This is the mail archive of the mailing list for the Archer project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [python][patch] Add options length parameter to value.string(...)

Thiago Jung Bauermann wrote:
El lun, 06-04-2009 a las 16:14 -0600, Tom Tromey escribiÃ:
"Phil" == Phil Muldoon <> writes:
The rest looks good to me.

>From what we talked on IRC, in this block:

      /* If we know the size of the array, we can use it as a limit on the
         number of characters to be fetched.  */
      if (TYPE_NFIELDS (type) == 1
          && TYPE_CODE (TYPE_FIELD_TYPE (type, 0)) == TYPE_CODE_RANGE)

we want to set *length and not fetchlimit from the array size. But
probably only if *length is -1. If the caller specified a *length
already, we shouldn't override it and continue using the array size for
the fetchlimit.
I'm not sure what benefit setting the length to arraysize in the case of length -1 would achieve? Fetchlimit will be set to either unint_max or the actual bounds of the array, and -1 means return at first null or fetchlimit length within the fetchlimit? So sending length of -1, and a fetchlimit as described in the current existing logic would "do the right thing"? If I were to set length to the array bound in the -1 case, it would fetch the whole array, and not stop on the first null (which we want for -1)?



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]