Re: pgsql: Fix strftime usage on Win32 when trying to fetch the locale-aware

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Magnus Hagander <magnus(at)hagander(dot)net>
Cc: pgsql-committers(at)postgresql(dot)org
Subject: Re: pgsql: Fix strftime usage on Win32 when trying to fetch the locale-aware
Date: 2009-01-09 13:55:51
Message-ID: 19567.1231509351@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-committers

Magnus Hagander <magnus(at)hagander(dot)net> writes:
> Tom Lane wrote:
>> The Single Unix Spec's definition of wcsftime says that the above
>> risks a buffer overrun, and the correct second argument would be
>> MAX_L10N_DATA. Now I realize that SUS is a poor guide for
>> Windows-specific code, but are you sure this is right?

> Now that I read it again, I think you're right. What MS says is:
> "If the total number of characters, including the terminating null, is
> more than maxsize, both strftime and wcsftime return 0 and the contents
> of strDest are indeterminate."

> The important difference being "character" vs "bytes", right?

SUS phrases it as

"If the total number of resulting wide-character codes including the
terminating null wide-character code is no more than maxsize, wcsftime()
returns the number of wide-character codes placed into the array pointed
to by wcs, not including the terminating null wide-character
code. Otherwise 0 is returned and the contents of the array are
indeterminate."

so it's very clear that maxsize is counted in wchars.

Perhaps someone could experiment to double-check what Windows does.

regards, tom lane

In response to

Responses

Browse pgsql-committers by date

  From Date Subject
Next Message Magnus Hagander 2009-01-09 14:05:56 Re: pgsql: Fix strftime usage on Win32 when trying to fetch the locale-aware
Previous Message Magnus Hagander 2009-01-09 13:52:05 Re: pgsql: Fix strftime usage on Win32 when trying to fetch the locale-aware