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

string.patternMatch, and small improvements

Expand Messages
  • Seth Dillingham
    I ve just given string.patternMatch an optional parameter called startingAt , so that it now looks like this: on patternMatch (pattern, s, startingAt = 1)
    Message 1 of 6 , Dec 27, 2004
      I've just given string.patternMatch an optional parameter called
      "startingAt", so that it now looks like this:

      on patternMatch (pattern, s, startingAt = 1)
      kernel (string.patternmatch)

      The old behavior is maintained in exactly the same way, but now you
      can also give it a position in the string at which it will start
      scanning.

      This is a very common language feature, I'm sure you don't need
      further explanation, but here are some examples just in case:

      string.patternMatch( "b", "abcde abcde" )
      >> 2

      string.patternMatch( "b", "abcde abcde", 3 )
      >> 8

      So this allows you to find matches anywhere in the string, rather than
      only the first match in the string. Before, if you wanted to match any
      occurrence of a pattern after the first match, you'd have to
      continually chop off the beginning of the string (up to and including
      the first matched character). This change will make it quite easy for
      some scripts to be immediately optimized.

      Anyway, this leads me to another question. Some of my comments a
      couple weeks ago were taken badly, perhaps sarcastically, but my point
      was sincere. I no longer feel completely comfortable checking in my
      changes because there has been so much resistance to my ideas... even
      small ideas which Andre supported, like min(), max(), and the math
      verbs.

      Here's an example. Dave requested a verb like bumpcounter( adrCounter
      ), here: <http://groups.yahoo.com/group/frontierkernel/message/393>,
      because they'd be faster and cleaner than the oft-used script
      equivalent.

      Yet, my own suggestions of min() and max() -- which I've already
      implemented, just like the improvement to patternMatch, but which also
      remain only on my computer -- were shot down even though the
      improvements and reasoning for them were identical to bumpCounter().

      I don't know what to do, and I'm a bit frustrated. Are my contributions wanted?

      Seth
    • Seth Dillingham
      ... I should have said right up front that this change has not been checked in yet. I did say it later in the email, but not everybody will read the whole
      Message 2 of 6 , Dec 27, 2004
        On Mon, 27 Dec 2004 16:11:51 -0500, I wrote:

        > I've just given string.patternMatch an optional parameter called
        > "startingAt", so that it now looks like this:

        I should have said right up front that this change has not been
        checked in yet. I did say it later in the email, but not everybody
        will read the whole thing.

        Seth
      • David Brown
        ... For things like your min and max and the bumpCounter request, I d put them in a separate branch of the source code, like Andre with his table improvements
        Message 3 of 6 , Dec 27, 2004
          Seth Dillingham wrote:
          > Yet, my own suggestions of min() and max() -- which I've already
          > implemented, just like the improvement to patternMatch, but which also
          > remain only on my computer -- were shot down even though the
          > improvements and reasoning for them were identical to bumpCounter().

          For things like your min and max and the bumpCounter request, I'd put
          them in a separate branch of the source code, like Andre with his table
          improvements and me with the Python work.

          I interpretted the original observations as "Let's work and make 10.0 as
          stable as we can, and leave the new features to a different version."

          We never did decide what the next "new feature version" would be, but I
          would agree that the "Frontier 10.0" release should be a stabilization
          of the previous commercial releases.

          > I don't know what to do, and I'm a bit frustrated. Are my contributions wanted?

          I know I can't speak for everyone, but I would say emphatically "YES!!!!"

          I want your contributions. We want your contributions. Branch the code
          and start a "small new features" code line. Lay the ground work for
          10.1, or 11.0 or whatever the version number is going to be for the
          first version of Frontier that starts to build on the base.

          Please please please, keep contributing.
        • Thomas A. Creedon
          ... contributions wanted? I appreciate the energy you are putting into this project. Please continue to do so. Toodle-loooooooo.......... Thomas editHere.com
          Message 4 of 6 , Dec 27, 2004
            --- In frontierkernel@yahoogroups.com, Seth Dillingham
            <seth.dillingham@g...> wrote:

            >I no longer feel completely comfortable checking in my
            > changes because there has been so much resistance to my ideas... even
            > small ideas which Andre supported, like min(), max(), and the math
            > verbs.

            > I don't know what to do, and I'm a bit frustrated. Are my
            contributions wanted?

            I appreciate the energy you are putting into this project. Please
            continue to do so.

            Toodle-loooooooo..........
            Thomas

            editHere.com < http://www.editHere.com/ > website hosting service.
            Websites as easy as see it, edit it, save it! (sm)
          • Philippe Martin
            ... Exactly what I think. Philippe -- ______________________________________________________________________ Philippe (Flip) MARTIN
            Message 5 of 6 , Dec 27, 2004
              At 13:26 -0800 27/12/04, David Brown wrote:
              >I want your contributions. We want your contributions. Branch the code
              >and start a "small new features" code line. Lay the ground work for
              >10.1, or 11.0 or whatever the version number is going to be for the
              >first version of Frontier that starts to build on the base.
              >
              >Please please please, keep contributing.

              Exactly what I think.

              Philippe
              --
              ______________________________________________________________________
              Philippe (Flip) MARTIN mailto:flip@...
              http://flip.macrobyte.net http://www.Free-Conversant.com
            • Steve Hooker
              ... Yes, yes! Though for the moment, I ve too many fish to fry in other directions to want to use them right now. I love what you are doing Seth ;-)
              Message 6 of 6 , Dec 30, 2004
                Seth Dillingham wrote:

                > I don't know what to do, and I'm a bit frustrated. Are my
                > contributions wanted?

                Yes, yes! Though for the moment, I've too many fish to fry in other
                directions to want to use them right now.

                I love what you are doing Seth ;-)
              Your message has been successfully submitted and would be delivered to recipients shortly.