Online Archives

Online Archives

[Mark Mitchell]

The online archives of the egcs mailing lists have one flaw: the
patches posted there don't work when saved from Netscape.  I get
messages like:

  patch: **** malformed patch at line 11: set gccpath [file dirname [file dirname ${base_dir}]]

  supernova% patch < ~/tmp/patch
  patch: **** malformed patch at line 34: set gccpath [file dirname [file dirname ${base_dir}]]

I even tried this one of my own patches, which I knew must have been
wellformed! :-)  I assume that the archiving software is screwing this
up; does anyone know if there's a way around it?  This happens both
with cut-and-paste from the broswer, and with File/Save from the menu
when selecting either "Source" or "Text".

Thanks,

-- 
Mark Mitchell		mmitchell@usa.net
Stanford University	http://www.stanford.edu

Re: Suggestion re: 970825 Snapshot Available

[Julian Seward]

Joe Buck wrote:
> 
> > BWT was invented by people working for digital, and to my information they
> > either already hold a patent or want to apply for one.
> 
> If they still "want to apply for one", they cannot.  In the US, they
> have one year from date of first publication to apply.  In most other
> countries, they must have filed the patent before first publication.
> (This is why the RSA patent is only good in the US -- they delayed the
> patent application).
> 
> So either they have a patent, or they have completed an application and
> it is working its way through the process.  There are no other possibilities.
> 
> > DEC "said" they don't want to exploit the BWT commercially, but that's what
> > unisys said about LZW...
> 
> Do you have contacts for any of the BWT people?  We could ask ...

Perhaps I should be clearer.

To the best of my knowledge, bzip2 doesn't use any patented algorithms.
However, clearly, as an individual, I don't have the resources to
conduct
a proper patent search myself.  So I cannot actually guarantee this.

The "Wheeler" in BWT is a prof at Cambridge.  email is 
djw3@cl.cam.ac.uk.  I don't have an email contact for Mike Burrows.

J

970828 snapshot

[Jeffrey A Law]

I'm away from the office right now, but I figured I'd go ahead and
put the 970828 snapshot up for ftp.

Diffs are included to bring 970825 up to 970828.  There are no changes
in the testsuite directory itself, so there's no need to fetch the
diff.gz files for the testsuites.

I'll update the snapshot web page when I get back to the office later
tonight.  I'll also look through some of the submissions from today
(like the sh rtems port).


Enjoy!
Jeff

Jeff Law (law@cygnus.com)
Cygnus Solutions		EGCS GNU Compiler System
http://www.cygnus.com		http://www.cygnus.com/egcs

Re: Online Archives

[Jason Molenda]

I wrote:

> I'll see if me and hypermail can reach an understanding about this.


OK, I won.  Hypermail isn't munging the  character any more.
Things that look like SGML elements may confuse your web browser, so
for instance "Jason Molenda " is likely to be
rendered as just "Jason Molenda " because the browser doesn't know any
element called crash@cygnus.co.jp.  (previously this would be changed
into "Jason Molenda "... this causes obvious
problems with C source code.)


Anyway, patches should now be archived safely.  Let me know if anyone
has a problem with the stup.

Jason

Re: Online Archives

[Jason Molenda]

I wrote:

> 
> Fixed.  



I lied.  Hypermail is still converting < & > characters into SGML
entities.  It is the correct thing to do, but it means patches won't
match.

I'll see if me and hypermail can reach an understanding about this.

Jason

Re: Online Archives

[Jason Molenda]

On Thu, Aug 28, 1997 at 04:43:42PM -0700, Mark Mitchell wrote:

> The online archives of the egcs mailing lists have one flaw: the
> patches posted there don't work when saved from Netscape.  


Fixed.  You'll have to save the files as "Source".  TABs etc are all
preserved now.  Patch should be able to see through all the HTML cruft
at the beginning and end.

I'm not going to bother with regenerating the already-formatted
messages, but anything that is sent to the egcs* lists will be handled
correctly from now on.


Jason