Hi Jim (and Michael)

It is already detect by Petri Dish but need to check log...

packet-ncp2222.c:7885:28: warning: unused variable 'info_str_0xc_req'
[-Wunused-const-variable]
packet-ncp2222.c:7903:28: warning: unused variable
'info_str_0x1109_req' [-Wunused-const-variable]
packet-ncp2222.c:7921:28: warning: unused variable
'info_str_0x1509_req' [-Wunused-const-variable]
...

Wunused-const-variable is no enable by default on Ubuntu buildbot..
(or may be using a old release of Clang)

Regards,

On Tue, Sep 22, 2015 at 11:29 AM, Jim Young <[email protected]> wrote:

>
> Hello Michael,
>
> Absolutely amazing how much effort you must have put into your "Journey of
> a thousand miles".   But it looks like we have a regression with regards to
> commit 38b6f306: Refactor NCP Python data so that INFO column can be
> generated on the fly.
>
> Not all of info_string_t objects defined in packet-ncp2222.c are
> referenced.  On my OS X system, Apple's clang compiler is failing to
> compile packet-ncp2222.c because of the clang compiler option
> -Wunused-const-variable.
>
> I'm seeing messages similar to the following (see attached for more
> context):
>
> > error: unused variable 'info_str_0xc_req'
>
> Currently there are 207 info_string_t objects created by
> ./tools/ncp2222.py in packet-ncp2222.c, but 43 of these no longer have any
> references.  If I revert the three files modified by commit 38b6f306 I see
> references to all 207 info_string_t objects.
>
> Here's a quick-and-dirty bash "one-liner" that I used to extract the
> orphaned info_str objects.
>
> for i in $(grep 'static const info_string_t' packet-ncp2222.c | awk '{
> print $4 }')
> do
>     echo -n "$i :"
>     grep $i packet-ncp2222.c | wc -l
> done | egrep ' 1$'
>
> This "one-liner" makes multiple passes over the file packet-ncp2222.c.
> First it simply extracts all the info_string_t objects.  In the main loop,
> after echoing the current info_string_t item, it greps for any lines having
> a specific info_string_t object name and pipes that output to wc -l to get
> a count of matching lines.  The output of the loop is then piped to egrep
> to select just the items where a single matching line was found (which
> implies a definition but no usage).
>
> I haven't (yet) been able to understand enough about how
> ./tools/ncp2222.py python script goes about its magic when generating
> packet-ncp2222.c to determine why 43 of the 207 info_string_t definitions
> no longer have references.
>
> But I'm guessing you might have a much better clue than me.
>
> Best regards,
>
> Jim Y.
> ___________________________________________________________________________
> Sent via:    Wireshark-dev mailing list <[email protected]>
> Archives:    https://www.wireshark.org/lists/wireshark-dev
> Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
>              mailto:[email protected]
> ?subject=unsubscribe
>
___________________________________________________________________________
Sent via:    Wireshark-dev mailing list <[email protected]>
Archives:    https://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
             mailto:[email protected]?subject=unsubscribe

Reply via email to