... [because none of the previous views appeared to be from ISE or ISE users] I've not seen any further comments. If there is no further discussion in the next...
Roger Browne
egroups@...
Jul 4, 2000 1:41 pm
1438
You should soon receive a message from eGroups inviting you to vote on the following proposal: Proposal for class STRING: Select the name of the ELKS2001...
Roger Browne
egroups@...
Jul 10, 2000 8:07 pm
1439
Enter your vote today! Check out the new poll for the eiffel-nice-library group: Proposal for class STRING: Select the name of the ELKS2001 feature ...
eiffel-nice-library@e...
Jul 10, 2000 8:10 pm
1440
I would now like to bring two related features into our discussion: 'index_of' and 'substring_index'. Here are their ELKS95 definitions: index_of (c:...
Roger Browne
egroups@...
Jul 11, 2000 12:16 pm
1441
... This signature is fine ... This is a change of words which sounds awkward to my ears. "from" goes with start_index, not c. ... This is a change of...
Peter Horan
peter@...
Jul 12, 2000 12:55 am
1442
... Here are my base ideas how searching should look like in general, independently of prior art: 1) The search features for ARRAY and STRING should be the...
Durchholz, Joachim
joachim.durchholz@...
Jul 12, 2000 6:57 am
1443
Roger Browne wrote: ... I'd better preserve ELKS'1995 naming for the feature "index_of". Otherwise the new standard would be incompatible with the old one. ......
Alexander Kogtenkov
kwaxer@...
Jul 12, 2000 7:55 am
1444
Good morning. ... I agree there's no obvious interpretation. ... Talking of short cuts, perhaps 'start_index' is a shortcut too. Here's a sideways look... ...
Simon Parker
sparker@...
Jul 12, 2000 11:37 am
1445
... It goes from 'last_index' towards 1. Dominique included the relevant assertions in his September 99 proposal, which is at ...
Roger Browne
egroups@...
Jul 12, 2000 3:18 pm
1446
This message describes an incompatibility in the existing behaviour of 'index_of' and 'substring_index'. ELKS 95 includes these header comments: index_of (c:...
Roger Browne
egroups@...
Jul 12, 2000 3:18 pm
1447
Friedrich, I have updated my copy of the Q compiler with your latest work. I haven't tried to compile it yet, but I have some comments after the merge process:...
Roger Browne
egroups@...
Jul 12, 2000 7:13 pm
1448
... I'm inclined to refine only the existing features and to defer the new ones for the future, because even the new proposed features do not cover all the...
Alexander Kogtenkov
kwaxer@...
Jul 13, 2000 8:53 am
1449
I can and have lived with either, but I prefer SmallEiffel's version. To me it's just a whole lot more intutive. Best, -- Jim ... ...
James McKim
jcm@...
Jul 13, 2000 11:18 pm
1450
The following eiffel-nice-library poll is now closed. Here are the final results: POLL QUESTION: Proposal for class STRING: Select the name of the ELKS2001...
eiffel-nice-library@e...
Jul 21, 2000 3:21 pm
1451
We discussed earlier this month whether the search functions need to be changed from 'index_of' and 'substring_index'. There seemed to be a clear feeling that...
Roger Browne
egroups@...
Jul 21, 2000 8:21 pm
1452
Just a few notes I made while perusing the proposed ELKS 2000 STRING class. The version I obtained is the one found at...
Jason Voegele
jason@...
Jul 22, 2000 12:20 am
1453
... I have a strong preference for returning count + 1 on failure of index_of based on the continuity of this choice. I quote my mail of 14 December, 1999 with...
Peter Horan
peter@...
Jul 23, 2000 12:47 pm
1454
... I have a strong preference for returning 0, for various reasons: 1) Returning "count + 1" is a sort of "imperative thinking" - it reflects the idea that...
Durchholz, Joachim
joachim.durchholz@...
Jul 24, 2000 7:35 am
1455
There seems to be some confusion wether to return '0', 'count+1' or something else on failed searches. IMHO the problem is that the result of 'index_of' and...
Arno Wagner
wagner@...
Jul 24, 2000 5:00 pm
1456
... Agreed. ... This would require a serious redesign of STRING, something that we have been trying to avoid. (If the arguments are compelling enough, we can...
Durchholz, Joachim
joachim.durchholz@...
Jul 24, 2000 9:33 pm
1457
... Unfortunately, words are like this - the letters which form them have a sequence. I suspect that any models which structure strings about which nothing is...
Peter Horan
peter@...
Jul 25, 2000 1:43 am
1458
Joachim Durchholz and Peter Horan debated on 0 against count+1: JD >> I have a strong preference for returning 0, for various reasons: ... PH> I would not say...
Pierre Métras
genepi@...
Jul 25, 2000 1:00 pm
1459
... A preliminary note: Jason, please set your mail client to Text-only format when posting here. The embedded formatting is making it more difficult to quote...
Durchholz, Joachim
joachim.durchholz@...
Jul 25, 2000 1:35 pm
1460
... to be ... are not ... are ... or "count + ... (i.e. ... features ... discuss ... I attempted to post a reply last week, but it doesn't appear in egroups....
Greg Compestine
gmc333@...
Jul 25, 2000 3:15 pm
1461
... <joachim.durchholz@h...> wrote: [snip] ... considered ... Agreed that it doesn't make sense. I'd recommend that the contract for this feature state that...
Greg Compestine
gmc333@...
Jul 25, 2000 3:17 pm
1462
... This doesn't mean that you have to go through it sequentially. (This is always an option of course, but it's not necessarily the only way to see a string.)...
Durchholz, Joachim
joachim.durchholz@...
Jul 25, 2000 5:11 pm
1463
... See my original example. -- Peter Horan School of Computing and Mathematics peter@... Deakin University ...
Peter Horan
peter@...
Jul 25, 2000 11:55 pm
1464
... My mistake, sorry. Due to too much C programming in the last week, I thought that the index of the first character in a STRING was '0'. After having read...
Arno Wagner
wagner@...
Jul 26, 2000 4:42 pm
1465
... Unfortunately, I see a very bad problem here: it's too easy to write code that seems to be correct (i.e. passes all tests) but is nonportable (i.e. it ...
Durchholz, Joachim
joachim.durchholz@...
Jul 26, 2000 5:50 pm
1466
... I was thinking perhaps index_of should not have a precondition of valid_index, but just return the off the end result. YMMV. ... This is the additional...
