Kelvin Eldridge wrote:
Russell Butler wrote:
Kelvin Eldridge wrote:
<snip>
Hi Kelvin
Thanks for the pointer to the new conversion utility.
I just tried it on my medical list (OOo 2.2 on gentoo Linux)
Unfortunately it produces big blocks of words run together without
spacing, and having to go through to insert spaces between them is
more of a problem than I had just doing a search/replace on a copy of
the file. IIRC there were a series of # between words, and just
replacing them with spaces (and/or carriage returns) gave me a list I
could sort alphabetically.
Also there doesn't seem to be any mechanism to export the list to a
new file, other than highlight, copy and paste.
Thanks anyway for working on it. Now we perhaps need a user friendly
way of editing Dictionary.lst.
Regards
Russell
Hi Russell,
I did notice the standard.dic file for Australian English appears not
to use Unicode characters. Yet if I create my own additional custom
dictionary as you have, that file uses Unicode characters.
From what I could see the utility I wrote worked with standard.dic,
but won't work directly with another custom .dic file like the medical
file.
When I saw this behaviour I took a punt the majority of people may
only use the standard.dic custom dictionary and so the utility may
prove to be useful. If it isn't, all I have done is waste some time.
If it proves useful then a lot more people can more easily contribute
words and that would be a good thing.
Does the utility work with your standard.dic file?
With your custom medical dictionary file I think you will find if you
open the file using MS Word, save the file as a text file, then use
the utility, it may work. It did for the test I did. You may have a
alternative package under Linux you can use. If you want me to try
this under Windows, send me a copy of your .dic medical file and I
will try it.
I used OpenOffice.org Basic and whilst I could be wrong, I don't think
I can get it to process Unicode characters correctly. Always happy to
be proven wrong. OpenOffice.org Basic isn't one of my stronger points.
The other issue to keep in mind is I use Windows and line end
characters can be an issue between operating systems.
With output going to another file, what I did was write a quick and
dirty program (quick and direct in name only, as it did take a fair
amount of time to analyse, write and test even this program). I felt
the copy and paste should be a sufficient compromise. It was only
meant to be an aid, not a professional tool.
The bottom line is I think I may have used the wrong tool to create
the utility. I used OpenOffice.org Basic as I felt it was appropriate
for this list.
Thanks again for the feedback. I will take on board the comments and
revisit this later if there is sufficient need.
Kevin:
I made a couple of changes in your code. You were trying to start at a
certain character (255) to start parsing the text. Because of a number
of other control characters in the file, it would not put the 'crlf' in
the correct places. I had it skip the first ll passes in the loop
before parsing the text. I then changed the 'Select Case' to look for 0
(null),instead of 3,6,8, to insert the 'crlf'. I only had some small
samples to work with for user dictionaries, but it seemed to work okay
with them. Also of note is that the code will give an error if the
number of lines to parse is too great from what I could gather. I am
not that familiar with StarBasic so I do not know what the character
limit is for a textbox.
TomW
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
