Hi,
2016/1/25 Mon 8:13:42 UTC+9 Ken Takata wrote:
> Hi,
>
> 2016/1/22 Fri 21:00:16 UTC+9 Ken Takata wrote:
> > Hi,
> >
> > 2016/1/21 Thu 21:48:44 UTC+9 Ken Takata wrote:
> > > Hi mattn and all,
> > >
> > > I think that supporting 64-bit Number (even on 32-bit systems) is very
> > > useful.
> > > One of the best examples is getfsize() as mattn said. 32-bit Number is
> > > too
> > > small to represent a filesize nowadays.
> > >
> > >
> > > 2014/1/6 Mon 15:34:44 UTC+9 mattn wrote:
> > > > https://gist.github.com/mattn/8278843
> > > >
> > > > This is in progress.
> > >
> > > I have updated mattn's patch:
> > > https://bitbucket.org/k_takata/vim-ktakata-mq/src/fd966d51f56b69e02d9ebb984355f692b73b2260/num64.patch?at=default&fileviewer=file-view-default
> > >
> > > Changes are:
> > >
> > > * Sync with the latest codebase.
> > > * VC9 or earlier doesn't have stdint.h, int64_t and uint64_t.
> > > Use __int64 for old Windows compilers.
> > > * Define FEAT_NUM64 in feature.h.
> > > (If someone doesn't like this feature, s/he can disable this by editing
> > > feature.h.)
> > > * Display +/-num64 in :version.
> > > * Add has('num64').
> > > * Adjust the behaviour when dividing by 0. (:help expr-/)
> > > * Adjust the behaviour of float2nr().
> > >
> > > I confirmed that getfsize() returned a right value even if the filesize
> > > was
> > > over 2 GiB. (My largefile patches are also needed.)
> > > I also confirmed that all existing tests passed.
> > >
> > > Of cause, more tests are needed, and documents should be updated.
> > >
> > > (See also: [patch] Always use long for vimscript Numbers
> > > https://groups.google.com/d/topic/vim_dev/Ju0DDCqeuMc/discussion )
> >
> > I have updated the patch and divided it into two:
> > https://bitbucket.org/k_takata/vim-ktakata-mq/src/ec8fa710e76975724c30d0f84c02e1460e5bb705/num64.patch?at=default&fileviewer=file-view-default
> > https://bitbucket.org/k_takata/vim-ktakata-mq/src/ec8fa710e76975724c30d0f84c02e1460e5bb705/incdec64.patch?at=default&fileviewer=file-view-default
> >
> > If incdec64.patch is applied, incrementing/decrementing with <C-A>/<C-X>
> > uses
> > 64-bit integer on supported platforms. (Currently, 64-bit int is used on
> > 64-bit
> > Linux but not on 64-bit Windows.)
>
> I have updated the paches:
> https://bitbucket.org/k_takata/vim-ktakata-mq/src/d1b6fe7c1f38ffc8d7fda03fda31d0dec6a5cbd1/num64.patch?fileviewer=file-view-default
> https://bitbucket.org/k_takata/vim-ktakata-mq/src/d1b6fe7c1f38ffc8d7fda03fda31d0dec6a5cbd1/incdec64.patch?fileviewer=file-view-default
>
> Now, vim_snprintf() supports 64-bit formats (%lld, etc.).
Here are the updated num64 patches for 7.4.1202 (including docs).
Also available on my bitbucket repository:
https://bitbucket.org/k_takata/vim-ktakata-mq/src
Regards,
Ken Takata
--
--
You received this message from the "vim_dev" maillist.
Do not top-post! Type your reply below the text you are replying to.
For more information, visit http://www.vim.org/maillist.php
---
You received this message because you are subscribed to the Google Groups
"vim_dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
# HG changeset patch
# Parent 144cb8c9cc01942b9183d72f53e9ecbe9aa0967f
diff --git a/src/Make_cyg_ming.mak b/src/Make_cyg_ming.mak
--- a/src/Make_cyg_ming.mak
+++ b/src/Make_cyg_ming.mak
@@ -378,7 +378,7 @@ endif # RUBY
# Any other defines can be included here.
DEF_GUI=-DFEAT_GUI_W32 -DFEAT_CLIPBOARD
DEFINES=-DWIN32 -DWINVER=$(WINVER) -D_WIN32_WINNT=$(WINVER) \
- -DHAVE_PATHDEF -DFEAT_$(FEATURES)
+ -DHAVE_PATHDEF -DFEAT_$(FEATURES) -DHAVE_STDINT_H
ifeq ($(ARCH),x86-64)
DEFINES+=-DMS_WIN64
endif
diff --git a/src/Make_mvc.mak b/src/Make_mvc.mak
--- a/src/Make_mvc.mak
+++ b/src/Make_mvc.mak
@@ -491,6 +491,11 @@ OPTFLAG = $(OPTFLAG) /GL
CFLAGS=$(CFLAGS) $(WP64CHECK)
!endif
+# VC10 or later has stdint.h.
+!if $(MSVC_MAJOR) >= 10
+CFLAGS = $(CFLAGS) -DHAVE_STDINT_H
+!endif
+
# Static code analysis generally available starting with VS2012 (VC11) or
# Windows SDK 7.1 (VC10)
!if ("$(ANALYZE)" == "yes") && ($(MSVC_MAJOR) >= 10)
diff --git a/src/charset.c b/src/charset.c
--- a/src/charset.c
+++ b/src/charset.c
@@ -1897,14 +1897,14 @@ vim_str2nr(start, prep, len, what, nptr,
is bin */
int *len; /* return: detected length of number */
int what; /* what numbers to recognize */
- long *nptr; /* return: signed result */
- unsigned long *unptr; /* return: unsigned result */
+ varnumber_T *nptr; /* return: signed result */
+ uvarnumber_T *unptr; /* return: unsigned result */
int maxlen; /* max length of string to check */
{
char_u *ptr = start;
int pre = 0; /* default is decimal */
int negative = FALSE;
- unsigned long un = 0;
+ uvarnumber_T un = 0;
int n;
if (ptr[0] == '-')
@@ -1973,7 +1973,7 @@ vim_str2nr(start, prep, len, what, nptr,
/* octal */
while ('0' <= *ptr && *ptr <= '7')
{
- un = 8 * un + (unsigned long)(*ptr - '0');
+ un = 8 * un + (uvarnumber_T)(*ptr - '0');
++ptr;
if (n++ == maxlen)
break;
@@ -1986,7 +1986,7 @@ vim_str2nr(start, prep, len, what, nptr,
n += 2; /* skip over "0x" */
while (vim_isxdigit(*ptr))
{
- un = 16 * un + (unsigned long)hex2nr(*ptr);
+ un = 16 * un + (uvarnumber_T)hex2nr(*ptr);
++ptr;
if (n++ == maxlen)
break;
@@ -1997,7 +1997,7 @@ vim_str2nr(start, prep, len, what, nptr,
/* decimal */
while (VIM_ISDIGIT(*ptr))
{
- un = 10 * un + (unsigned long)(*ptr - '0');
+ un = 10 * un + (uvarnumber_T)(*ptr - '0');
++ptr;
if (n++ == maxlen)
break;
@@ -2011,9 +2011,9 @@ vim_str2nr(start, prep, len, what, nptr,
if (nptr != NULL)
{
if (negative) /* account for leading '-' for decimal numbers */
- *nptr = -(long)un;
+ *nptr = -(varnumber_T)un;
else
- *nptr = (long)un;
+ *nptr = (varnumber_T)un;
}
if (unptr != NULL)
*unptr = un;
diff --git a/src/eval.c b/src/eval.c
--- a/src/eval.c
+++ b/src/eval.c
@@ -817,7 +817,7 @@ static int handle_subscript(char_u **arg
static typval_T *alloc_tv(void);
static typval_T *alloc_string_tv(char_u *string);
static void init_tv(typval_T *varp);
-static long get_tv_number(typval_T *varp);
+static varnumber_T get_tv_number(typval_T *varp);
#ifdef FEAT_FLOAT
static float_T get_tv_float(typval_T *varp);
#endif
@@ -1315,7 +1315,7 @@ eval_to_bool(arg, error, nextcmd, skip)
int skip; /* only parse, don't execute */
{
typval_T tv;
- int retval = FALSE;
+ varnumber_T retval = FALSE;
if (skip)
++emsg_skip;
@@ -1333,7 +1333,7 @@ eval_to_bool(arg, error, nextcmd, skip)
if (skip)
--emsg_skip;
- return retval;
+ return (int)retval;
}
/*
@@ -1459,12 +1459,12 @@ eval_to_string_safe(arg, nextcmd, use_sa
* Evaluates "expr" silently.
* Returns -1 for an error.
*/
- int
+ varnumber_T
eval_to_number(expr)
char_u *expr;
{
typval_T rettv;
- int retval;
+ varnumber_T retval;
char_u *p = skipwhite(expr);
++emsg_off;
@@ -1577,7 +1577,7 @@ get_spellword(list, pp)
li = li->li_next;
if (li == NULL)
return -1;
- return get_tv_number(&li->li_tv);
+ return (int)get_tv_number(&li->li_tv);
}
#endif
@@ -1620,7 +1620,7 @@ call_vim_function(func, argc, argv, safe
typval_T *rettv;
{
typval_T *argvars;
- long n;
+ varnumber_T n;
int len;
int i;
int doesrange;
@@ -1686,7 +1686,7 @@ call_vim_function(func, argc, argv, safe
* Returns -1 when calling the function fails.
* Uses argv[argc] for the function arguments.
*/
- long
+ varnumber_T
call_func_retnr(func, argc, argv, safe)
char_u *func;
int argc;
@@ -1694,7 +1694,7 @@ call_func_retnr(func, argc, argv, safe)
int safe; /* use the sandbox */
{
typval_T rettv;
- long retval;
+ varnumber_T retval;
/* All arguments are passed as strings, no conversion to number. */
if (call_vim_function(func, argc, argv, safe, TRUE, &rettv) == FAIL)
@@ -1834,7 +1834,7 @@ eval_foldexpr(arg, cp)
int *cp;
{
typval_T tv;
- int retval;
+ varnumber_T retval;
char_u *s;
int use_sandbox = was_set_insecurely((char_u *)"foldexpr",
OPT_LOCAL);
@@ -1869,7 +1869,7 @@ eval_foldexpr(arg, cp)
--sandbox;
--textlock;
- return retval;
+ return (int)retval;
}
#endif
@@ -3090,7 +3090,7 @@ tv_op(tv1, tv2, op)
typval_T *tv2;
char_u *op;
{
- long n;
+ varnumber_T n;
char_u numbuf[NUMBUFLEN];
char_u *s;
@@ -4420,7 +4420,7 @@ eval4(arg, rettv, evaluate)
exptype_T type = TYPE_UNKNOWN;
int type_is = FALSE; /* TRUE for "is" and "isnot" */
int len = 2;
- long n1, n2;
+ varnumber_T n1, n2;
char_u *s1, *s2;
char_u buf1[NUMBUFLEN], buf2[NUMBUFLEN];
regmatch_T regmatch;
@@ -4727,7 +4727,7 @@ eval5(arg, rettv, evaluate)
typval_T var2;
typval_T var3;
int op;
- long n1, n2;
+ varnumber_T n1, n2;
#ifdef FEAT_FLOAT
float_T f1 = 0, f2 = 0;
#endif
@@ -4912,7 +4912,7 @@ eval6(arg, rettv, evaluate, want_string)
{
typval_T var2;
int op;
- long n1, n2;
+ varnumber_T n1, n2;
#ifdef FEAT_FLOAT
int use_float = FALSE;
float_T f1 = 0, f2;
@@ -5033,12 +5033,21 @@ eval6(arg, rettv, evaluate, want_string)
{
if (n2 == 0) /* give an error message? */
{
+#ifdef FEAT_NUM64
+ if (n1 == 0)
+ n1 = -0x7fffffffffffffff - 1; /* similar to NaN */
+ else if (n1 < 0)
+ n1 = -0x7fffffffffffffff;
+ else
+ n1 = 0x7fffffffffffffff;
+#else
if (n1 == 0)
n1 = -0x7fffffffL - 1L; /* similar to NaN */
else if (n1 < 0)
n1 = -0x7fffffffL;
else
n1 = 0x7fffffffL;
+#endif
}
else
n1 = n1 / n2;
@@ -5092,7 +5101,7 @@ eval7(arg, rettv, evaluate, want_string)
int evaluate;
int want_string UNUSED; /* after "." operator */
{
- long n;
+ varnumber_T n;
int len;
char_u *s;
char_u *start_leader, *end_leader;
@@ -6372,7 +6381,7 @@ list_find_nr(l, idx, errorp)
*errorp = TRUE;
return -1L;
}
- return get_tv_number_chk(&li->li_tv, errorp);
+ return (long)get_tv_number_chk(&li->li_tv, errorp);
}
/*
@@ -7629,7 +7638,7 @@ get_dict_string(d, key, save)
* Get a number item from a dictionary.
* Returns 0 if the entry doesn't exist or out of memory.
*/
- long
+ varnumber_T
get_dict_number(d, key)
dict_T *d;
char_u *key;
@@ -11514,12 +11523,21 @@ f_float2nr(argvars, rettv)
if (get_float_arg(argvars, &f) == OK)
{
+# ifdef FEAT_NUM64
+ if (f < -0x7fffffffffffffff)
+ rettv->vval.v_number = -0x7fffffffffffffff;
+ else if (f > 0x7fffffffffffffff)
+ rettv->vval.v_number = 0x7fffffffffffffff;
+ else
+ rettv->vval.v_number = (varnumber_T)f;
+# else
if (f < -0x7fffffff)
rettv->vval.v_number = -0x7fffffff;
else if (f > 0x7fffffff)
rettv->vval.v_number = 0x7fffffff;
else
rettv->vval.v_number = (varnumber_T)f;
+# endif
}
}
@@ -13448,6 +13466,9 @@ f_has(argvars, rettv)
"mzscheme",
#endif
#endif
+#ifdef FEAT_NUM64
+ "num64",
+#endif
#ifdef FEAT_OLE
"ole",
#endif
@@ -17003,7 +17024,7 @@ search_cmn(argvars, match_pos, flagsp)
#ifdef FEAT_RELTIME
if (argvars[3].v_type != VAR_UNKNOWN)
{
- time_limit = get_tv_number_chk(&argvars[3], NULL);
+ time_limit = (long)get_tv_number_chk(&argvars[3], NULL);
if (time_limit < 0)
goto theend;
}
@@ -17104,8 +17125,8 @@ f_screenattr(argvars, rettv)
int col;
int c;
- row = get_tv_number_chk(&argvars[0], NULL) - 1;
- col = get_tv_number_chk(&argvars[1], NULL) - 1;
+ row = (int)get_tv_number_chk(&argvars[0], NULL) - 1;
+ col = (int)get_tv_number_chk(&argvars[1], NULL) - 1;
if (row < 0 || row >= screen_Rows
|| col < 0 || col >= screen_Columns)
c = -1;
@@ -17127,8 +17148,8 @@ f_screenchar(argvars, rettv)
int off;
int c;
- row = get_tv_number_chk(&argvars[0], NULL) - 1;
- col = get_tv_number_chk(&argvars[1], NULL) - 1;
+ row = (int)get_tv_number_chk(&argvars[0], NULL) - 1;
+ col = (int)get_tv_number_chk(&argvars[1], NULL) - 1;
if (row < 0 || row >= screen_Rows
|| col < 0 || col >= screen_Columns)
c = -1;
@@ -19080,7 +19101,7 @@ f_str2nr(argvars, rettv)
{
int base = 10;
char_u *p;
- long n;
+ varnumber_T n;
int what;
if (argvars[1].v_type != VAR_UNKNOWN)
@@ -19486,14 +19507,14 @@ f_synID(argvars, rettv)
{
int id = 0;
#ifdef FEAT_SYN_HL
- long lnum;
- long col;
+ linenr_T lnum;
+ colnr_T col;
int trans;
int transerr = FALSE;
lnum = get_tv_lnum(argvars); /* -1 on type error */
- col = get_tv_number(&argvars[1]) - 1; /* -1 on type error */
- trans = get_tv_number_chk(&argvars[2], &transerr);
+ col = (linenr_T)get_tv_number(&argvars[1]) - 1; /* -1 on type error */
+ trans = (int)get_tv_number_chk(&argvars[2], &transerr);
if (!transerr && lnum >= 1 && lnum <= curbuf->b_ml.ml_line_count
&& col >= 0 && col < (long)STRLEN(ml_get(lnum)))
@@ -19519,7 +19540,7 @@ f_synIDattr(argvars, rettv)
char_u modebuf[NUMBUFLEN];
int modec;
- id = get_tv_number(&argvars[0]);
+ id = (int)get_tv_number(&argvars[0]);
what = get_tv_string(&argvars[1]);
if (argvars[2].v_type != VAR_UNKNOWN)
{
@@ -19609,7 +19630,7 @@ f_synIDtrans(argvars, rettv)
int id;
#ifdef FEAT_SYN_HL
- id = get_tv_number(&argvars[0]);
+ id = (int)get_tv_number(&argvars[0]);
if (id > 0)
id = syn_get_final_id(id);
@@ -19629,8 +19650,8 @@ f_synconcealed(argvars, rettv)
typval_T *rettv;
{
#if defined(FEAT_SYN_HL) && defined(FEAT_CONCEAL)
- long lnum;
- long col;
+ linenr_T lnum;
+ colnr_T col;
int syntax_flags = 0;
int cchar;
int matchid = 0;
@@ -19642,7 +19663,7 @@ f_synconcealed(argvars, rettv)
#if defined(FEAT_SYN_HL) && defined(FEAT_CONCEAL)
lnum = get_tv_lnum(argvars); /* -1 on type error */
- col = get_tv_number(&argvars[1]) - 1; /* -1 on type error */
+ col = (colnr_T)get_tv_number(&argvars[1]) - 1; /* -1 on type error */
vim_memset(str, NUL, sizeof(str));
@@ -19691,8 +19712,8 @@ f_synstack(argvars, rettv)
typval_T *rettv;
{
#ifdef FEAT_SYN_HL
- long lnum;
- long col;
+ linenr_T lnum;
+ colnr_T col;
int i;
int id;
#endif
@@ -19702,7 +19723,7 @@ f_synstack(argvars, rettv)
#ifdef FEAT_SYN_HL
lnum = get_tv_lnum(argvars); /* -1 on type error */
- col = get_tv_number(&argvars[1]) - 1; /* -1 on type error */
+ col = (colnr_T)get_tv_number(&argvars[1]) - 1; /* -1 on type error */
if (lnum >= 1 && lnum <= curbuf->b_ml.ml_line_count
&& col >= 0 && col <= (long)STRLEN(ml_get(lnum))
@@ -20681,29 +20702,29 @@ f_winrestview(argvars, rettv)
else
{
if (dict_find(dict, (char_u *)"lnum", -1) != NULL)
- curwin->w_cursor.lnum = get_dict_number(dict, (char_u *)"lnum");
+ curwin->w_cursor.lnum = (linenr_T)get_dict_number(dict, (char_u *)"lnum");
if (dict_find(dict, (char_u *)"col", -1) != NULL)
- curwin->w_cursor.col = get_dict_number(dict, (char_u *)"col");
+ curwin->w_cursor.col = (colnr_T)get_dict_number(dict, (char_u *)"col");
#ifdef FEAT_VIRTUALEDIT
if (dict_find(dict, (char_u *)"coladd", -1) != NULL)
- curwin->w_cursor.coladd = get_dict_number(dict, (char_u *)"coladd");
+ curwin->w_cursor.coladd = (colnr_T)get_dict_number(dict, (char_u *)"coladd");
#endif
if (dict_find(dict, (char_u *)"curswant", -1) != NULL)
{
- curwin->w_curswant = get_dict_number(dict, (char_u *)"curswant");
+ curwin->w_curswant = (colnr_T)get_dict_number(dict, (char_u *)"curswant");
curwin->w_set_curswant = FALSE;
}
if (dict_find(dict, (char_u *)"topline", -1) != NULL)
- set_topline(curwin, get_dict_number(dict, (char_u *)"topline"));
+ set_topline(curwin, (linenr_T)get_dict_number(dict, (char_u *)"topline"));
#ifdef FEAT_DIFF
if (dict_find(dict, (char_u *)"topfill", -1) != NULL)
- curwin->w_topfill = get_dict_number(dict, (char_u *)"topfill");
+ curwin->w_topfill = (int)get_dict_number(dict, (char_u *)"topfill");
#endif
if (dict_find(dict, (char_u *)"leftcol", -1) != NULL)
- curwin->w_leftcol = get_dict_number(dict, (char_u *)"leftcol");
+ curwin->w_leftcol = (colnr_T)get_dict_number(dict, (char_u *)"leftcol");
if (dict_find(dict, (char_u *)"skipcol", -1) != NULL)
- curwin->w_skipcol = get_dict_number(dict, (char_u *)"skipcol");
+ curwin->w_skipcol = (colnr_T)get_dict_number(dict, (char_u *)"skipcol");
check_cursor();
win_new_height(curwin, curwin->w_height);
@@ -21384,7 +21405,7 @@ eval_isnamec1(c)
void
set_vim_var_nr(idx, val)
int idx;
- long val;
+ varnumber_T val;
{
vimvars[idx].vv_nr = val;
}
@@ -21392,7 +21413,7 @@ set_vim_var_nr(idx, val)
/*
* Get number v: variable value.
*/
- long
+ varnumber_T
get_vim_var_nr(idx)
int idx;
{
@@ -21924,7 +21945,7 @@ init_tv(varp)
* caller of incompatible types: it sets *denote to TRUE if "denote"
* is not NULL or returns -1 otherwise.
*/
- static long
+ static varnumber_T
get_tv_number(varp)
typval_T *varp;
{
@@ -21933,17 +21954,17 @@ get_tv_number(varp)
return get_tv_number_chk(varp, &error); /* return 0L on error */
}
- long
+ varnumber_T
get_tv_number_chk(varp, denote)
typval_T *varp;
int *denote;
{
- long n = 0L;
+ varnumber_T n = 0L;
switch (varp->v_type)
{
case VAR_NUMBER:
- return (long)(varp->vval.v_number);
+ return varp->vval.v_number;
#ifdef FEAT_FLOAT
case VAR_FLOAT:
EMSG(_("E805: Using a Float as a Number"));
@@ -22023,12 +22044,12 @@ get_tv_lnum(argvars)
typval_T rettv;
linenr_T lnum;
- lnum = get_tv_number_chk(&argvars[0], NULL);
+ lnum = (linenr_T)get_tv_number_chk(&argvars[0], NULL);
if (lnum == 0) /* no valid number, try using line() */
{
rettv.v_type = VAR_NUMBER;
f_line(argvars, &rettv);
- lnum = rettv.vval.v_number;
+ lnum = (linenr_T)rettv.vval.v_number;
clear_tv(&rettv);
}
return lnum;
@@ -22049,7 +22070,7 @@ get_tv_lnum_buf(argvars, buf)
&& argvars[0].vval.v_string[0] == '$'
&& buf != NULL)
return buf->b_ml.ml_line_count;
- return get_tv_number_chk(&argvars[0], NULL);
+ return (linenr_T) get_tv_number_chk(&argvars[0], NULL);
}
/*
@@ -22101,7 +22122,8 @@ get_tv_string_buf_chk(varp, buf)
switch (varp->v_type)
{
case VAR_NUMBER:
- sprintf((char *)buf, "%ld", (long)varp->vval.v_number);
+ sprintf((char *)buf, "%"VARNUMBER_FMT"d",
+ (varnumber_T)varp->vval.v_number);
return buf;
case VAR_FUNC:
EMSG(_("E729: using Funcref as a String"));
diff --git a/src/ex_cmds.c b/src/ex_cmds.c
--- a/src/ex_cmds.c
+++ b/src/ex_cmds.c
@@ -295,10 +295,10 @@ typedef struct
union {
struct
{
- long start_col_nr; /* starting column number */
- long end_col_nr; /* ending column number */
+ varnumber_T start_col_nr; /* starting column number */
+ varnumber_T end_col_nr; /* ending column number */
} line;
- long value; /* value if sorting by integer */
+ varnumber_T value; /* value if sorting by integer */
#ifdef FEAT_FLOAT
float_T value_flt; /* value if sorting by float */
#endif
diff --git a/src/ex_getln.c b/src/ex_getln.c
--- a/src/ex_getln.c
+++ b/src/ex_getln.c
@@ -5930,7 +5930,7 @@ get_list_range(str, num1, num2)
{
int len;
int first = FALSE;
- long num;
+ varnumber_T num;
*str = skipwhite(*str);
if (**str == '-' || vim_isdigit(**str)) /* parse "from" part of range */
diff --git a/src/feature.h b/src/feature.h
--- a/src/feature.h
+++ b/src/feature.h
@@ -382,12 +382,16 @@
* +eval Built-in script language and expression evaluation,
* ":let", ":if", etc.
* +float Floating point variables.
+ * +num64 64-bit Number.
*/
#ifdef FEAT_NORMAL
# define FEAT_EVAL
# if defined(HAVE_FLOAT_FUNCS) || defined(WIN3264) || defined(MACOS)
# define FEAT_FLOAT
# endif
+# if defined(HAVE_STDINT_H) || defined(WIN3264) || (VIM_SIZEOF_LONG >= 8)
+# define FEAT_NUM64
+# endif
#endif
/*
diff --git a/src/fileio.c b/src/fileio.c
--- a/src/fileio.c
+++ b/src/fileio.c
@@ -9535,7 +9535,7 @@ apply_autocmds_group(event, fname, fname
#ifdef FEAT_EVAL
/* set v:cmdarg (only when there is a matching pattern) */
- save_cmdbang = get_vim_var_nr(VV_CMDBANG);
+ save_cmdbang = (long)get_vim_var_nr(VV_CMDBANG);
if (eap != NULL)
{
save_cmdarg = set_cmdarg(eap, NULL);
diff --git a/src/fold.c b/src/fold.c
--- a/src/fold.c
+++ b/src/fold.c
@@ -3108,7 +3108,7 @@ foldlevelExpr(flp)
/* KeyTyped may be reset to 0 when calling a function which invokes
* do_cmdline(). To make 'foldopen' work correctly restore KeyTyped. */
save_keytyped = KeyTyped;
- n = eval_foldexpr(flp->wp->w_p_fde, &c);
+ n = (int)eval_foldexpr(flp->wp->w_p_fde, &c);
KeyTyped = save_keytyped;
switch (c)
diff --git a/src/json.c b/src/json.c
--- a/src/json.c
+++ b/src/json.c
@@ -114,8 +114,8 @@ json_encode_item(garray_T *gap, typval_T
break;
case VAR_NUMBER:
- vim_snprintf((char *)numbuf, NUMBUFLEN, "%ld",
- (long)val->vval.v_number);
+ vim_snprintf((char *)numbuf, NUMBUFLEN, "%lld",
+ val->vval.v_number);
ga_concat(gap, numbuf);
break;
@@ -361,7 +361,7 @@ json_decode_string(js_read_T *reader, ty
int len;
char_u *p = reader->js_buf + reader->js_used + 1;
int c;
- long nr;
+ varnumber_T nr;
char_u buf[NUMBUFLEN];
ga_init2(&ga, 1, 200);
@@ -387,7 +387,7 @@ json_decode_string(js_read_T *reader, ty
buf[(*mb_char2bytes)((int)nr, buf)] = NUL;
ga_concat(&ga, buf);
#else
- ga_append(&ga, nr);
+ ga_append(&ga, (int)nr);
#endif
break;
default: c = p[1]; break;
@@ -480,7 +480,7 @@ json_decode_item(js_read_T *reader, typv
else
#endif
{
- long nr;
+ varnumber_T nr;
res->v_type = VAR_NUMBER;
vim_str2nr(reader->js_buf + reader->js_used,
diff --git a/src/message.c b/src/message.c
--- a/src/message.c
+++ b/src/message.c
@@ -3882,7 +3882,7 @@ do_browse(flags, title, dflt, ext, initd
#if defined(FEAT_EVAL)
static char *e_printf = N_("E766: Insufficient arguments for printf()");
-static long tv_nr(typval_T *tvs, int *idxp);
+static varnumber_T tv_nr(typval_T *tvs, int *idxp);
static char *tv_str(typval_T *tvs, int *idxp);
# ifdef FEAT_FLOAT
static double tv_float(typval_T *tvs, int *idxp);
@@ -3891,13 +3891,13 @@ static double tv_float(typval_T *tvs, in
/*
* Get number argument from "idxp" entry in "tvs". First entry is 1.
*/
- static long
+ static varnumber_T
tv_nr(tvs, idxp)
typval_T *tvs;
int *idxp;
{
int idx = *idxp - 1;
- long n = 0;
+ varnumber_T n = 0;
int err = FALSE;
if (tvs[idx].v_type == VAR_UNKNOWN)
@@ -3954,7 +3954,7 @@ tv_float(tvs, idxp)
if (tvs[idx].v_type == VAR_FLOAT)
f = tvs[idx].vval.v_float;
else if (tvs[idx].v_type == VAR_NUMBER)
- f = tvs[idx].vval.v_number;
+ f = (double)tvs[idx].vval.v_number;
else
EMSG(_("E807: Expected Float argument for printf()"));
}
@@ -4212,7 +4212,11 @@ vim_vsnprintf(str, str_m, fmt, ap, tvs)
if (length_modifier == 'l' && *p == 'l')
{
/* double l = long long */
+# ifdef FEAT_NUM64
+ length_modifier = 'L';
+# else
length_modifier = 'l'; /* treat it as a single 'l' */
+# endif
p++;
}
}
@@ -4341,6 +4345,12 @@ vim_vsnprintf(str, str_m, fmt, ap, tvs)
long int long_arg = 0;
unsigned long int ulong_arg = 0;
+# ifdef FEAT_NUM64
+ /* only defined for length modifier ll */
+ varnumber_T llong_arg = 0;
+ uvarnumber_T ullong_arg = 0;
+# endif
+
/* pointer argument value -only defined for p
* conversion */
void *ptr_arg = NULL;
@@ -4385,6 +4395,19 @@ vim_vsnprintf(str, str_m, fmt, ap, tvs)
else if (long_arg < 0)
arg_sign = -1;
break;
+# ifdef FEAT_NUM64
+ case 'L':
+ llong_arg =
+# if defined(FEAT_EVAL)
+ tvs != NULL ? tv_nr(tvs, &arg_idx) :
+# endif
+ va_arg(ap, varnumber_T);
+ if (llong_arg > 0)
+ arg_sign = 1;
+ else if (llong_arg < 0)
+ arg_sign = -1;
+ break;
+# endif
}
}
else
@@ -4413,6 +4436,18 @@ vim_vsnprintf(str, str_m, fmt, ap, tvs)
if (ulong_arg != 0)
arg_sign = 1;
break;
+# ifdef FEAT_NUM64
+ case 'L':
+ ullong_arg =
+# if defined(FEAT_EVAL)
+ tvs != NULL ? (uvarnumber_T)
+ tv_nr(tvs, &arg_idx) :
+# endif
+ va_arg(ap, uvarnumber_T);
+ if (ullong_arg != 0)
+ arg_sign = 1;
+ break;
+# endif
}
}
@@ -4457,17 +4492,27 @@ vim_vsnprintf(str, str_m, fmt, ap, tvs)
}
else
{
- char f[5];
+ char f[6];
int f_l = 0;
/* construct a simple format string for sprintf */
f[f_l++] = '%';
if (!length_modifier)
;
- else if (length_modifier == '2')
+ else if (length_modifier == 'L')
{
+# ifdef FEAT_NUM64
+# ifdef WIN3264
+ f[f_l++] = 'I';
+ f[f_l++] = '6';
+ f[f_l++] = '4';
+# else
f[f_l++] = 'l';
f[f_l++] = 'l';
+# endif
+# else
+ f[f_l++] = 'l';
+# endif
}
else
f[f_l++] = length_modifier;
@@ -4488,6 +4533,11 @@ vim_vsnprintf(str, str_m, fmt, ap, tvs)
case 'l': str_arg_l += sprintf(
tmp + str_arg_l, f, long_arg);
break;
+# ifdef FEAT_NUM64
+ case 'L': str_arg_l += sprintf(
+ tmp + str_arg_l, f, llong_arg);
+ break;
+# endif
}
}
else
@@ -4502,6 +4552,11 @@ vim_vsnprintf(str, str_m, fmt, ap, tvs)
case 'l': str_arg_l += sprintf(
tmp + str_arg_l, f, ulong_arg);
break;
+# ifdef FEAT_NUM64
+ case 'L': str_arg_l += sprintf(
+ tmp + str_arg_l, f, ullong_arg);
+ break;
+# endif
}
}
diff --git a/src/misc1.c b/src/misc1.c
--- a/src/misc1.c
+++ b/src/misc1.c
@@ -9309,7 +9309,7 @@ get_expr_indent()
if (use_sandbox)
++sandbox;
++textlock;
- indent = eval_to_number(curbuf->b_p_inde);
+ indent = (int)eval_to_number(curbuf->b_p_inde);
if (use_sandbox)
--sandbox;
--textlock;
diff --git a/src/misc2.c b/src/misc2.c
--- a/src/misc2.c
+++ b/src/misc2.c
@@ -2820,7 +2820,7 @@ find_special_key(srcp, modp, keycode, ke
int modifiers;
int bit;
int key;
- unsigned long n;
+ uvarnumber_T n;
int l;
src = *srcp;
diff --git a/src/ops.c b/src/ops.c
--- a/src/ops.c
+++ b/src/ops.c
@@ -4778,7 +4778,7 @@ fex_format(lnum, count, c)
*/
if (use_sandbox)
++sandbox;
- r = eval_to_number(curbuf->b_p_fex);
+ r = (int)eval_to_number(curbuf->b_p_fex);
if (use_sandbox)
--sandbox;
diff --git a/src/option.c b/src/option.c
--- a/src/option.c
+++ b/src/option.c
@@ -4192,7 +4192,7 @@ do_set(arg, opt_flags)
int afterchar; /* character just after option name */
int len;
int i;
- long value;
+ varnumber_T value;
int key;
long_u flags; /* flags for current option */
char_u *varp = NULL; /* pointer to variable for current option */
@@ -9153,7 +9153,7 @@ get_option_value(name, numval, stringval
if ((int *)varp == &curbuf->b_changed)
*numval = curbufIsChanged();
else
- *numval = *(int *)varp;
+ *numval = (long) *(varnumber_T *)varp;
}
return 1;
}
diff --git a/src/proto/charset.pro b/src/proto/charset.pro
--- a/src/proto/charset.pro
+++ b/src/proto/charset.pro
@@ -52,7 +52,7 @@ char_u *skiptowhite(char_u *p);
char_u *skiptowhite_esc(char_u *p);
long getdigits(char_u **pp);
int vim_isblankline(char_u *lbuf);
-void vim_str2nr(char_u *start, int *prep, int *len, int what, long *nptr, unsigned long *unptr, int maxlen);
+void vim_str2nr(char_u *start, int *prep, int *len, int what, varnumber_T *nptr, uvarnumber_T *unptr, int maxlen);
int hex2nr(int c);
int hexhex2nr(char_u *p);
int rem_backslash(char_u *str);
diff --git a/src/proto/eval.pro b/src/proto/eval.pro
--- a/src/proto/eval.pro
+++ b/src/proto/eval.pro
@@ -19,12 +19,12 @@ char_u *eval_to_string_skip(char_u *arg,
int skip_expr(char_u **pp);
char_u *eval_to_string(char_u *arg, char_u **nextcmd, int convert);
char_u *eval_to_string_safe(char_u *arg, char_u **nextcmd, int use_sandbox);
-int eval_to_number(char_u *expr);
+varnumber_T eval_to_number(char_u *expr);
list_T *eval_spell_expr(char_u *badword, char_u *expr);
int get_spellword(list_T *list, char_u **pp);
typval_T *eval_expr(char_u *arg, char_u **nextcmd);
int call_vim_function(char_u *func, int argc, char_u **argv, int safe, int str_arg_only, typval_T *rettv);
-long call_func_retnr(char_u *func, int argc, char_u **argv, int safe);
+varnumber_T call_func_retnr(char_u *func, int argc, char_u **argv, int safe);
void *call_func_retstr(char_u *func, int argc, char_u **argv, int safe);
void *call_func_retlist(char_u *func, int argc, char_u **argv, int safe);
void *save_funccal(void);
@@ -78,7 +78,7 @@ int dict_add_nr_str(dict_T *d, char *key
int dict_add_list(dict_T *d, char *key, list_T *list);
dictitem_T *dict_find(dict_T *d, char_u *key, int len);
char_u *get_dict_string(dict_T *d, char_u *key, int save);
-long get_dict_number(dict_T *d, char_u *key);
+varnumber_T get_dict_number(dict_T *d, char_u *key);
int string2float(char_u *text, float_T *value);
char_u *get_function_name(expand_T *xp, int idx);
char_u *get_expr_name(expand_T *xp, int idx);
@@ -88,8 +88,8 @@ void dict_extend(dict_T *d1, dict_T *d2,
void mzscheme_call_vim(char_u *name, typval_T *args, typval_T *rettv);
float_T vim_round(float_T f);
long do_searchpair(char_u *spat, char_u *mpat, char_u *epat, int dir, char_u *skip, int flags, pos_T *match_pos, linenr_T lnum_stop, long time_limit);
-void set_vim_var_nr(int idx, long val);
-long get_vim_var_nr(int idx);
+void set_vim_var_nr(int idx, varnumber_T val);
+varnumber_T get_vim_var_nr(int idx);
char_u *get_vim_var_str(int idx);
list_T *get_vim_var_list(int idx);
void set_vim_var_char(int c);
@@ -103,7 +103,7 @@ char_u *v_throwpoint(char_u *oldval);
char_u *set_cmdarg(exarg_T *eap, char_u *oldarg);
void free_tv(typval_T *varp);
void clear_tv(typval_T *varp);
-long get_tv_number_chk(typval_T *varp, int *denote);
+varnumber_T get_tv_number_chk(typval_T *varp, int *denote);
char_u *get_tv_string_chk(typval_T *varp);
char_u *get_tv_string_buf_chk(typval_T *varp, char_u *buf);
char_u *get_var_value(char_u *name);
diff --git a/src/quickfix.c b/src/quickfix.c
--- a/src/quickfix.c
+++ b/src/quickfix.c
@@ -4068,11 +4068,11 @@ set_errorlist(wp, list, action, title)
continue;
filename = get_dict_string(d, (char_u *)"filename", TRUE);
- bufnum = get_dict_number(d, (char_u *)"bufnr");
- lnum = get_dict_number(d, (char_u *)"lnum");
- col = get_dict_number(d, (char_u *)"col");
- vcol = get_dict_number(d, (char_u *)"vcol");
- nr = get_dict_number(d, (char_u *)"nr");
+ bufnum = (int)get_dict_number(d, (char_u *)"bufnr");
+ lnum = (int)get_dict_number(d, (char_u *)"lnum");
+ col = (int)get_dict_number(d, (char_u *)"col");
+ vcol = (int)get_dict_number(d, (char_u *)"vcol");
+ nr = (int)get_dict_number(d, (char_u *)"nr");
type = get_dict_string(d, (char_u *)"type", TRUE);
pattern = get_dict_string(d, (char_u *)"pattern", TRUE);
text = get_dict_string(d, (char_u *)"text", TRUE);
diff --git a/src/structs.h b/src/structs.h
--- a/src/structs.h
+++ b/src/structs.h
@@ -1101,11 +1101,34 @@ typedef struct hashtable_S
typedef long_u hash_T; /* Type for hi_hash */
-#if VIM_SIZEOF_INT <= 3 /* use long if int is smaller than 32 bits */
-typedef long varnumber_T;
+#ifdef FEAT_NUM64
+/* Use 64-bit Number. */
+# ifdef HAVE_STDINT_H
+typedef int64_t varnumber_T;
+typedef uint64_t uvarnumber_T;
+# define VARNUMBER_FMT "ll"
+# elif defined(WIN3264)
+typedef __int64 varnumber_T;
+typedef unsigned __int64 uvarnumber_T;
+# define VARNUMBER_FMT "I64"
+# else
+typedef long varnumber_T;
+typedef unsigned long uvarnumber_T;
+# define VARNUMBER_FMT "l"
+# endif
#else
-typedef int varnumber_T;
+/* Use 32-bit Number. */
+# if VIM_SIZEOF_INT <= 3 /* use long if int is smaller than 32 bits */
+typedef long varnumber_T;
+typedef unsigned long uvarnumber_T;
+# define VARNUMBER_FMT "l"
+# else
+typedef int varnumber_T;
+typedef unsigned int uvarnumber_T;
+# define VARNUMBER_FMT ""
+# endif
#endif
+
typedef double float_T;
typedef struct listvar_S list_T;
diff --git a/src/version.c b/src/version.c
--- a/src/version.c
+++ b/src/version.c
@@ -454,6 +454,11 @@ static char *(features[]) =
#else
"-netbeans_intg",
#endif
+#ifdef FEAT_NUM64
+ "+num64",
+#else
+ "-num64",
+#endif
#ifdef FEAT_GUI_W32
# ifdef FEAT_OLE
"+ole",
# HG changeset patch
# Parent 7122a1b0a7aa78d56f2547986edfb79eceaddc36
diff --git a/src/ops.c b/src/ops.c
--- a/src/ops.c
+++ b/src/ops.c
@@ -5475,8 +5475,8 @@ do_addsub(op_type, pos, length, Prenum1)
char_u buf2[NUMBUFLEN];
int pre; /* 'X'/'x': hex; '0': octal; 'B'/'b': bin */
static int hexupper = FALSE; /* 0xABC */
- unsigned long n;
- long_u oldn;
+ uvarnumber_T n;
+ uvarnumber_T oldn;
char_u *ptr;
int c;
int todel;
@@ -5682,9 +5682,9 @@ do_addsub(op_type, pos, length, Prenum1)
oldn = n;
if (subtract)
- n -= (unsigned long)Prenum1;
+ n -= (uvarnumber_T)Prenum1;
else
- n += (unsigned long)Prenum1;
+ n += (uvarnumber_T)Prenum1;
/* handle wraparound for decimal numbers */
if (!pre)
{
@@ -5692,7 +5692,7 @@ do_addsub(op_type, pos, length, Prenum1)
{
if (n > oldn)
{
- n = 1 + (n ^ (unsigned long)-1);
+ n = 1 + (n ^ (uvarnumber_T)-1);
negative ^= TRUE;
}
}
@@ -5701,7 +5701,7 @@ do_addsub(op_type, pos, length, Prenum1)
/* add */
if (n < oldn)
{
- n = (n ^ (unsigned long)-1);
+ n = (n ^ (uvarnumber_T)-1);
negative ^= TRUE;
}
}
@@ -5777,7 +5777,7 @@ do_addsub(op_type, pos, length, Prenum1)
{
int i;
int bit = 0;
- int bits = sizeof(unsigned long) * 8;
+ int bits = sizeof(uvarnumber_T) * 8;
/* leading zeros */
for (bit = bits; bit > 0; bit--)
@@ -5789,13 +5789,13 @@ do_addsub(op_type, pos, length, Prenum1)
buf2[i] = '\0';
}
else if (pre == 0)
- sprintf((char *)buf2, "%lu", n);
+ sprintf((char *)buf2, "%"VARNUMBER_FMT"u", n);
else if (pre == '0')
- sprintf((char *)buf2, "%lo", n);
+ sprintf((char *)buf2, "%"VARNUMBER_FMT"o", n);
else if (pre && hexupper)
- sprintf((char *)buf2, "%lX", n);
+ sprintf((char *)buf2, "%"VARNUMBER_FMT"X", n);
else
- sprintf((char *)buf2, "%lx", n);
+ sprintf((char *)buf2, "%"VARNUMBER_FMT"x", n);
length -= (int)STRLEN(buf2);
/*
# HG changeset patch
# Parent 26d10cf05f1a4e7109dabbebbb60be3d234e824c
diff --git a/runtime/doc/eval.txt b/runtime/doc/eval.txt
--- a/runtime/doc/eval.txt
+++ b/runtime/doc/eval.txt
@@ -40,6 +40,8 @@ 1.1 Variable types ~
There are six types of variables:
Number A 32 or 64 bit signed number. |expr-number| *Number*
+ 64-bit Number is avaliable only when compiled with the
+ |+num64| feature.
Examples: -123 0x10 0177
Float A floating point number. |floating-point-format| *Float*
@@ -836,6 +838,11 @@ When dividing a Number by zero the resul
<0 / 0 = -0x7fffffff (like negative infinity)
(before Vim 7.2 it was always 0x7fffffff)
+When 64-bit Number support is enabled:
+ 0 / 0 = -0x8000000000000000 (like NaN for Float)
+ >0 / 0 = 0x7fffffffffffffff (like positive infinity)
+ <0 / 0 = -0x7fffffffffffffff (like negative infinity)
+
When the righthand side of '%' is zero, the result is 0.
None of these work for |Funcref|s.
@@ -3213,8 +3220,10 @@ float2nr({expr}) *float2nr()*
decimal point.
{expr} must evaluate to a |Float| or a Number.
When the value of {expr} is out of range for a |Number| the
- result is truncated to 0x7fffffff or -0x7fffffff. NaN results
- in -0x80000000.
+ result is truncated to 0x7fffffff or -0x7fffffff (or when
+ 64-bit Number support is enabled, 0x7fffffffffffffff or
+ -0x7fffffffffffffff. NaN results in -0x80000000 (or when
+ 64-bit Number support is enabled, -0x8000000000000000).
Examples: >
echo float2nr(3.95)
< 3 >
@@ -7085,6 +7094,7 @@ multi_lang Compiled with support for mu
mzscheme Compiled with MzScheme interface |mzscheme|.
netbeans_enabled Compiled with support for |netbeans| and connected.
netbeans_intg Compiled with support for |netbeans|.
+num64 Compiled with 64-bit |Number| support.
ole Compiled with OLE automation support for Win32.
os2 OS/2 version of Vim.
path_extra Compiled with up/downwards search in 'path' and 'tags'
diff --git a/runtime/doc/various.txt b/runtime/doc/various.txt
--- a/runtime/doc/various.txt
+++ b/runtime/doc/various.txt
@@ -393,6 +393,7 @@ N *+multi_lang* non-English language su
m *+mzscheme* Mzscheme interface |mzscheme|
m *+mzscheme/dyn* Mzscheme interface |mzscheme-dynamic| |/dyn|
m *+netbeans_intg* |netbeans|
+ *+num64* 64-bit Number support |Number|
m *+ole* Win32 GUI only: |ole-interface|
N *+path_extra* Up/downwards search in 'path' and 'tags'
m *+perl* Perl interface |perl|
