here's the patch

CommentFileSizeAuthor
#7 locale-inc.patch739 byteskilles@www.drop.org
#4 pgsql_locales_fix_0.diff2.64 KBadrian
#3 pgsql_taxo_bug_0.diff1.96 KBadrian
pgsql_locales_fix.diff2.09 KBadrian

Comments

killes@www.drop.org’s picture

killes@www.drop.org commented
Priority: Critical » Normal

The patch should not be neccessary becuase the location field is pruned to contain 128 chars. But - at least for pgsql- it doesn#t hurt either.

adrian’s picture

adrian commented

here is an updated patch, which fixes the problem with too many unique restraints on locales_target.

adrian’s picture

adrian commented
StatusFileSize
new pgsql_taxo_bug_0.diff1.96 KB

err. here is the patch.

adrian’s picture

adrian commented
StatusFileSize
new pgsql_locales_fix_0.diff2.64 KB

here is a corrected patch, it contained some cruft fromt he taxonomy patch i made earlier

gábor hojtsy’s picture

gábor hojtsy
he/him
Primary language Hungarian
Location Hungary
commented
Version: 4.5.0-rc »

Killes, the 128 char check is improperly implemented (the length is checked before a new substring of unknown length is added), so that check needs to be fixed in the first place.

gábor hojtsy’s picture

gábor hojtsy
he/him
Primary language Hungarian
Location Hungary
commented

Killes, the 128 char check is improperly implemented (the length is checked before a new substring of unknown length is added), so that check needs to be fixed in the first place.

killes@www.drop.org’s picture

killes@www.drop.org commented
Component: postgresql database » locale.module
Assigned: vertice@www.drop.org » killes@www.drop.org
StatusFileSize
new locale-inc.patch739 bytes

How about this?

Steven’s picture

Steven commented

The first patch was apparently applied to CVS already. But it seems that in the pgsql code, the temporary column "old_location" is not dropped?

Still, we should decide on this... either we:
- Remove the substring checking completely and change "location" to text in both pgsql and mysql.
or
- We go back to a 128-char location, change the db columns back, and implement killes' improved checker.

By the way killes, I think the strlen($comm) < 128 in the while() condition is unnecessary with your later check in place.

Steven’s picture

Steven commented
Status: Needs review » Fixed

I don't think there's going to be any more follow-ups on this. Committed.

Anonymous’s picture

(not verified) commented
Anonymous’s picture

(not verified) commented
Anonymous’s picture

(not verified) commented
Anonymous’s picture

(not verified) commented
Anonymous’s picture

(not verified) commented
Status: Fixed » Closed (fixed)