Account Options

  1. Sign in
The old Google Groups will be going away soon, but your browser is incompatible with the new version.
Google Groups Home
« Groups Home
comp.lang.prolog
+ new post
About this group
Subscribe to this group
This is a Usenet group - learn more
Message from discussion This was memory fragmentation
The group you are posting to is a Usenet group. Messages posted to this group will make your email address visible to anyone on the Internet.
Your reply message has not been sent.
Your post was successful
 
From:
To:
Cc:
Followup To:
Add Cc | Add Followup-to | Edit Subject
Subject:
Validation:
For verification purposes please type the characters you see in the picture below or the numbers you hear by clicking the accessibility icon. Listen and type the numbers you hear
 
Jan Wielemaker  
View profile  
 More options Feb 8 2008, 9:50 am
Newsgroups: comp.lang.prolog
From: Jan Wielemaker <j...@nospam.ct.xs4all.nl>
Date: 08 Feb 2008 14:50:20 GMT
Local: Fri, Feb 8 2008 9:50 am
Subject: Re: This was memory fragmentation
Print | View thread | Show original | Report this message | Find messages by this author
On 2008-02-08, A.L <alewa...@zanoza.com> wrote:

> On 08 Feb 2008 08:18:47 GMT, Jan Wielemaker <j...@nospam.ct.xs4all.nl>
> wrote:

>>I'm a bit lost now.  Are we talking lines of source code or execution
>>time?  First I though I misunderstood and it is time instead of lines,
>>but the remarks below on Java testing indicate it is lines, but it
>>includes testing code.

> I messed up. There are two types of testing that I am doing: unit
> testing (that is done by writing EXTERNAL tests using framework that I
> call pUnit, as analogy to jUnit and I had to write myself). I don't
> count this code.

> The other testing is a) sort of a testing that could be done by type
> system, and this is analogy to my former Smalltalk
> "must_be_instance_of" assertions. I must be getting list when I expect
> a list, and structure of this list must be as it is expected, b) data
> integrity test that type system would not do (like time of effect of
> some action cannot precede time of cause of this action)

> This 75% does not cover pUnits, pUnit tests is additional XXXX lines
> of code.

So my first reading was correct after all.  There is 75% characters
(or lines) of the source code that does not add to the logic but is
merely there to validate assumptions?  That is really huge.

Ok, Prolog source is overall so much shorter that the total amount of
code and readability might still not be bad ...

What is interesting now is how much of the 75% could be replaced by a
type system and how much lines are needed for the declarations that
replace these tests.

        Cheers --- Jan


 
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
Create a group - Google Groups - Google Home - Terms of Service - Privacy Policy
©2012 Google