d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

sort: replace home made line reader by getdelim(3)

Browse Source 
The previous code had bug when reading lines with an unexpected
encoding, returning without the full line being captured.
This result in sort complaining with "sort: Illegal byte sequence"

Using getdelim(3) instead of the home made code, fixes the situation.

PR:		241679
Reported by:	Ronald F. Guilmette <rfg-freebsd@tristatelogic.com>
MFC After:	1 week
Reviewed by:	markj, imp
Differential Revision:	https://reviews.freebsd.org/D36948
This commit is contained in:
Baptiste Daroussin 2022-10-12 15:31:50 +02:00
parent 45c11d5401
commit b58094c0d9
2 changed files with 12 additions and 115 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

109
usr.bin/sort/bwstring.c
View File

@ -470,115 +470,6 @@ bwsfwrite(struct bwstring *bws, FILE *f, bool zero_ended)
}
}
/*
* Allocate and read a binary string from file.
* The strings are nl-ended or zero-ended, depending on the sort setting.
*/
struct bwstring *
bwsfgetln(FILE *f, size_t *len, bool zero_ended, struct reader_buffer *rb)
{
wint_t eols;
eols = zero_ended ? btowc('\0') : btowc('\n');
if (!zero_ended && (mb_cur_max > 1)) {
wchar_t *ret;
ret = fgetwln(f, len);
if (ret == NULL) {
if (!feof(f))
err(2, NULL);
return (NULL);
}
if (*len > 0) {
if (ret[*len - 1] == (wchar_t)eols)
--(*len);
}
return (bwssbdup(ret, *len));
} else if (!zero_ended && (mb_cur_max == 1)) {
char *ret;
ret = fgetln(f, len);
if (ret == NULL) {
if (!feof(f))
err(2, NULL);
return (NULL);
}
if (*len > 0) {
if (ret[*len - 1] == '\n')
--(*len);
}
return (bwscsbdup((unsigned char *)ret, *len));
} else {
*len = 0;
if (feof(f))
return (NULL);
if (2 >= rb->fgetwln_z_buffer_size) {
rb->fgetwln_z_buffer_size += 256;
rb->fgetwln_z_buffer = sort_realloc(rb->fgetwln_z_buffer,
sizeof(wchar_t) * rb->fgetwln_z_buffer_size);
}
rb->fgetwln_z_buffer[*len] = 0;
if (mb_cur_max == 1)
while (!feof(f)) {
int c;
c = fgetc(f);
if (c == EOF) {
if (*len == 0)
return (NULL);
goto line_read_done;
}
if (c == eols)
goto line_read_done;
if (*len + 1 >= rb->fgetwln_z_buffer_size) {
rb->fgetwln_z_buffer_size += 256;
rb->fgetwln_z_buffer = sort_realloc(rb->fgetwln_z_buffer,
SIZEOF_WCHAR_STRING(rb->fgetwln_z_buffer_size));
}
rb->fgetwln_z_buffer[*len] = c;
rb->fgetwln_z_buffer[++(*len)] = 0;
}
else
while (!feof(f)) {
wint_t c;
c = fgetwc(f);
if (c == WEOF) {
if (*len == 0)
return (NULL);
goto line_read_done;
}
if (c == eols)
goto line_read_done;
if (*len + 1 >= rb->fgetwln_z_buffer_size) {
rb->fgetwln_z_buffer_size += 256;
rb->fgetwln_z_buffer = sort_realloc(rb->fgetwln_z_buffer,
SIZEOF_WCHAR_STRING(rb->fgetwln_z_buffer_size));
}
rb->fgetwln_z_buffer[*len] = c;
rb->fgetwln_z_buffer[++(*len)] = 0;
}
line_read_done:
/* we do not count the last 0 */
return (bwssbdup(rb->fgetwln_z_buffer, *len));
}
}
int
bwsncmp(const struct bwstring *bws1, const struct bwstring *bws2,
size_t offset, size_t len)

18
usr.bin/sort/file.c
View File

@ -75,7 +75,7 @@ struct file_reader
struct reader_buffer rb;
FILE *file;
char *fname;
unsigned char *buffer;
char *buffer;
unsigned char *mmapaddr;
unsigned char *mmapptr;
size_t bsz;
@ -713,7 +713,7 @@ file_reader_readline(struct file_reader *fr)
}
} else if (fr->file != stdin) {
unsigned char *strend;
char *strend;
size_t bsz1, remsz, search_start;
search_start = 0;
@ -785,10 +785,16 @@ file_reader_readline(struct file_reader *fr)
fr->strbeg = (strend - fr->buffer) + 1;
} else {
size_t len = 0;
ret = bwsfgetln(fr->file, &len, sort_opts_vals.zflag,
&(fr->rb));
int delim = sort_opts_vals.zflag ? '\0' : '\n';
ssize_t len = getdelim(&fr->buffer, &fr->bsz, delim, fr->file);
if (len < 0) {
if (!feof(fr->file))
err(2, NULL);
return (NULL);
}
if (len > 0 && fr->buffer[len - 1] == delim)
len--;
ret = bwscsbdup(fr->buffer, len);
}
return (ret);