Name: krC82822 Date: 05/17/2001


(1.3.1 final release)

NOTE: while the presence of CTL-M's in the Solaris download
".sh" files has been confirmed, this does NOT preclude extracting
the distribution here at Sun (using what appear to be the same
tools the user is employing). However, the CTL-M's DO preclude extraction
at the user's site. This bug is being filed as a "marker" to clarify
what is going wrong at the user's site. (see notes below)
-------------

User's Description of the problem:
----------------------------------------------

The new JDK 1.3.1 software distribution will not unpack. Get an error message
of "./j2sdk-1_3_1-solsparc.sh: syntax error at line 205: `in^M' unexpected".
An examination of the file shows carriage returns (0x0d) at the end of each line
as if some one built this file on a WIntel box. Attempts to strip the offending
trailing '^M' results in a file which fails at checksum since one cannot tell
which 0x0d values are valid and which are not.

------------

17 May 2001, eval1127@eng -- works fine for me from the following locations
on download page (using Netscape 4.76 on a Solaris machine):

1) FTP (main java.sun.com one)
2) HTTP (from bottom of page)

Will try on PC with MSIE, followed by local FTP to a Solaris box...

1) FTP will not save .sh to file...file contents get "spewed" to the main browser window (MSIE 4.x);
     checked all browser options, and could not figure out how to disable this
2) HTTP, followed by a bin mode FTP to my Solaris box, works fine

-----------

19:25 PDT: user called, and reported that he's using Netscape 4.76 on a Solaris box.
The only download that shows up without CTL-M's at the end of each line is the HTTP one.

I verified that CTL-M's are included at the end of every line in the FTP downloads.

However, HERE, I am able to extract successfully -- consistently (I made one error
while on the phone with the user, unfortunately, which confused things -- I failed to
delete the previous extraction before re-attempting).

I failed to ask the user what his default SHELL is (since the script runs in its own,
it may not matter).
(Review ID: 124594)
======================================================================