• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Advice request from newbie
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Advice request from newbie


  • Subject: Advice request from newbie
  • From: Andrea Perego <email@hidden>
  • Date: Fri, 21 Sep 2001 18:22:03 +0200

Hi,

I'm just starting to learn Cocoa, with no previous experience in OOP (but for some Java tutorial). My goal would be to port a rather large application I developed some years ago (better said, I've been developing since....;-> )
using MPW and C. I'd prefer a Cocoa porting rather than a plain "carbonization" since:
- the application relies on rather older APIs and I should work a lot just
to upgrade it to the level where carbonization may begin.
- I'm rather bored of spending 80%+ of my programming time just for the GUI.

My concerns lie in the fact that this application deals with its data in a way
that does not conform to the standard "document" paradigm:

- its main purpose is to perform analysis of nuclear spectra, i.e. histograms
of, typically, 4096-8192 channels.

- in many cases, the output data are just the result of some calculation, and may be stored as plain text files. Sometimes, however, new spectra are created
by manipulating existing ones, and must be archived on disk.

- the spectra are stored on disk in files that contain many of them. Each file has a fixed-size header with a "jump table" to the individual, sequentially stored, spectra. Presently, the routines to load-store spectra
walk through this internal file structure, reading and writing at the
proper file offsets.

- during the process of selecting the spectra to be read, the user has access to some info pertaining to each spectrum (e.g. read from the spectra file) before
making her/his choice.

These peculiarities (and my lack of Cocoa experience!) make me dubious about
embracing the standard Cocoa model for document driven applications, with due
subclassing, or skipping it altogether. Of course, a rather easy way (?) would be to equate: a spectra file <-> a Cocoa document, but I don't find the solution appealing, because the application should read/write/keep in [virtual] memory a large amount of unused data. After all, I've started programming in the early '70, when 32Kbytes were a dream, and economy has become a second nature...

Therefore, I'd like the opinion of some more expert member of the Cocoa community.

TIA
Andrea Perego
Univ. of Florence - Phys. Dept.


  • Follow-Ups:
    • Re: Advice request from newbie
      • From: Ondra Cada <email@hidden>
  • Prev by Date: Re: NSArray behavior changes ?
  • Next by Date: NSSound--description forthcoming
  • Previous by thread: Re: bug in NSBundle when using QuickTime ?
  • Next by thread: Re: Advice request from newbie
  • Index(es):
    • Date
    • Thread