home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!uknet!icdoc!swan.doc.ic.ac.uk!ijp
- From: ijp@swan.doc.ic.ac.uk (Ian Palmer)
- Newsgroups: comp.sys.acorn
- Subject: Re: WordHound
- Message-ID: <1992Jul28.101659.20367@doc.ic.ac.uk>
- Date: 28 Jul 92 10:16:59 GMT
- References: <20Pfcjj010n@st-andrews.ac.uk>
- Sender: usenet@doc.ic.ac.uk
- Organization: Department of Computing, Imperial College, University of London, UK.
- Lines: 76
- Nntp-Posting-Host: swan.doc.ic.ac.uk
- X-Newsreader: Tin 1.1 PL3
-
- gta@st-andrews.ac.uk (Graham Allan) writes:
- : tim@parcplace.com (Tim Rowledge) writes:
- :
- [ et, al. various comments about Word Hound ]
-
- This is attempt number 3 (our news server has all but died here) at
- replying to the various comments about Word Hound.
-
- First of all, no "antidisestablishmentarianism" is not there, but then
- I didn't create the dictionary data :-). And yes (at some time in the
- future, well that's my current intention anyway) be able to add more
- data files to Word Hound.
-
- Second, yes I know there are quite a few bugs in it, many have already
- been corrected. My main intention for releasing a version so soon in
- it's developement (it's only about 3 weeks old, and most of that time
- has been spent with my first attempt at fighting with RiscOSLib :-( )
- is that I wanted (a) to let everyone know of it's exsistence, and (b)
- to make sure I could get peoples "wants" before I got too far and they
- became too difficult to add.
-
- To answer the question about the compression method used, it's a
- simple(ish) substitution compression where each character is
- substituted by a bit pattern (patterns being of varying lenghts). This
- is not as good at compression as (say) the LZW (and the like), but
- these are no good as you don't want to sit through a complete
- decompression every time you want to look up a word.
-
- Next, here is a list of what I have already done to it for version
- 1.10 (available probable fairly soon(ish)) :
-
- o Corrected a few major bugs
-
- o Compressed the index files (now less than 50% their old size)
-
- o Added highlighting of selected word/phrase
-
- o General tidy up of internals
-
- o Will add a space after it outputs a word/phrase if it needs to
- (ie. if one was passed to it by dragging a selection)
-
-
- Problems I know about and working on :
-
- o The scrolling problem
-
- o The 'Enter' window disappearing (see below)
-
- o Allowing addition of own data files
-
-
- I have one major problem (apart from finding out how to hotlink to
- Impression better) which I would like some info on.
-
- When you have entered a word into the 'Enter' window it would be nice
- if the caret returned to whereever it was before you entered the word
- into that window. This is currently provided by the fact that to be
- able to enter a word you need to first bring up the 'Enter' window
- which allows the program to note where the caret is before this
- happens. And as the window disappears each time you enter a word there
- is no real problem with this, except it's annoying.
-
- The problem is that if I make the window remain open, next time you
- use it the caret might be somewhere else, and so the program will be
- unable to get a correct caret location. My question is, is there any
- way that every time you gain the caret, note where it was just before
- you gained it?
-
- Ian
-
- --
- E-mail : ijp@doc.ic.ac.uk
- ___ __ Snail mail : Department of Computing, Huxley Building,
- / _ _ /_/ _ / _ _ _ _ Imperial College, 180 Queens Gate,
- _/_ (_|_| ) / (_|_(_| ) )_|/_) London. SW7 2BZ. England.
-