fluencelabs/redis
1
0
Fork 0
mirror of https://github.com/fluencelabs/redis synced 2025-06-21 21:11:33 +00:00
Code Issues Projects Releases Wiki Activity

Fix HINCRBYFLOAT to work with long doubles.

Browse Source 
During the refactoring needed for lazy free, specifically the conversion
of t_hash from struct robj to plain SDS strings, HINCRBFLOAT was
accidentally moved away from long doubles to doubles for internal
processing of increments and formatting.

The diminished precision created more obvious artifacts in the way small
numbers are formatted once we convert from decimal number in radix 10 to
double and back to its string in radix 10.

By using more precision, we now have less surprising results at least
with small numbers like "1.23", exactly like in the previous versions of
Redis.

See issue #2846.
This commit is contained in:
antirez
2015-11-04 17:16:34 +01:00
parent f6255703b0
commit 71aa9b75f2
3 changed files with 8 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
src/t_hash.c
View File

@ -596,23 +596,23 @@ void hincrbyCommand(client *c) {
}
void hincrbyfloatCommand(client *c) {
double value, incr;
long double value, incr;
long long ll;
robj *o;
sds new;
unsigned char *vstr;
unsigned int vlen;
if (getDoubleFromObjectOrReply(c,c->argv[3],&incr,NULL) != C_OK) return;
if (getLongDoubleFromObjectOrReply(c,c->argv[3],&incr,NULL) != C_OK) return;
if ((o = hashTypeLookupWriteOrCreate(c,c->argv[1])) == NULL) return;
if (hashTypeGetValue(o,c->argv[2]->ptr,&vstr,&vlen,&ll) == C_OK) {
if (vstr) {
if (string2d((char*)vstr,vlen,&value) == 0) {
if (string2ld((char*)vstr,vlen,&value) == 0) {
addReplyError(c,"hash value is not a float");
return;
}
} else {
value = (double)ll;
value = (long double)ll;
}
} else {
value = 0;