[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: gdb-4.18 on HP-UX 11.00 with C++ (gcc-2.95.2): autocompleting method
From: |
Daniel Berlin |
Subject: |
Re: gdb-4.18 on HP-UX 11.00 with C++ (gcc-2.95.2): autocompleting method |
Date: |
21 Feb 2001 10:12:33 -0500 |
User-agent: |
Gnus/5.090001 (Oort Gnus v0.01) XEmacs/21.2 (Thalia) |
address@hidden (Ulrich Windl) writes:
> When auto-completing a breakpoint in GNU gdb 4.18 configured as
> "hppa2.0n-hp-hpux11.00", there is a problem:
>
> After adding the class and "::", all symbols are suggested. If the
> right method is typed, plus a TAB, an illegal name is shown:
>
> (gdb) b HCM_M
> HCM_MOVE_TYPE
> HCM_MSG_CONTAINER
> HCM_MSG_INFO
> HCM_MT_ADMISS
> HCM_MT_BEG_ABS
> HCM_MT_DISMISS
> HCM_MT_END_ABS
> HCM_MT_MOVE
> HCM_MT_VISIT
> HCM_Message
> HCM_Message type_info function
> HCM_Message type_info node
> HCM_Message virtual table
> HCM_Message::do_process(char const *, int, generate &)
> (gdb) b HCM_Message
> HCM_Message
> HCM_Message type_info function
> HCM_Message type_info node
> HCM_Message virtual table
> HCM_Message::do_process(char const *, int, generate &)
> (gdb) b HCM_Message::
> Display all 3854 possibilities? (y or n)
> (gdb) b HCM_Message::do
> do_cleanup
> do_copy_pa20
> do_exitcu
> do_fcvt
> do_osfx__7ostream
> do_process__11HCM_MessagePCciR8generate
> do_process__12HCM_AufnahmePCciR8generate
> do_process__12HCM_ZusfuehrPCciR8generate
> do_process__13HCM_VerlegungPCciR8generate
> do_process__14HCM_EntlassungPCciR8generate
> do_seg_Angeh(char const **, unsigned int *, HCM_Angeh *, MESSAGE_ID
> const *)
> do_seg_Angeh__FPPCcPUiP9HCM_AngehPC10MESSAGE_ID
> do_seg_Ende(char const **, unsigned int *, HCM_Ende *, MESSAGE_ID const
> *)
> do_seg_Ende__FPPCcPUiP8HCM_EndePC10MESSAGE_ID
> do_seg_Fall(char const **, unsigned int *, HCM_Fall *, MESSAGE_ID const
> *)
> do_seg_FallKlassi(char const **, unsigned int *, HCM_FallKlassi *,
> MESSAGE_ID co
> nst *)
> do_seg_FallKlassi__FPPCcPUiP14HCM_FallKlassiPC10MESSAGE_ID
> do_seg_Fall__FPPCcPUiP8HCM_FallPC10MESSAGE_ID
> do_seg_Kopf(char const **, unsigned int *, HCM_Kopf *, MESSAGE_ID *)
> do_seg_Kopf__FPPCcPUiP8HCM_KopfP10MESSAGE_ID
> do_seg_PatStamm(char const **, unsigned int *, HCM_PatStamm *,
> MESSAGE_ID const
> *)
> do_seg_PatStamm__FPPCcPUiP12HCM_PatStammPC10MESSAGE_ID
> do_seg_VersVerh(char const **, unsigned int *, HCM_VersVerh *,
> MESSAGE_ID const
> *)
> do_seg_VersVerh__FPPCcPUiP12HCM_VersVerhPC10MESSAGE_ID
> do_seg_ZusFuehr(char const **, unsigned int *, HCM_ZusFuehr *,
> MESSAGE_ID const
> *)
> do_seg_ZusFuehr__FPPCcPUiP12HCM_ZusFuehrPC10MESSAGE_ID
> doallocate__7filebuf
> doallocate__9streambuf
> doallocbuf__9streambuf
> done
> done20
> done_mm
> done_pa20
> dont_mask
> double
> double type_info function
> double type_info node
> dowrite
> dowritec
> (gdb) b HCM_Message::do_process__1
> the class HCM_Message does not have any method named do_process__1
> Hint: try 'HCM_Message::do_process__1<TAB> or
> 'HCM_Message::do_process__1<ESC-?>
> (Note leading single quote.)
> (gdb) b HCM_Message::do_process
> Breakpoint 1 at 0xb358: file HCMprocess.cc, line 1372.
> (gdb)
>
> Well it works when entered with a single quote at the beginning, but
> why won't it work without?
Because of bugs in the completer.
I fixed this once, but the changes aren't in cvs yet.
>
> Ulrich