develooper Front page | perl.perl5.porters | Postings from January 2000

Re: On Pseudohashes

Thread Previous
January 14, 2000 10:07
Re: On Pseudohashes
Message ID:

Tom Christiansen wrote:
> I propose that we collectively consider the needs that pseudohashes
> were designed to meet, then assess the extent to which those needs
> have been adequately met.  it offers a process we can explore to help 
> decide whether the experiment should be deemed a success or a failure-

the hundred or so emails on pseudohashes are due to
differnces around what people are putting under the three headings
of "what I'd want to improve"  "what I'd be willing to sacrifice" and 
"what I'd be unwilling to sacrifice".

my pseudohash design spec:

"what I'd want to improve"
1) use less memory (compared to regular hash for the same values of data)

"what I'd be willing to sacrifice"
1) dynamic keys (keys are static, 
      declared up front or somehow determined by compiler)
2) inheritability (modules that use pseudo hashes cannot be inherited)
3) speed (use code if needed to compensate for any internal weirdness so that
        pseudohashes always behave like real hashes)

"what I'd be unwilling to sacrifice"
1) conceptual model of normal array and normal hash.
2) ease of conversion from pseudohash to normal hash in perl code.
        perl code should be same for operations on a 
        hash as on a pseudohash. only difference might be
        the declaration line that declares a hash to be
        pseudohash and possibly predeclare keys

if everyone put measurable items under these headings, it might 
be easier to sort out what's what.

I dont mean to go off on a rant here, but 
That's just my opinion, I could be wrong
        -- Dennis Miller

Thread Previous Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at | Group listing | About