vim-patch:8.1.1122: char2nr() does not handle composing characters

Problem:    char2nr() does not handle composing characters.
Solution:   Add str2list() and list2str(). (Ozaki Kiichi, closes vim/vim#4190)
9d40128afd

'utf8' optional param is noop unlike Vim.
This commit is contained in:
Jan Edmund Lazo 2020-02-20 22:53:38 -05:00
parent 13b6f7a806
commit 7ed4837298
No known key found for this signature in database
GPG Key ID: 64915E6E9F735B15
5 changed files with 119 additions and 2 deletions

View File

@ -2240,6 +2240,7 @@ libcallnr({lib}, {func}, {arg}) Number idem, but return a Number
line({expr}) Number line nr of cursor, last line or mark
line2byte({lnum}) Number byte count of line {lnum}
lispindent({lnum}) Number Lisp indent for line {lnum}
list2str({list} [, {utf8}]) String turn numbers in {list} into a String
localtime() Number current time
log({expr}) Float natural logarithm (base e) of {expr}
log10({expr}) Float logarithm of Float {expr} to base 10
@ -2394,6 +2395,8 @@ sqrt({expr}) Float square root of {expr}
stdioopen({dict}) Number open stdio in a headless instance.
stdpath({what}) String/List returns the standard path(s) for {what}
str2float({expr}) Float convert String to Float
str2list({expr} [, {utf8}]) List convert each character of {expr} to
ASCII/UTF8 value
str2nr({expr} [, {base}]) Number convert String to Number
strchars({expr} [, {skipcc}]) Number character length of the String {expr}
strcharpart({str}, {start} [, {len}])
@ -5687,6 +5690,20 @@ lispindent({lnum}) *lispindent()*
When {lnum} is invalid or Vim was not compiled the
|+lispindent| feature, -1 is returned.
list2str({list} [, {utf8}]) *list2str()*
Convert each number in {list} to a character string can
concatenate them all. Examples: >
list2str([32]) returns " "
list2str([65, 66, 67]) returns "ABC"
< The same can be done (slowly) with: >
join(map(list, {nr, val -> nr2char(val)}), '')
< |str2list()| does the opposite.
When {utf8} is omitted or zero, the current 'encoding' is used.
With {utf8} is 1, always return utf-8 characters.
With utf-8 composing characters work as expected: >
list2str([97, 769]) returns "á"
<
localtime() *localtime()*
Return the current time, measured as seconds since 1st Jan
1970. See also |strftime()| and |getftime()|.
@ -8197,6 +8214,18 @@ str2float({expr}) *str2float()*
|substitute()|: >
let f = str2float(substitute(text, ',', '', 'g'))
str2list({expr} [, {utf8}]) *str2list()*
Return a list containing the number values which represent
each character in String {expr}. Examples: >
str2list(" ") returns [32]
str2list("ABC") returns [65, 66, 67]
< |list2str()| does the opposite.
When {utf8} is omitted or zero, the current 'encoding' is used.
With {utf8} set to 1, always treat the String as utf-8
characters. With utf-8 composing characters are handled
properly: >
str2list("á") returns [97, 769]
str2nr({expr} [, {base}]) *str2nr()*
Convert string {expr} to a number.

View File

@ -578,8 +578,10 @@ used for. You can find an alphabetical list here: |functions|. Use CTRL-] on
the function name to jump to detailed help on it.
String manipulation: *string-functions*
nr2char() get a character by its ASCII value
char2nr() get ASCII value of a character
nr2char() get a character by its number value
list2str() get a character string from a list of numbers
char2nr() get number value of a character
str2list() get list of numbers from a string
str2nr() convert a string to a Number
str2float() convert a string to a Float
printf() format a string according to % items

View File

@ -214,6 +214,7 @@ return {
line={args=1},
line2byte={args=1},
lispindent={args=1},
list2str={args={1, 2}},
localtime={},
log={args=1, func="float_op_wrapper", data="&log"},
log10={args=1, func="float_op_wrapper", data="&log10"},
@ -321,6 +322,7 @@ return {
sqrt={args=1, func="float_op_wrapper", data="&sqrt"},
stdpath={args=1},
str2float={args=1},
str2list={args={1, 2}},
str2nr={args={1, 2}},
strcharpart={args={2, 3}},
strchars={args={1,2}},

View File

@ -5246,6 +5246,36 @@ static void f_lispindent(typval_T *argvars, typval_T *rettv, FunPtr fptr)
}
}
// "list2str()" function
static void f_list2str(typval_T *argvars, typval_T *rettv, FunPtr fptr)
{
garray_T ga;
rettv->v_type = VAR_STRING;
rettv->vval.v_string = NULL;
if (argvars[0].v_type != VAR_LIST) {
EMSG(_(e_invarg));
return;
}
list_T *const l = argvars[0].vval.v_list;
if (l == NULL) {
return; // empty list results in empty string
}
ga_init(&ga, 1, 80);
char_u buf[MB_MAXBYTES + 1];
TV_LIST_ITER_CONST(l, li, {
buf[utf_char2bytes(tv_get_number(TV_LIST_ITEM_TV(li)), buf)] = NUL;
ga_concat(&ga, buf);
});
ga_append(&ga, NUL);
rettv->v_type = VAR_STRING;
rettv->vval.v_string = ga.ga_data;
}
/*
* "localtime()" function
*/
@ -9353,6 +9383,17 @@ static void f_str2float(typval_T *argvars, typval_T *rettv, FunPtr fptr)
rettv->v_type = VAR_FLOAT;
}
// "str2list()" function
static void f_str2list(typval_T *argvars, typval_T *rettv, FunPtr fptr)
{
tv_list_alloc_ret(rettv, kListLenUnknown);
const char_u *p = (const char_u *)tv_get_string(&argvars[0]);
for (; *p != NUL; p += utf_ptr2len(p)) {
tv_list_append_number(rettv->vval.v_list, utf_ptr2char(p));
}
}
// "str2nr()" function
static void f_str2nr(typval_T *argvars, typval_T *rettv, FunPtr fptr)
{

View File

@ -60,3 +60,46 @@ func Test_getvcol()
call assert_equal(2, virtcol("'["))
call assert_equal(2, virtcol("']"))
endfunc
func Test_list2str_str2list_utf8()
" One Unicode codepoint
let s = "\u3042\u3044"
let l = [0x3042, 0x3044]
call assert_equal(l, str2list(s, 1))
call assert_equal(s, list2str(l, 1))
if &enc ==# 'utf-8'
call assert_equal(str2list(s), str2list(s, 1))
call assert_equal(list2str(l), list2str(l, 1))
endif
" With composing characters
let s = "\u304b\u3099\u3044"
let l = [0x304b, 0x3099, 0x3044]
call assert_equal(l, str2list(s, 1))
call assert_equal(s, list2str(l, 1))
if &enc ==# 'utf-8'
call assert_equal(str2list(s), str2list(s, 1))
call assert_equal(list2str(l), list2str(l, 1))
endif
" Null list is the same as an empty list
call assert_equal('', list2str([]))
" call assert_equal('', list2str(test_null_list()))
endfunc
func Test_list2str_str2list_latin1()
" When 'encoding' is not multi-byte can still get utf-8 string.
" But we need to create the utf-8 string while 'encoding' is utf-8.
let s = "\u3042\u3044"
let l = [0x3042, 0x3044]
let save_encoding = &encoding
" set encoding=latin1
let lres = str2list(s, 1)
let sres = list2str(l, 1)
let &encoding = save_encoding
call assert_equal(l, lres)
call assert_equal(s, sres)
endfunc