15606cf281
com_print(), fix a memory leak
2012-10-17 19:51:30 +02:00
99b0e78c84
DCtrCurv(), fix a memory leak
2012-10-17 19:39:30 +02:00
f038d59fbe
mkfnode(), fix a memory leak
2012-10-17 19:39:02 +02:00
31fa48cf17
com_write(), com_write_sparam(), fix a big memory leak
2012-10-15 20:55:55 +02:00
c6e631f30a
fourier.c: rename stored vectors
2012-10-15 18:37:07 +02:00
ae25f54709
inpcom.c: plug small memory leak
2012-10-15 18:35:12 +02:00
0336e31c57
vector.c, rewrite vec_basename(), FIXME !
...
This rewritten code is functionally equivalent to the previous code,
yet the result is queer,
because the char *v_name slot of struct dvec is read
beyond its terminating '\0'
Is there indeed some code in ngspice which creates
v_name strings with an embedded '\0' ?
2012-10-13 19:09:49 +02:00
ebc50b8639
inpcom.c, rewrite
2012-10-13 16:48:15 +02:00
339a002c55
fourier.c: output also to vector feature request tracker no. 3575841
2012-10-13 15:56:19 +02:00
374b716fdd
fourier.c: remove even more memory leaks
2012-10-13 13:25:00 +02:00
a4e167a657
inpcom.c, use strchr() in favour of strstr()
2012-10-13 12:27:20 +02:00
3b2b130177
inpcom.c, inp_fix_ternary_operator_str(), use copy_substring()
2012-10-13 12:00:29 +02:00
680e45f7f3
inpcom.c, inp_fix_ternary_operator_str(), rewrite
2012-10-13 11:47:22 +02:00
9352a4b72d
inpcom.c, get_instance_subckt(), more carefull when there is trailing whitespace
2012-10-13 11:31:37 +02:00
f61f323be0
inpcom.c, get_instance_subckt(), rewrite
2012-10-13 11:31:37 +02:00
2754771a61
fourier.c: remove memory leaks
2012-10-13 11:21:29 +02:00
7f0065e7fb
inpcom.c, make use of copy_substring()
2012-10-13 11:03:52 +02:00
69bf624097
inpcom.c, rewrite
2012-10-13 11:02:34 +02:00
3f4e2eed49
inpcom.c, bug fix in inp_get_subckt_name()
2012-10-13 10:45:09 +02:00
021a5dd871
inpcom.c, rewrite
2012-10-12 19:23:46 +02:00
e0b796f3c5
inpcom.c, rewrite
2012-10-12 18:10:52 +02:00
3085c471ac
inpcom.c, cleanup
2012-10-12 17:58:05 +02:00
dbf288bae7
inpcom.c, fix a bug
...
which was introduced in commit
skip-ws, #5/6, unify macro versions, obj not invariant
where the SKIP.*back.* macros have been unified
FIXME, the check for `\0' in the backwards direction if of course nonsense.
These backward functions either shall not check it at all (being optimistic),
or they shell check against a start of string pointer.
2012-10-11 19:33:32 +02:00
caf50ba4d8
inpcom.c, cleanup
2012-10-10 20:48:42 +02:00
3ca1235602
gnuplot.c: improve scaling of y axis
2012-10-09 19:47:24 +02:00
2b72fbbc5e
skip_ws(), #2/2, drop macros, obj-invariant
2012-10-08 17:48:27 +02:00
9f3f794ced
skip_ws(), #1/2, use functions instead of macros
2012-10-08 17:28:52 +02:00
701942cebf
skip-ws, #6/6, unify macro versions, obj-invariant
...
checked for object file invariance
2012-10-07 21:53:42 +02:00
532ff76dac
skip-ws, #5/6, unify macro versions, obj not invariant
2012-10-07 21:53:29 +02:00
6a5f234519
skip-ws, #4/6, use macros, obj-invariant
...
checked for object file invariance
2012-10-07 21:52:54 +02:00
4bf52410fb
skip-ws, #3/6, introduce macros, obj-invariant
...
checked for object file invariance
2012-10-07 21:52:35 +02:00
e3b5ff94c5
skip-ws, #2/6, prepare for rewrite, obj-invariant
...
checked for object file invariance
2012-10-07 21:52:22 +02:00
841db86e00
skip-ws, #1/6, prepare for rewrite, obj-invariant
...
checked for object file invariance
2012-10-07 21:52:09 +02:00
996dedaa71
inpcom.c: remove cccs from F, ccvs from H source
2012-10-07 17:12:42 +02:00
e69f50270d
inpcom.c: correct replacement of vccs, vcvs in E, G-sources
...
search for 'par(' instead of 'par'
2012-10-07 15:19:28 +02:00
01ed0ac5d9
com_fft(), com_psd(), com_spec(), cleanup storage more thoroughly
...
use free_pnode() instead of free_pnode_o()
drop free_pnode_o()
2012-10-06 21:07:27 +02:00
dfe8342b91
invoke com_remcirc() in com_quit()
2012-10-06 15:20:17 +02:00
c7717e37fc
cleanup, whitespace
2012-10-06 12:30:31 +02:00
7acc91157a
cleanup, drop unused macros
2012-10-06 12:16:07 +02:00
eb931dc01a
cleanup, ngspice/config.h is already in ngspice/ngspice.h, #2/2
2012-10-06 12:12:17 +02:00
11579d9369
cleanup, ngspice/config.h is already in ngspice/ngspice.h, #1/2 (easy ones)
2012-10-06 12:12:08 +02:00
c336f5ec9d
outif.c, typesdef.c, sim.h: correct labeling of cap, cur, and charge data
2012-10-03 15:01:36 +02:00
7d82367666
b3ld.c, b4ld.c, b3v32ld.c: enable capacitance calculation during dc sweep
2012-10-03 14:28:19 +02:00
705e130868
b3v32.c: make additional parameters accessible by @m1[par]
2012-10-01 23:03:41 +02:00
c0bc0de4b5
prototype for controlled_exit()
2012-10-01 23:02:26 +02:00
7a65ea0318
com_write_sparam(), cleanup storage more thoroughly (`wl_sparam')
2012-10-01 21:28:48 +02:00
98b38556aa
com_alter_common(), cleanup storage more thoroughly (on error)
2012-10-01 21:09:26 +02:00
124f0f4727
com_write(), com_write_sparam(), com_cross(), cleanup storage more thoroughly
2012-10-01 21:03:53 +02:00
7b2c9db809
com_psd(), cleanup storage more thoroughly (`win')
2012-10-01 19:59:43 +02:00
a891e7dc9c
com_print(), cleanup storage more thoroughly
2012-10-01 19:44:49 +02:00
0f089d2bd6
com_compose(), cleanup storage more thoroughly
2012-10-01 18:31:02 +02:00
35c0e9eca3
variable rename, unify the source
...
checked for object file invariance
2012-10-01 17:59:40 +02:00
f95aaaafc1
com_print(), com_cross(), use for(;;) loops
2012-10-01 17:53:01 +02:00
fe828c5081
bsim3, make additonal parameters accessible with @m[par]
2012-09-30 23:49:18 +02:00
44c1380281
variable rename, unify the source
...
checked for object file invariance
2012-09-30 23:26:26 +02:00
bc89379aff
fourier(), use for(;;) loops
2012-09-30 23:11:26 +02:00
d06265142d
variable rename, unify the source
...
checked for object file invariance
2012-09-30 23:10:49 +02:00
e0d64e14cc
bug fix, fourier(), prevent endless loop when processing unexpected vectors
2012-09-30 23:10:38 +02:00
e712080579
com_compose(), use for(;;) loops
2012-09-30 22:50:23 +02:00
959b62e14d
variable rename, unify the source
...
checked for object file invariance
2012-09-30 22:36:12 +02:00
5a58281e7b
com_spec(), use for(;;) loops
2012-09-30 22:22:28 +02:00
5102735df9
variable rename, unify the source
...
checked for object file invariance
2012-09-30 22:22:20 +02:00
5826f3c0e6
com_fft(), com_psd(), use for(;;) loops
2012-09-30 21:44:28 +02:00
26bbb39a68
variable rename, unify the source
...
checked for object file invariance
2012-09-30 21:43:16 +02:00
f52ae6eccd
bug fix, invalid read when the title line is the empty
2012-09-30 18:38:36 +02:00
2506fe3858
changed the #include order - ngspice.h must be first
...
because icc complains about declaration visibility of mmsghdr struct
2012-09-29 20:44:12 +02:00
ebfeaf72f1
bjttemp.c: prevent uninitialized variables
2012-09-29 17:57:35 +02:00
119be60fdc
polish a very old and unused artifact in WANT_SENSE2
2012-09-27 20:43:27 +02:00
ae5de92c69
cleanup db_print_pnode_tree()
2012-09-27 19:49:42 +02:00
44e7f7f7bd
bug fix, dvec in a `define' can't be member of any plot
2012-09-27 19:46:06 +02:00
0280a49b69
bug fix, avoid segfault when the refered plot is still empty
2012-09-27 19:46:06 +02:00
986f9b7535
grid.c: prevent overflow of div_list
2012-09-26 21:12:26 +02:00
e096101133
ifeval.c: reduce fudge factor for div and atanh
2012-09-26 21:00:40 +02:00
0bdc33db40
grid.c: prevent crash if real low and high values are close
2012-09-26 20:58:58 +02:00
68c4de76b7
new option noopac, don't do op simulation before ac if circuit is linear
...
May simplify ac simulation of passive RLC networks
(No need to add series resistors in Ls, allows to C inseries)
2012-09-26 16:59:59 +02:00
998c7f2934
disable some more cleanup code in com_quit()
...
destroy_ivars() is reported to freeze
2012-09-25 21:31:45 +02:00
ee0b602dc4
subckt.c, cleanup, use `for' instead of `while'
2012-09-22 22:12:56 +02:00
fe1acd3757
src/frontend/** remove superfluous `return' invocations
2012-09-22 19:33:55 +02:00
ab98f03622
src/frontend/** whitespace, indentation, ...
...
checked for object file invariance on linux
2012-09-22 19:33:50 +02:00
d1d4eb11a5
invoke cleanup functions in com_quit()
...
trying to get the valgrind reports down.
2012-09-22 10:53:05 +02:00
71240d19be
free spice_init_devices() allocated memory in com_quit()
2012-09-21 21:19:44 +02:00
811bf5664a
destroy `const' plot in com_quit()
2012-09-21 21:19:44 +02:00
7a5eeb2d9c
drop a superfluous cast
2012-09-21 21:19:44 +02:00
fb23e1f024
correct the diode temperature model
2012-09-20 21:06:34 +02:00
c0e2d4abef
src/frontend/**, whitespace,
...
object files will change, due to __LINE__ usage
2012-09-20 20:32:02 +02:00
7454a6d486
src/frontend/**, whitespace, indentation, ...
...
untabify
delete-trailing-whitespace
braces
...
checked for object file invariance on linux
2012-09-20 20:30:53 +02:00
f3b5336d3a
cleanup some macros, try to be more careful
2012-09-19 20:16:27 +02:00
ec1a71249a
fix ownership of `cp_promptstring' and `cp_program'
2012-09-15 14:58:39 +02:00
198604f605
define.c, plug a memleak in com_undefine()
2012-09-13 21:52:03 +02:00
6a3c20706a
define.c, cleanup
2012-09-13 21:21:46 +02:00
b68259ff2d
define.c, cleanup
2012-09-13 21:21:08 +02:00
1fbc4d62cc
define.c, whitespace cleanup
2012-09-13 21:18:02 +02:00
38ce3b6142
define.c, bug fix, when undefining more than one udfunc
2012-09-13 21:05:56 +02:00
adc9ee09ce
api change for ngdirname() to fix a memory leak
...
ngdirname did `own' the returned string.
now the invoker is responsible for the returned string.
note, this is contrary to the POSIX dirname() implementation,
which *might* return pointers to statical allocated memory.
2012-09-13 20:07:33 +02:00
7c9b90c9ea
XSPICE lcouple: remove unused cm_analog_integrate
2012-09-11 21:29:16 +02:00
cb220b2d6f
XSPICE core: remove memory leak
2012-09-10 23:40:40 +02:00
8d374157fe
XSPICE square: remove memory leak, initialize phase1
2012-09-10 22:52:10 +02:00
d6698c5f6c
XSPICE square: indentations
2012-09-10 22:20:41 +02:00
2379bd577a
XSPICE triangle: remove memory leak remove initialization error for phase1 FIXME: there are still more initialization errors
2012-09-09 19:07:29 +02:00
8ece02a1cd
XSPICE triangle: indentations
2012-09-09 17:56:12 +02:00
feb545d8a7
XSPICE oneshot: remove memory leak remove using uninitialzed variable old_clock correct waveform shape by using permanent breakpoints
2012-09-09 16:55:10 +02:00
36898f9921
indentations
2012-09-09 16:44:58 +02:00
bcbd506ccc
XSPICE sine: out-of-bounds removed
2012-09-08 01:38:25 +02:00
10d97d222e
XSPICE sine source: memory leak removed (bug no. 3564166)
2012-09-08 00:45:59 +02:00
9f8c73d271
cktdest.c: reset for XSPICE allocated memory (not yet complete)
2012-09-07 23:50:38 +02:00
5ce51f148e
XSPICE filesource: remove segfault with op before tran
2012-09-07 23:48:09 +02:00
9de9336581
cktload.c, indentations
2012-09-02 16:11:46 +02:00
4e45f4cba0
inp.c: unify syntax of .save and save commands: allow node1 and v(node1), v2#branch and i(v2), @dev[param]
2012-08-30 11:03:48 +02:00
0efaeac8b4
evtqueue.c: plug memory leak
2012-08-27 10:42:12 +02:00
417d109764
resask.c: avoid segfault if @R[i] is asked before analysis is run
2012-08-27 00:12:14 +02:00
9eaff645f5
vectors.c, resask.c: indentation
2012-08-26 23:32:21 +02:00
a5a1029998
d_source: Use STATIC_VAR_TABLE to save memory
2012-08-25 13:39:27 +02:00
3d7ab6a297
cleanup keyword databases on proram termination
2012-08-23 21:19:15 +02:00
80bc975e33
plug a memory leak in inp_spsource/inp_nutsource (deck title string)
2012-08-23 21:13:44 +02:00
84283b43d4
cleanup ivars() related globals on program termination
2012-08-23 21:09:25 +02:00
f4ad29a17c
d_source: allow multiple instances
2012-08-22 23:14:26 +02:00
83c99bece4
cleanup wallace arrays on program termination
2012-08-22 22:21:33 +02:00
590fc543f8
avoid error messages when finally closing the display device.
...
these messages creeped in with commit
2012-08-19 18:58:31
"plug memory leak in x11.c"
when the "printf" device Close and Update
functions might be invoked on program termination.
use the silent "nop" implementation
instead of the error reporting "nodev" implementation
2012-08-20 21:26:08 +02:00
7191fdce63
d_source cfunc.mod: plug memory leaks
2012-08-20 08:54:26 +02:00
488bc6a8c8
bug no. 3456221: new d_source model
2012-08-19 21:36:33 +02:00
b5b0ec7302
cleanup, white space changes
2012-08-19 21:36:31 +02:00
6c01089be1
plug memory leak in x11.c
...
try to obey the Xt toolkit protocol when closing widgets and application
2012-08-19 19:18:34 +02:00
cb42989204
mifgetvalue.c: need a copy to allow freeing token
2012-08-19 15:49:24 +02:00
789edee376
rewrite inp_list() avoid duplicate code
2012-08-19 14:04:37 +02:00
c9e2a21a9c
reimplement tcl_vfprintf()
2012-08-19 12:53:55 +02:00
c4d0feb6cc
bug fix, out_send() erroneously interpreting printf %-style sequences
2012-08-19 12:20:33 +02:00
b9b9f6fd1c
cleanup (checked for code invariance)
2012-08-19 12:14:22 +02:00
5ca213b481
part of temperature level=3 model
2012-08-18 21:10:48 +02:00
67132108ad
out_printf() is not supposed to be a preprocessor macro
2012-08-18 18:40:45 +02:00
0af75f3bd9
subckt.c, mif_inp2.c: plug XSPICE setup memory leaks
2012-08-18 18:31:23 +02:00
d4ced47e8e
drop out_pbuf, which was an ancient workaround
...
obviously an old implementation of out_printf()
has been unreliable.
2012-08-18 18:27:13 +02:00
0825c10471
bug fix, print complex yet real valued frequency
...
when printing a complex frequency vector, yet with real value
some left over nonsens was printed instead of the actual real value.
this bug was introduced in commit:
Date: Thu Jan 15 21:08:09 2009 +0000
Improvements in vector derivative computation and new functions to
compute group delay and moving average. From espice (A. Roldan).
2012-08-18 17:10:06 +02:00
b02a497490
mif_inp2.c, subckt.c: plug memory leaks at XSPICE setup
2012-08-15 23:11:07 +02:00
9390a9a9ec
rewrite com_quit()
...
emphasize user confirmation
and if we are going to remove dangling circuits,
then do so unconditionally
2012-08-15 20:15:15 +02:00
be183407cd
cleanup (checked for code invariance)
2012-08-15 19:18:15 +02:00
ae6a16e3e4
plug a memory leak when executing `shell'
2012-08-14 21:00:21 +02:00
fa81e9d007
plug a memory leak when (un)setting variables
2012-08-14 20:39:11 +02:00
64139a243c
spiceif.c, cleanup (checked for code invariance)
2012-08-14 20:38:25 +02:00
0c230b2970
spiceif.c, cleanup (checked for code invariance)
2012-08-12 20:28:42 +02:00
10a9ae32c5
options.c cleanup
2012-08-12 12:48:27 +02:00
a2f58da0d0
options.c, whitespace
2012-08-12 12:43:09 +02:00
e93960f9ed
cleanup, if_options() will copy() anyway
2012-08-12 12:25:00 +02:00
7126dd6f1d
display.c, cleanup
2012-08-08 18:54:52 +02:00
0beb95757f
cancer of the semicolon
2012-08-07 21:20:36 +02:00
d50ffdd987
swallow warnings
2012-08-06 19:50:23 +02:00
1a2477bdcc
avoid `meas' syntax error causing segfault
...
Invalid meas crashes ngspice with SIGSEV - ID: 3554330
http://sourceforge.net/tracker/?func=detail&aid=3554330&group_id=38962&atid=423915
2012-08-06 19:42:49 +02:00
46980f2cc6
plotit.c #9/9, whitespace
2012-08-05 21:41:25 +02:00
4803d8bba2
plotit.c #8/9, braces
2012-08-05 21:41:25 +02:00
6a334da2b4
plotit.c #7/9, braces
2012-08-05 21:41:25 +02:00
f587553217
plotit.c #6/9, braces
2012-08-05 21:41:25 +02:00
088e88d9b2
plotit.c #5/9, localize
2012-08-05 21:41:25 +02:00
5f1387892c
plotit.c #4/9, line wrap
2012-08-05 21:41:25 +02:00
ed4ad1c0dd
plotit.c #3/9, drop some comments
2012-08-05 21:41:25 +02:00
5c43e8f2fc
plotit.c #2/9, emphasize maximize/minimize
2012-08-05 21:41:25 +02:00
406fc29007
plotit.c #1/9, think positive
2012-08-05 21:41:25 +02:00
11ae043c3d
com_measure2.c: enable 'meas dc' without 'from' and 'to' given
...
measure.c: no error printout during autostop
2012-08-05 20:03:00 +02:00
d6ba5e5b3f
cleanup free_struct_variable()
2012-08-05 12:22:47 +02:00
75c59a26fa
variable.c, cleanup (mostly whitespace)
2012-08-05 12:22:47 +02:00
99b167b272
correct alter of mos geometries for binned models
2012-08-05 12:11:16 +02:00
51271d2a0b
cp_oddcomm(), plug a memory leak
2012-08-04 22:27:04 +02:00
acfc7a2a27
remove memory leaks
2012-08-04 19:09:13 +02:00
2516a118a3
indentation
2012-08-04 16:24:38 +02:00
713091ea22
wlist #3/3, whitespace
2012-08-04 11:40:35 +02:00
fe5ac28b92
wlist #2/3, cleanup
2012-08-04 11:40:35 +02:00
406dcc3519
wlist #1/3, cleanup
2012-08-04 11:40:35 +02:00
05437b1d48
inpcom.c: safely uncomment unused subcircuits (top level only)
...
if they contain further nested subcircuits
2012-08-03 22:42:33 +02:00
9ed7ce4e9e
wlist.c, `const' ness
2012-08-03 21:46:13 +02:00
e4c7a8df32
use wl_find()
2012-08-03 20:48:11 +02:00
cf2a0f3df3
cleanup, linewrap
2012-08-03 20:43:12 +02:00
191ae9ea90
cleanup, char 0
2012-08-03 20:43:10 +02:00
5a69b9efc4
cleanup, void return
2012-08-03 20:43:06 +02:00
ef37252155
lexical, use wl_append_word()
2012-08-03 20:24:58 +02:00
62a4ee77d5
lexical #6/6, whitespace
2012-08-03 20:20:18 +02:00
ea558fbc7f
lexical #5/6, cleanup
2012-08-03 20:18:17 +02:00
aa188c0766
lexical #4/6, parentheses and braces
2012-08-03 20:16:47 +02:00
c73c0f4318
lexical #3/6, rewrite
2012-08-03 20:12:58 +02:00
5656203a6c
lexical #2/6, linewrap
2012-08-03 20:12:19 +02:00
5d245c9dab
lexical #1/6, cleanup
2012-08-03 20:09:34 +02:00
96d2ff96e0
alloc.c: exlude controlled_exit if not MS Windows
...
not available in some tests in cmaths and executable like ngproc2mod
2012-08-03 08:28:50 +02:00
64d00a6b43
lexical #3/3, collapse append(NULL) into cp_ccom()
...
the sequence
orig_wlist = wlist;
append(NULL);
cp_ccom(wlist, ...);
is transformed.
expressions to care about in cp_ccom()
C1 := (wlist->wl_next != NULL)
C2 := (wlist->wl_word != NULL)
C3 := wl_length(wlist)
D1 := (orig_wlist != NULL)
D2 := (orig_wlist->wl_word != NULL)
D3 := wl_length(orig_wlist);
cases and their significance in cp_ccom():
orig_wlist = NULL;
wlist = wl_cons(NULL, NULL);
C1 == false
D1 == false
C2,C3 dont care
orig_wlist = wl_cons(E1, wl_cons(E2, ... wl_cons(En, NULL)))
wlist = wl_cons(E1, wl_cons(E2, .... wl_cons(En, wl_cons(NULL, NULL))))
C1 == true
D1 == true
C2 === D2
C3 === D3 + 1
2012-08-02 18:55:54 +02:00
fec5b00a0d
lexical #2/3, fix memory leaks
2012-08-02 18:55:54 +02:00
042311d0c3
lexical #1/3, move append(NULL) invokations into position
2012-08-02 18:55:54 +02:00
67e0cf0608
undo controlled_exit: not compatible with ancient tests in cmath under LINUX
2012-08-01 22:10:13 +02:00
1195498978
lexical #19/19, rewrite
2012-07-31 20:49:34 +02:00
5b67ecda79
lexical #18/19, rewrite
...
`cw' won't be looked at, thus the append + removal can be dropped
`wl_word' is already NULL
2012-07-31 20:49:34 +02:00
f990e517b5
lexical #17/19, rewrite
2012-07-31 20:49:34 +02:00
fa9be4336a
lexical #16/19, transform append --> append(word)
2012-07-31 20:49:34 +02:00
43a6d59c72
lexical #15/19, drop `delayed' which is never zero ...
2012-07-31 20:49:33 +02:00
47b5e9576d
lexical #14/19, drop `delayed = 0' wherever its effect isn't visible
...
without effect, because we return or will overwrite it anyway
2012-07-31 20:49:33 +02:00
01df4e9437
lexical #13/19, delay `append'
...
delay its execution, until
either `wlist' or `cw' will be accessed,
or
a return or goto nloop is inevitable
2012-07-31 20:49:33 +02:00
cff6188fe3
lexical #12/19, use `append'
...
some `assumption' are no longer valid
2012-07-31 20:49:32 +02:00
5ad8382cf1
lexical #11/19, aproach `append' here
2012-07-31 20:49:32 +02:00
d61f3c10a0
lexical #10/19, split off `append' from `newword'
2012-07-31 20:49:32 +02:00
af0a0c2c03
lexical #9/19, aproach the `append'
...
cw has never been NULL here,
because we used to survive an update of cw->wl_next here
wlist hasn't been NULL either,
because there is no place where wlist becomes NULL without
setting cw to NULL as well. yet cw isn't NULL (see above)
aproach the `append'
2012-07-31 20:49:31 +02:00
74cc38f67a
lexical #8/19, express where `wlist' and `cw' start being != NULL
2012-07-31 20:49:31 +02:00
3cbc4b0773
lexical #7/19, express verbosely where `wlist' and `cw' are NULL
2012-07-31 20:49:31 +02:00
ecb2eec279
lexical #6/19, express cp_ccom() property
2012-07-31 20:49:31 +02:00
3b575d5e3d
lexical #5/19, collapse into `newword'
2012-07-31 20:49:30 +02:00
348dc2d7d1
lexical #4/19, the `else' can take that job (unconditionally)
2012-07-31 20:49:30 +02:00
80191f3d44
lexical #3/19, attach and instantly remove a NULL wl_word here
...
this modification is faithful iff cw != NULL and cw->wl_next == NULL
cw != NULL because we have just survived an update of cw->wl_word
cw->wl_next == NULL is witnessed by:
1) cw either NULL or pointing to a wl_cons(,NULL)
2) cw not being modified, and no wl_next being modified
3) check of all function invocations with wlist/cw argument
not to modify any wl_next slot
which are:
1) cp_ccom(wlist, ...)
2) wl_free() always followed by goto nloop, which does a cw = wl_cons(,NULL)
and cp_ccom() doesn't modify any wl_next slot
the first part of the patch is intentionally similar to newword
`i' and `buf' can be modified here, because we will inevitably `goto done'
attach and instantly remove a NULL wl_word here
2012-07-31 20:49:30 +02:00
rlar
h_vogt
dwarning