Re: Bug Fix Session

[Jason Bechtel <jason@xxxxxxxxxxx>]

Scott-

Sorry I didn't see your message.  Netscape did not *bold* my tn5250
folder for the last few weeks even though I was receiving messages
there.  I thought the list was just really quiet.  :)

So, thank you for taking a day off work to do bugfixing!  That's really
awesome!  I wish my boss would let me do that... <sigh>

The "Deep Bug" with the AS/400 I/O error is still outstanding as far as
I know.  I will try to download and compile the CVS version from
Sourceforge soon and let you know, unless you can tell me what the
status of that bug is before I spend the time.

The main problem now is that no matter what version I use (0.15, 0.16,
or 0.17-cvs) KP_Add is mapped to <Enter>!  This should be interpreted as
"Field-Exit".  I'm sorry if this is a stupid question, but how do I map
this?  I've looked at Xmodmap, .Xdefaults,
/usr/local/tn5250/share/tn5250/XTerm, and
/usr/local/tn5250/share/tn5250/us5250.map and I'm getting confused. 
Xmodmap just assigns a keysym to a physical key on the keyboard (e.g.
keycode 22 = BackSpace).  That occurs at the X level.  Then .Xdefaults
contains some assertions about X-Window properties.  Here is the
relevant portion of the .Xdefaults file we are using:

xt5250*VT100.Translations: #override\n\
        <Key>KP_Enter: string("\015") \n\
        <Key>KP_Add: string("\033OM") \n\
        <Key>BackSpace: string("\177") \n\
        <Key>End: string("\033[4~") \n\
        Shift <Key>Escape: string("\001") \n\
        <Key>Down: string("\033[B") \n\
        <Key>Up: string("\033[A") \n\
        <Key>Left: string("\033[D") \n\
        <Key>Right: string("\033[C") \n\
        ~Shift <Key>F1:  string("\033[[A") \n\
        ~Shift <Key>F2:  string("\033[[B") \n\
        ~Shift <Key>F3:  string("\033[[C") \n\
        ~Shift <Key>F4:  string("\033[[D") \n\
        ~Shift <Key>F5:  string("\033[[E") \n\
        ~Shift <Key>F6:  string("\033[17~") \n\
        ~Shift <Key>F7:  string("\033[18~") \n\
        ~Shift <Key>F8:  string("\033[19~") \n\
        ~Shift <Key>F9:  string("\033[20~") \n\
        ~Shift <Key>F10: string("\033[21~") \n\
        ~Shift <Key>F11: string("\033[23~") \n\
        ~Shift <Key>F12: string("\033[24~") \n\
        Shift <Key>F1: string("\033[25~") \n\
        Shift <Key>F2: string("\033[26~") \n\
        Shift <Key>F3: string("\033[28~") \n\
        Shift <Key>F4: string("\033[29~") \n\
        Shift <Key>F5: string("\033[31~") \n\
        Shift <Key>F6: string("\033[32~") \n\
        Shift <Key>F7: string("\033[33~") \n\
        Shift <Key>F8: string("\033[34~") \n\
        Shift <Key>F9: string("\033[35~") \n\
        Shift <Key>F10: string("\033[36~") \n\
        Shift <Key>F11: string("\033[37~") \n\
        Shift <Key>F12: string("\033[38~") \n\
        Shift <Key>Tab: string("\033[Z") \n\
        ~Shift <Key>Delete: string("\033[3~") \n\
        ~Shift <Key>Insert: string("\033[2~") 
xt5250.VT100*underLine: True
xt5250.VT100*colorULMode: off

The diff between your XTerm file and our .Xdefaults file reveals the
following:

<       <Key>KP_Enter: string("\033OM") \n\
<       <Key>BackSpace: string("\177") \n\
<       <Key>End: string("\033[4~") \n\
<       <Key>Home: string("\033[1~") \n\
---
>         <Key>KP_Enter: string("\015") \n\
>         <Key>KP_Add: string("\033OM") \n\
>         <Key>BackSpace: string("\177") \n\
>         <Key>End: string("\033[4~") \n\
>         Shift <Key>Escape: string("\001") \n\
>         <Key>Down: string("\033[B") \n\
>         <Key>Up: string("\033[A") \n\
>         <Key>Left: string("\033[D") \n\
>         <Key>Right: string("\033[C") \n\

It looks like someone (not me) put the KP_Enter string on KP_Add.  This
is causing the KP_Add to behave like KP_Enter.  I take this line out and
KP_Add goes back to being a plain old '+' (i.e. not Field-Exit).  What
am I missing?

Jason

PS:  Thanks for remembering my bug report and keeping track of that
trace file!  All your help is very appreciated.


Scott Klement wrote:
> 
> Aside from this S/36 business, the only other outstanding bug that I can
> think of is the "Deep Bug" reported by Jason Bechtel.
> 
> I'm not certain if Mike Madore was ever able to solve it.  (Or, at any
> rate, I didn't hear of anything)...  Maybe if Jason is reading this, he
> can comment?
> 
> Otherwise, you can find it in the archives here:
> http://archive.midrange.com/linux5250/200105/msg00008.html
> 
> Every other bug that I've come across, I've fixed and committed myself.
> (Which I'm kinda proud of...)
> 
> I think I've still got the tracefile for the "Deep Bug" problem, I'll
> send it to you privately.

-- 
  __________________________________  __________________________
 /         Jason Bechtel            \/  Phone:  (419) 861-3331  \
/       Unique Systems, Inc.         \    FAX:  (419) 861-3340   \
) 6920 Spring Valley Drive, Ste. 106 (                           (
\        Holland, OH 43528           / mailto:jason@uniqsys.com  /
 \__________________________________/\__________________________/
+---
| This is the LINUX5250 Mailing List!
| To submit a new message, send your mail to LINUX5250@midrange.com.
| To subscribe to this list send email to LINUX5250-SUB@midrange.com.
| To unsubscribe from this list send email to LINUX5250-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---