.  I get the same error again.  Why?  Because
plrro> /home/faux/plucker-1.8/viewer-GTK+2-POSIX/Makefile was overwritten by
plrro> some portion of theinstall-plucker script.


plrro> -- 
plrro>   .-'))       fauxascii.com        ('-.  | It's a damn poor mind that
plrro>  ' ..-  .:" )               ( ":.  -.. ' | can only think of one way to
plrro>   ((,,_;'.;'   UIN=66618055  ';. ';_,,)) | spell a word.
plrro>         ((_.YIM=Faux_Pseudo :._))        | - Andrew Jackson

plrro> --__--__--

plrro> Message: 4
plrro> Date: Thu, 03 Mar 2005 16:30:47 +1100
plrro> From: Kornelis Sietsma <[EMAIL PROTECTED]>
plrro> To: [email protected]
plrro> Subject: Re: dataMgr.c, Line:11231, Index out of range (t5)
plrro> Reply-To: [email protected]

plrro> David A. Desrosiers wrote:
>>>Yes and no.  Yes, it repairs Plucker so I can load other documents*,
>>>but no, it doesn't allow me to view the actual document in question.
>> 
>>      Are you sure the document does not contain corrupted records? 

plrro> After some digging, I can clearly say: I don't know :)

plrro> If I re-convert the file (using jpluck) I get a different binary file
plrro> each time (must be datestamps or something).  But they all crash plucker
plrro> in one way or another.
plrro> *Unless* I rename the document to a short simple name first (it has a 48
plrro> character file name with spaces and dashes in it) and then convert, in
plrro> which case the document loads fine.

plrro> But it's not just the file name - if I create another document of
plrro> similar length with the same (long) file name, it also works fine.  As
plrro> do many other documents with long file names.

plrro> Sadly the original document is an internal company document so I can't
plrro> post it.  If I change it in any significant way, such as replacing all
plrro> text in the document with 'x' characters, the resulting document
plrro> converts and reads with no problems.

plrro> So I have no idea what the problem is, I suspect it is probably JPluck,
plrro> but can't be sure.  Certainly plucker, when it crashes, crashes really
plrro> impressively - not even CleanPlkr has fixed all my crash problems,
plrro> renaming the /Palm/Ebook folder has been my only recovery path more than
plrro> once...

plrro> Anyway, when I get a document that exhibits this behaviour that I *can*
plrro> upload somewhere, I'll do so, and then maybe we can progress further.

plrro> - Korny

plrro> --__--__--

plrro> Message: 5
plrro> Date: Thu, 3 Mar 2005 00:41:31 -0500 (EST)
plrro> From: "David A. Desrosiers" <[EMAIL PROTECTED]>
plrro> To: Plucker General List <[email protected]>
plrro> Subject: Re: dataMgr.c, Line:11231, Index out of range (t5)
plrro> Reply-To: [email protected]


>> Anyway, when I get a document that exhibits this behaviour that I 
>> *can* upload somewhere, I'll do so, and then maybe we can progress 
>> further.

plrro>  How about using the Plucker distiller, instead of JPluck?
 

plrro> David A. Desrosiers
plrro> [EMAIL PROTECTED]
plrro> http://gnu-designs.com

plrro> --__--__--

plrro> Message: 6
plrro> To: [email protected]
plrro> Subject: plucker-build and tidy
plrro> From: Justus Piater <[EMAIL PROTECTED]>
plrro> Date: Thu, 03 Mar 2005 10:06:39 +0100
plrro> Reply-To: [email protected]

plrro> Hi,

plrro> The issue of Web pages whose HTML is fouled up to the point of
plrro> impluckability (add this to Merriam-Webster!) comes up over and over
plrro> again.

plrro> The standard solution would be to use wget with the right options to
plrro> download all that's needed, then run tidy on the file(s) in question,
plrro> and then pluck the local files.  This is quite cumbersome, and one
plrro> loses the original URL in the plucked PDB.

plrro> How about adding an option to plucker-build for filtering each
plrro> downloaded file through tidy?

plrro> This should only be a minor hack, the tidying occurs in the right
plrro> place in the pipeline, and it increases plucker-build's practical
plrro> usability without placing additional burden on the user.

plrro> Justus

plrro> --=20
plrro> Justus H. Piater, Ph.D.        
plrro> http://www.montefiore.ulg.ac.be/~piater/
plrro> Institut Montefiore, B28        Phone: +32-4-366-2279
plrro> Universit=E9 de Li=E8ge, Belgium    Fax:   +32-4-366-2620


plrro> --__--__--

plrro> Message: 7
plrro> Date: Thu, 3 Mar 2005 08:47:43 -0500 (EST)
plrro> From: "David A. Desrosiers" <[EMAIL PROTECTED]>
plrro> To: Plucker General List <[email protected]>
plrro> Subject: Re: plucker-build and tidy
plrro> Reply-To: [email protected]


>> How about adding an option to plucker-build for filtering each 
>> downloaded file through tidy?

plrro>  You mean like using the --filter option of plucker-build? 

plrro>     --filter=<filter name>:
plrro>           Pass fetched documents through filter prior to parsing.


plrro> David A. Desrosiers
plrro> [EMAIL PROTECTED]
plrro> http://gnu-designs.com

plrro> --__--__--

plrro> Message: 8
plrro> To: [email protected]
plrro> Subject: Re: plucker-build and tidy
plrro> From: Justus Piater <[EMAIL PROTECTED]>
plrro> Date: Thu, 03 Mar 2005 17:44:48 +0100
plrro> Reply-To: [email protected]

plrro> "David A. Desrosiers" <[EMAIL PROTECTED]> wrote on Thu, 3 Mar
plrro> 2005 08:47:43 -0500 (EST):

>>> How about adding an option to plucker-build for filtering each=20
>>> downloaded file through tidy?
>>
>>      You mean like using the --filter option of plucker-build?=20
>>
>>     --filter=3D<filter name>:
>>           Pass fetched documents through filter prior to parsing.

plrro> Doh! Yes.

plrro> Let me shake off a small fraction of my shame by noting that neither
plrro> the man page delivered with plucker 1.8, nor the HTML manual pages
plrro> contain any trace of this option (rpm from plkr.org).

plrro> Justus

plrro> --=20
plrro> Justus H. Piater, Ph.D.        
plrro> http://www.montefiore.ulg.ac.be/~piater/
plrro> Institut Montefiore, B28        Phone: +32-4-366-2279
plrro> Universit=E9 de Li=E8ge, Belgium    Fax:   +32-4-366-2620


plrro> --__--__--

plrro> Message: 9
plrro> Date: Thu, 3 Mar 2005 11:56:19 -0500 (EST)
plrro> From: "David A. Desrosiers" <[EMAIL PROTECTED]>
plrro> To: Plucker General List <[email protected]>
plrro> Subject: Re: plucker-build and tidy
plrro> Reply-To: [email protected]


>> Let me shake off a small fraction of my shame by noting that neither
>> the man page delivered with plucker 1.8, nor the HTML manual pages 
>> contain any trace of this option (rpm from plkr.org).

plrro>  Care to help us fix tha discrepancy? We need people like 
plrro> yourself to go through the parts of Plucker to find these kinds of
plrro> missing bits, so we can add/fix them before the next release. 


plrro> David A. Desrosiers
plrro> [EMAIL PROTECTED]
plrro> http://gnu-designs.com

plrro> --__--__--

plrro> Message: 10
plrro> Date: Thu, 3 Mar 2005 18:14:32 +0100
plrro> From: Michael Nordstrom <[EMAIL PROTECTED]>
plrro> To: [email protected]
plrro> Subject: Re: dataMgr.c, Line:11231, Index out of range (t5)
plrro> Reply-To: [email protected]

plrro> On Thu, Mar 03, 2005, Kornelis Sietsma wrote:
>> *Unless* I rename the document to a short simple name first (it has a 48
>> character file name with spaces and dashes in it) and then convert, in
>> which case the document loads fine.

plrro> If jpluck doesn't make sure that the name is at most 32 bytes
plrro> (including the terminating NULL character) then you could run 
plrro> into problems if you try to use a 48 character name...

plrro> /Mike


plrro> --__--__--

plrro> Message: 11
plrro> From: Jim Anderson <[EMAIL PROTECTED]>
plrro> To: <[email protected]>
plrro> Date: Thu, 3 Mar 2005 11:28:31 -0600
plrro> Subject: Re: dataMgr.c, Line:11231, Index out of range (t5)
plrro> Reply-To: [email protected]

plrro> So I decided to take a look at line 11231 of dataMgr.c to see what it 
was=
plrro>  doing, and get a hint of what might be the problem.  I downloaded the=
plrro>  latest source snapshop, unpacked it, and started looking for dataMgr.c. 
 I=
plrro>  can't find it?  Anybody know where the source to dataMgr.c might 
reside,=
plrro>  and if we are allowed to look at it?




plrro> --__--__--

plrro> Message: 12
plrro> Subject: IR Beaming a plucker document?
plrro> From: Dan Stromberg <[EMAIL PROTECTED]>
plrro> To: [email protected]
plrro> Cc: [EMAIL PROTECTED]
plrro> Date: Thu, 03 Mar 2005 10:18:49 -0800
plrro> Reply-To: [email protected]


plrro> Has anyone run across a way of beaming a plucker document to another
plrro> palm over infrared?

plrro> Thanks!



plrro> --__--__--

plrro> Message: 13
plrro> Date: Thu, 3 Mar 2005 19:21:03 +0100
plrro> From: Michael Nordstrom <[EMAIL PROTECTED]>
plrro> To: [email protected]
plrro> Subject: Re: What is /required/ for an install
plrro> Reply-To: [email protected]

plrro> On Thu, Mar 03, 2005, Faux_Pseudo wrote:

>> make install bombs because 
>> cp AUTHORS BUGREPORT COPYING CREDITS ChangeLog FAQ NEWS README
>> REQUIREMENTS TODO /usr/local/share/plucker/doc
>> cp: cannot stat `TODO': No such file or directory
>> make: *** [install-data] Error 1

plrro> This is my fault; I removed the TODO file from the source package,
plrro> but forgot to remove it from the Makefile. Still, most users that
plrro> "build" Plucker on their own would probably use the snapshot to
plrro> get a more recent package than what is included in 1.8. In that
plrro> case you wouldn't run into this problem.

>> Each of those links is a dead link. The actual locations should point
>> to files in /usr/lib/python2.3/site-packages/PyPlucker/ 

plrro> 'make install' doesn't create any symlinks, so those links you
plrro> probably created when running install-plucker.

>> So then we try and continue 
>> plucker-build -f file
>> 
>> Which results in:
>> exactly the same error as above

plrro> I don't know why you can't remove files from your /tmp dir; never
plrro> seen that problem before.

>> But this does work.

plrro> Well, when you run plucker-build with the -c argument it will just
plrro> write the data to the cache directory; don't know *why* you want to
plrro> do that, but if you look in ~/.plucker/cache I guess you will find
plrro> some files in it...

plrro> When you use -f it will, as you found out, create a Plucker document.

>> Version two:  this one is mostly a rant because using the 
>> install-plucker in the unix dir shouldn't have so many issues.

plrro> The install-plucker script has been deprecated for quite a while
plrro> and is not included in any of the binary packages any longer...

>> running it as sudo results in:
>> 
>> gmake: *** [palmos_client] Error 2
>> Traceback (most recent call last):
>>   File "./install-plucker", line 140, in ?
>>     os.symlink(os.path.join(MOD_DIR, "PyPlucker", "Spider.py"),
>> os.path.join(BIN_DIR, "plucker-build"))
>> OSError: [Errno 17] File exists

plrro> Once again, I don't know why you can't remove files on your system...

>> Why is it going into viewer-GTK+2-POSIX ? 

plrro> Because you told it to.

>> Didn't I tell it to not
>> build gtkviewer with the --disable-gtkviewer option?  

plrro> Nope, not when you *after* running the configure command decide to
plrro> run the install-plucker script (that will run the configure script
plrro> using the settings you select and since the script is deprecated
plrro> it has not been updated to include any options for the GTK viewer).

>> So I add a "install-prc-files: : " section in
>> /home/faux/plucker-1.8/viewer-GTK+2-POSIX/Makefile just to get past
>> it.  I get the same error again.  Why?  Because
>> /home/faux/plucker-1.8/viewer-GTK+2-POSIX/Makefile was overwritten by
>> some portion of theinstall-plucker script.

plrro> Because the script will run configure...

plrro> /Mike



plrro> --__--__--

plrro> _______________________________________________
plrro> plucker-list mailing list
plrro> [email protected]
plrro> http://lists.rubberchicken.org/mailman/listinfo/plucker-list


plrro> End of plucker-list Digest




---8<------------Your original message ends here-----------------

_______________________________________________
plucker-list mailing list
[email protected]
http://lists.rubberchicken.org/mailman/listinfo/plucker-list

Reply via email to