Loading ...
Sorry, an error occurred while loading the content.

RE: [XP] TF Challenge: how did you implement "circular reference detection"

Expand Messages
  • Dinwiddie, George
    Yes (now that I ve finally caught up with Challenge 3, I m reading the messages pertaining to it), that was my tack. The circular reference issue is what made
    Message 1 of 54 , Jan 31, 2002
    • 0 Attachment
      Yes (now that I've finally caught up with Challenge 3, I'm reading the
      messages pertaining to it), that was my tack. The circular reference issue
      is what made me factor out a "Cell" class. It sets a boolean when it starts
      evaluating itself and resets it on completion. If the boolean is set on
      entry to the evaluation of the cell, there is a circular reference.

      - George

      > -----Original Message-----
      > From: Ilja Preuß [mailto:ilja.preuss@...]
      > Sent: Thursday, January 31, 2002 12:23 PM
      > To: extremeprogramming@yahoogroups.com
      > Subject: RE: [XP] TF Challenge: how did you implement "circular
      > reference detection"
      >
      >
      > > I made a vector that got passed along as the expression is
      > > evaluated. As
      > > each cell is evaluated the cell's name is pushed onto the
      > > vector. Each
      > > time I evaluate a new cell I check to see if it is on the vector
      > > already. If so it returns "#Circular".
      >
      > We did the same, but then discovered that a cell didn't need to know
      > which cells already got visited, it only needed to know if it was
      > already visited before. Our code became much simpler by putting this
      > knowledge into the code...
      >
      > Regards, Ilja
      >
      >
      > To Post a message, send it to: extremeprogramming@...
      >
      > To Unsubscribe, send a blank message to:
      > extremeprogramming-unsubscribe@...
      >
      > ad-free courtesy of objectmentor.com
      >
      > Your use of Yahoo! Groups is subject to
      > http://docs.yahoo.com/info/terms/
      >
      >
    • Blum, Robert
      ... If you do not know the intention of the test, how can you make sure you gave the customer what he wanted? IMHO, XP does not mean that the customer just
      Message 54 of 54 , Feb 1, 2002
      • 0 Attachment
        > From: Brian C. Robinson [mailto:brian.c.robinson@...]

        > Blum, Robert made a strange utterance something like this:
        > > Getting just the minimum work in so
        > >that tests pass, without considering the intention of the
        > test qualifies as
        > >a relic of the code&fix days, in my book. (Oh, seems to
        > work. Let's ship it)
        >
        > Actually, that's XP. Why do more than the minimum? Do you
        > need more than
        > the minimum, maybe for the future?

        If you do not know the intention of the test, how can you make sure you gave
        the customer what he wanted? IMHO, XP does not mean that the customer just
        hands me a stack of stories and ATs, and I write code to satisfy those
        tests. It helps if you then communicate with the customer. ESPECIALLY if you
        are satisfying the test, but violating the intent of the test.

        > >We're actually paid to think about the solution we deliver -
        > not to just
        > >satisfy the tests blindly.
        >
        > Huh. I'm pretty sure I'm paid to satisfy the tests blindly.
        > And to write
        > the tests intelligently.

        Try adding communication to that. How can you come up with intelligent UTs
        without that?

        Which neatly condenses the whole problem of this excercise - the customer
        handed us a bunch of requirements, and everybody satisfied them. Slightly
        differently. Why? Because communication was limited. Which is the right one?
        As I said before, I don't know. Depends on your customer.
        Robert
      Your message has been successfully submitted and would be delivered to recipients shortly.