The great RISC OS spell-checking problems:
Roger Darlington
It seems to me that there are many huge defficiencies in
Spell-checking implementations on RISC OS.
1) The user-dictionaries are far too small to be useful. Even 8 x 8kB
is far too small.
2) There is no 'BACK' button to press if you accidentally go on to the
'NEXT' word that needs attention.
3) There is no way to see let alone edit or delete the words the user
has added (perhaps accidentally) to the user-dictionary(s).
Has anyone seen any Spell-checker that has NONE of these short-comings
on RISC OS?
Do please let me know, especially if it works on HTML files without
flagging up every tag it hasn't heard of. (My user dictionary is full
to over-flowing, I cannot add any more words to it!).
--
Cheers
Roger
Is Morph a Pleistocene man?
Peter Naulls
> It seems to me that there are many huge deficiencies in
Including that word ;-)
> Spell-checking implementations on RISC OS.
>
> 1) The user-dictionaries are far too small to be useful. Even 8 x 8kB
> is far too small.
>
> 2) There is no 'BACK' button to press if you accidentally go on to the
> 'NEXT' word that needs attention.
>
> 3) There is no way to see let alone edit or delete the words the user
> has added (perhaps accidentally) to the user-dictionary(s).
>
> Has anyone seen any Spell-checker that has NONE of these short-comings
> on RISC OS?
You'd do well to name specific examples of applications here. In any
case, we're really talking about two different problems here. One is
the actual dictionary itself - there should be a central one (other
rather system, allowing for different languages) - and the other is
the behaviour of applications which might interact with it - Zap/SE,
Impression, Ovation, etc.
> Do please let me know, especially if it works on HTML files without
> flagging up every tag it hasn't heard of.
That's a slightly different problem again. If an editor has a proper
HTML mode (Zap/SE) then it should know what is a tag and what is not,
and therefore what shouldn't be queried in the dictionary.
Dr Peter Young
> Roger Darlington wrote:
[snip]
> > Do please let me know, especially if it works on HTML files without
> > flagging up every tag it hasn't heard of.
> That's a slightly different problem again. If an editor has a proper
> HTML mode (Zap/SE) then it should know what is a tag and what is not,
> and therefore what shouldn't be queried in the dictionary.
Spell-checking in StrongED's HTML mode certainly ignores all the tags.
However, as it uses the Impression spelling checker, the number of
words that can be added to user dictionaries is limited.
With best wishes,
Peter.
--
Peter, \ / \ Prestbury, Cheltenham, Glos. GL52
Anne \ / __ __ \ England.
and / / \ | | |\ | / _ \ http://pnyoung.orpheusweb.co.uk
family / \__/ \_/ | \| \__/ \______________ pny...@ormail.co.uk.
Martin Bazley
> Roger Darlington wrote:
> > It seems to me that there are many huge deficiencies in
>
> Including that word ;-)
>
Really? StrongED thinks it's fine. But then again, spell checking on
RISC OS is fundementally flawed, so what does it know?
--
__<^>__ === RISC OS is a work of art. Some people adore it, ===
/ _ _ \ === others can't see the point of it, and it's really ===
( ( |_| ) ) === expensive. ===
\_> <_/ ======================= Martin Bazley ===================
Vince M Hudd
> The following bytes were arranged on 23 Dec 2009 by Peter Naulls :
> > Roger Darlington wrote:
> > > It seems to me that there are many huge deficiencies in
> > Including that word ;-)
> Really? StrongED thinks it's fine. But then again, spell checking on
> RISC OS is fundementally flawed, so what does it know?
It is fine - now. Peter corrected it from Roger's original "defficiencies".
--
Vince M Hudd - Soft Rock Software
http://www.softrock.co.uk
http://misc.vinceh.com
Roger Darlington
> On 23 Dec 2009 Peter Naulls <pe...@chocky.org> wrote:
>
>> Roger Darlington wrote:
>
> [snip]
>
>> > Do please let me know, especially if it works on HTML files without
>> > flagging up every tag it hasn't heard of.
>
>> That's a slightly different problem again. If an editor has a proper
>> HTML mode (Zap/SE) then it should know what is a tag and what is not,
>> and therefore what shouldn't be queried in the dictionary.
>
> Spell-checking in StrongED's HTML mode certainly ignores all the tags.
> However, as it uses the Impression spelling checker, the number of
> words that can be added to user dictionaries is limited.
Does it? Zap doesn't :-( [but then I am on Zap 1.46, not 1.47]
How does StrongEd get on with styles:
<style>
body { background-color:#000011;
color: #ffffaa; }
a:link { color: #aaaaff; }
a:visited { color: #ffffaa; }
a:active { color: #ff00ff; }
</style>
--
Cheers
Roger
Don't play with your food - play with somebody else's.
Steve Fryatt
<f88b44ce5...@rogerarm.freeuk.com>:
> On 23 Dec 2009, Dr Peter Young wrote:
>
> > Spell-checking in StrongED's HTML mode certainly ignores all the tags.
> > However, as it uses the Impression spelling checker, the number of words
> > that can be added to user dictionaries is limited.
>
> Does it? Zap doesn't :-(
Oh yes it does...
You /are/ using Zap's HoTMeaL mode, aren't you?
--
Steve Fryatt - Leeds, England
Dr Peter Young
> a:link { color: #aaaaff; }
> a:visited { color: #ffffaa; }
Mixed success here; perhaps I wrote over-confidently. I saved that
bit, set its type as HTML and checked it. It picks up "color" as a
mis-spelling, understandable as it's a Rightpondian checker, but also
picks up "ffffaa" and "aaaaff" but not ""ff00ff"; presumably the
latter escapes as it contains non-hex digits. MPro's checker does the
same, BTW, and that uses David Pilling's dictionaries and its own
checker. However, it does accept "color".
Roger, perhaps send me, off-list, a bigger segment of HTML, and I'll
see what StrongED makes of it. In the meantime, I've made a screenshot
of the check I did, in case you're interested.
Russell Hafter News
Peter Young <pny...@ormail.co.uk> wrote:
> It picks up "color" as a mis-spelling, understandable as
> it's a Rightpondian checker,
You mean, it expects real English?
I am currently in frequent contact with an elderly American
about a trip to Germany he has booked for himself and nine
others. He asks me to check the documents he sends me for
accuracy and I always run them through Easiwriter's spell
checker. I know I ought not to, but when EW complains at
'color' and 'favorite' I always let it insert the missing
'u's.
By the way, I noticed this time a possible error in EW's
original German dictionary. Does anyone know if there is a
way to remove this? (It is a proper name, and even if
correct, it is not one that I need. But it is very similar
to one that is important for me.) Or maybe I should just ask
Martin Wurthner directly.
--
Russell
http://www.russell-hafter-holidays.co.uk
Russell Hafter Holidays E-mail to enquiries at our domain
Need a hotel? <http://www.hrs.com/?client=en__blue&customerId=416873103>
Peter Naulls
> In article <d4c54ace5...@pnyoung.ormail.co.uk>, Dr
> Peter Young <pny...@ormail.co.uk> wrote:
>
>> It picks up "color" as a mis-spelling, understandable as
>> it's a Rightpondian checker,
>
> You mean, it expects real English?
>
> I am currently in frequent contact with an elderly American
> about a trip to Germany he has booked for himself and nine
> others. He asks me to check the documents he sends me for
> accuracy and I always run them through Easiwriter's spell
> checker. I know I ought not to, but when EW complains at
> 'color' and 'favorite' I always let it insert the missing
> 'u's.
Right, and my mail client is configured for American English,
although I make an effort to use British for RISC OS contexts,
and naturally those are always highlighted. Which returns to
the original point - you need several dictionaries, with
a consistent interface which all apps can be use. Although to
be entirely fair, my mail client doesn't really know the best
locale to use for a given audience.
Jeremy Nicoll - news posts
> On 23 Dec 2009 Roger Darlington <roge...@freeuk.com> wrote:
> > How does StrongEd get on with styles:
>
> > <style>
> > body { background-color:#000011;
> > color: #ffffaa; }
>
> > a:link { color: #aaaaff; }
>
> > a:visited { color: #ffffaa; }
>
> > a:active { color: #ff00ff; }
> > </style>
>
> Mixed success here; perhaps I wrote over-confidently. I saved that
> bit, set its type as HTML and checked it. It picks up "color" as a
> mis-spelling, understandable as it's a Rightpondian checker, but also
> picks up "ffffaa" and "aaaaff" but not ""ff00ff"; presumably the
> latter escapes as it contains non-hex digits.
Excuse me? "0" is a valid digit in all number bases, certainly including
base 16.
Perhaps you mean that a /spelling/ checker doesn't like digits at all. But
then, values like "#fffc23" shouldn't be passed to the spelling checker at
all.
If StrongED can be persuaded to colour the #xxxxxx strings differently from
stuff around them then it can probably also be configured to think of them
as 'punctuation' or some other non-word element. But you have to remember
that good as SE's syntax-colouring etc is, it's written in a generic way to
enable people to define colouring of a wide range of file formats.
It's NOT however parsing the file to see how it fits a formal HTML or XML or
whatever structure (which after all really needs the sort of syntax-
checking code that's normally in a compiler). Some editors don't have
generic syntax colouring routines and instead have language/format-specific
support - that is a programmer has written a simple or complex syntax-aware
parser for what's meant to be in the file... and therefore can tell that
something really isn't a word.
Sometimes systems like that have some flexibility, eg there might be a
parser that's written to understand reasonably well variants of a language
like C. And sometimes a user of such a system can tell the editor to use
the basic layout of - eg C - but a different set of keywords, or a few more
built-in functions - and they'll still get a pretty good parse of content
even if it's not actually in C. But if you try and colour, say, a syslog
file using a C-based parser it just doesn't work, whereas with cunning one
can make SE colour syslog data quite well.
--
Jeremy C B Nicoll - my opinions are my own.
Email sent to my from-address will be deleted. Instead, please reply
to newsre...@wingsandbeaks.org.uk replacing "nnn" by "284".
Dr Peter Young
<jn.nntp....@wingsandbeaks.org.uk> wrote:
> Dr Peter Young <pny...@ormail.co.uk> wrote:
>> On 23 Dec 2009 Roger Darlington <roge...@freeuk.com> wrote:
>>> How does StrongEd get on with styles:
>>
>>> <style>
>>> body { background-color:#000011;
>>> color: #ffffaa; }
>>
>>> a:link { color: #aaaaff; }
>>
>>> a:visited { color: #ffffaa; }
>>
>>> a:active { color: #ff00ff; }
>>> </style>
>>
>> Mixed success here; perhaps I wrote over-confidently. I saved that
>> bit, set its type as HTML and checked it. It picks up "color" as a
>> mis-spelling, understandable as it's a Rightpondian checker, but also
>> picks up "ffffaa" and "aaaaff" but not ""ff00ff"; presumably the
>> latter escapes as it contains non-hex digits.
> Excuse me? "0" is a valid digit in all number bases, certainly including
> base 16.
Sorry, me being unclear. What I meant is that it ignores anything
including the digits 0 to 9.
Roger Darlington
> On 23 Dec, Roger Darlington wrote in message
> <f88b44ce5...@rogerarm.freeuk.com>:
>
>> On 23 Dec 2009, Dr Peter Young wrote:
>>
>> > Spell-checking in StrongED's HTML mode certainly ignores all the tags.
>> > However, as it uses the Impression spelling checker, the number of words
>> > that can be added to user dictionaries is limited.
>>
>> Does it? Zap doesn't :-(
>
> Oh yes it does...
>
> You /are/ using Zap's HoTMeaL mode, aren't you?
>
You did read the bit about styles didn't you>
--
Cheers
Roger
Never ask a question for which you dont already know the answer.
Roger Darlington
>
> It seems to me that there are many huge deficiencies in
> Spell-checking implementations on RISC OS.
>
> 1) The user-dictionaries are far too small to be useful. Even 8 x 8kB
> is far too small.
>
> 2) There is no 'BACK' button to press if you accidentally go on to the
> 'NEXT' word that needs attention.
>
> 3) There is no way to see let alone edit or delete the words the user
> has added (perhaps accidentally) to the user-dictionary(s).
4) No option to /start/ the spell-checking from the cursor or other
arbitrary word (well, not in Zap HoTMeaL, anyway)
>
> Has anyone seen any Spell-checker that has NONE of these short-comings
> on RISC OS?
>
> Do please let me know, especially if it works on HTML files without
> flagging up every tag it hasn't heard of. (My user dictionary is full
> to over-flowing, I cannot add any more words to it!).
>
>
--
Cheers
Roger
I'm walking slowly because I may be going in the wrong direction
Tony Moore
<jn.nntp....@wingsandbeaks.org.uk> wrote:
> Dr Peter Young <pny...@ormail.co.uk> wrote:
> > On 23 Dec 2009 Roger Darlington <roge...@freeuk.com> wrote:
>
> > > How does StrongEd get on with styles:
> >
> > > <style>
> > > body { background-color:#000011;
> > > color: #ffffaa; }
> >
> > > a:link { color: #aaaaff; }
> >
> > > a:visited { color: #ffffaa; }
> >
> > > a:active { color: #ff00ff; }
> > > </style>
> >
> > Mixed success here; perhaps I wrote over-confidently. I saved that
> > bit, set its type as HTML and checked it. It picks up "color" as a
> > mis-spelling, understandable as it's a Rightpondian checker, but
> > also picks up "ffffaa" and "aaaaff" but not ""ff00ff"; presumably
> > the latter escapes as it contains non-hex digits.
>
> Excuse me? "0" is a valid digit in all number bases, certainly
> including base 16.
>
> Perhaps you mean that a /spelling/ checker doesn't like digits at all.
> But then, values like "#fffc23" shouldn't be passed to the spelling
> checker at all.
>
> If StrongED can be persuaded to colour the #xxxxxx strings differently
> from stuff around them then it can probably also be configured to
> think of them as 'punctuation' or some other non-word element.
In StrongED's StrongHTML ModeFile, the line
_SpellIgnore ("<" ? {'~<>'}+ ">") | " "
needs tweaking. (No time to mess with it, just now.)
[snip]
> It's NOT however parsing the file to see how it fits a formal HTML or
> XML or whatever structure (which after all really needs the sort of
> syntax-checking code that's normally in a compiler).
For HTML syntax-checking, try Alex Macfarlane Smith's port of Tidy
http://www.archifishal.co.uk/software/riscos/downloads/tidy/tidy-03012006.zip
Tony
Roger Darlington
>> <style>
>> body { background-color:#000011;
>> color: #ffffaa; }
>
>> a:link { color: #aaaaff; }
>
>> a:visited { color: #ffffaa; }
>
>> a:active { color: #ff00ff; }
>> </style>
>
> Mixed success here; perhaps I wrote over-confidently. I saved that
> bit, set its type as HTML and checked it. It picks up "color" as a
> mis-spelling, understandable as it's a Rightpondian checker, but also
> picks up "ffffaa" and "aaaaff" but not ""ff00ff"; presumably the
> latter escapes as it contains non-hex digits. MPro's checker does the
> same, BTW, and that uses David Pilling's dictionaries and its own
> checker. However, it does accept "color".
>
> Roger, perhaps send me, off-list, a bigger segment of HTML, and I'll
> see what StrongED makes of it. In the meantime, I've made a screenshot
> of the check I did, in case you're interested.
>
I have StrongED here, but it is very little used.
I tried the spell-check option of it, but am left puzzled. I thought
it would be self-explanatory, but it does not seem to be so.
1) When it high-lights mis-spellings, how do I get to see what are its
offerred choices are from which I should choose instead?
2) When it high-lights what it thinks is a mis-spelled word, how do I
enter into its user dictionary the word it is highlighting but which
actually is spelled correctly?
I have tried pressing all mousse buttons; not much happens...
Perhaps it is expecting keyboard input but there are 1000's of
combinations of those I could try. I know, RTMF, but I thought someone
here may tell me how easy it is, or, conversely, how impossible it
is...
[mis-spelled especially for Xmas amusement]
BTW Merry Xmas :-)
--
Cheers
Roger
Where there's a Will, there's a Hey.
Tony Moore
[snip]
> I have StrongED here, but it is very little used.
>
> I tried the spell-check option of it, but am left puzzled. I thought
> it would be self-explanatory, but it does not seem to be so.
>
> 1) When it high-lights mis-spellings, how do I get to see what are its
> offerred choices are from which I should choose instead?
Position the cursor on a highlighted word, then _adjust_ click on the
dictionary icon in the StrongED toolbar. The word will appear in the
'Spell Check' dialogue window, together with a list of suggestions as to
its correct spelling. The drop-down menu allows you to specify the user
dictionary to which the word should be added, or ignored.
Tony
Martin Bazley
> I have StrongED here, but it is very little used.
>
> I tried the spell-check option of it, but am left puzzled. I thought
> it would be self-explanatory, but it does not seem to be so.
>
It is - well, to me at any rate.
> 1) When it high-lights mis-spellings, how do I get to see what are its
> offerred choices are from which I should choose instead?
Click Adjust on the option button at the top of the window to open the
mode choices. The 'Misc' section contains the relevant options.
>
> 2) When it high-lights what it thinks is a mis-spelled word, how do I
> enter into its user dictionary the word it is highlighting but which
> actually is spelled correctly?
>
You use the spell dialogue, as detailed above. You obviously won't get
anywhere without that. It can also be opened by clicking Adjust on the
dictionary icon.
> Perhaps it is expecting keyboard input but there are 1000's of
> combinations of those I could try. I know, RTMF, but I thought someone
> here may tell me how easy it is, or, conversely, how impossible it
> is...
>
You imply you've owned a copy of StrongED for some time. It's very easy
(pretty much anyone who uses StrongED could tell you how). And it is in
the manual (Introduction > Spell checking). What's the problem?
--
__<^>__ Red sky in the morning: Shepherd's warning
/ _ _ \ Red sky at night: Shepherd's delight
( ( |_| ) ) Mince and potatoes: Shepherd's pie
\_> <_/ ======================= Martin Bazley ==========================
Dr Peter Young
[snip]
> I have StrongED here, but it is very little used.
> I tried the spell-check option of it, but am left puzzled. I thought
> it would be self-explanatory, but it does not seem to be so.
> 1) When it high-lights mis-spellings, how do I get to see what are its
> offerred choices are from which I should choose instead?
> 2) When it high-lights what it thinks is a mis-spelled word, how do I
> enter into its user dictionary the word it is highlighting but which
> actually is spelled correctly?
Not very intuitive, I'm afraid! Do you have the book icon on the top
bar? If so, clicking select on this will highlight the mis-spelled
words but not do anything else. Adjust clicking opens the spell
dialogue; this does take a bit of getting used to, but can do a lot.
Initially it offers to add a word to the "ignore" dictionary, which
only accepts the word for this session; if you click on the menu icon
next to "Ignore" it gives you the list of user dictionaries, to which
you can add the word, and from which you can make anew dictionaries.
You then have to click on "next" to find the next unknown word. This
is all documented in the "Spell checking" section in the Introduction
to StrongED in the StrongHelp manual, available from the icon bar
icon.
> BTW Merry Xmas :-)
And to you and to all.
HTH,
Theo Markettos
> You'd do well to name specific examples of applications here. In any
> case, we're really talking about two different problems here. One is
> the actual dictionary itself - there should be a central one (other
> rather system, allowing for different languages) - and the other is
> the behaviour of applications which might interact with it - Zap/SE,
> Impression, Ovation, etc.
It's been years since I played with ImpressionSpell, but it seems to me that
improving that module is part of the way forward. Or writing a new version
from scratch with the same API. As I recall the API is fairly simple.
That would presumably handle the question of user dictionaries, and might be
extended to multiple language support.
I have to say, I do prefer the MS Word idea of red squiggly underlines,
rather than having to go through word by word. Though a Zap-style 'list of
found' (maybe even an editable one) might be useful too.
Theo
Dave Symes
Theo Markettos <theom...@chiark.greenend.org.uk> wrote:
[Snippy]
> I have to say, I do prefer the MS Word idea of red squiggly underlines,
> rather than having to go through word by word. Though a Zap-style 'list
> of found' (maybe even an editable one) might be useful too.
> Theo
Ovation Pro does the colour underline business as well.
Dave
--
Dave Triffid
druck
> It seems to me that there are many huge defficiencies in
> Spell-checking implementations on RISC OS.
There is no single spell checking implementation in RISC OS, which of
the dozens of applications with this feature are you talking about?
[Other rest of post snipped as pointless without this information]
---druck
--
The ARM Club Free Software - http://www.armclub.org.uk/free/
32 bit Conversions Page - http://www.armclub.org.uk/32bit/
Hauke Wegner
Russell Hafter News <see...@walkingingermany.invalid>
wrote:
> In article <d4c54ace5...@pnyoung.ormail.co.uk>, Dr
> Peter Young <pny...@ormail.co.uk> wrote:
[snip]
> By the way, I noticed this time a possible error in EW's
> original German dictionary. Does anyone know if there is a
> way to remove this? (It is a proper name, and even if
> correct, it is not one that I need. But it is very similar
> to one that is important for me.) Or maybe I should just ask
> Martin Wurthner directly.
You can remove words from the dictionary by clicking menue on the
EW/TW icon on the icon bar. Then select "exceptions -> spell
checker"*. There you can add new words or remove unwanted/incorrect
ones.
* I just assume that the options are called like this as i am running
the german version of TW).
Sincerely Hauke
--
*** fatal error - no signatures found ***
Roger Darlington
> On 23 Dec 2009 Roger Darlington <roge...@freeuk.com> wrote:
>> It seems to me that there are many huge defficiencies in
>> Spell-checking implementations on RISC OS.
>
> There is no single spell checking implementation in RISC OS, which of
> the dozens of applications with this feature are you talking about?
>
> [Other rest of post snipped as pointless without this information]
>
Point taken.
OK, I admit, I'm using !ZapSpell.
First of all, I cannot even find my user dictionary, although I know
it must be there (on the hard drive) somewhere - because there are a
lot of extra words in it that ZapSpell/HoTMeaL is using.
When I do find it, wherever it might be (one place it is NOT is in the
ZapSpell.Dictionary itself, I have looked, it is empty apart from one
text file: !ReadMe) then how do I see all the words in it, and how do
I edit it. (I admit that when I do find it, looking and editing it may
be self-explanatory, but I can't even find it...)
The ZapWebsite:
http://zap.tartarus.org/documentation/manual/Chapter13.html
Where it says that the main window has a large number of buttons:
Previous, Next, Check Selection and Check Buffer.
However, in the version I have (just downloaded the latest version in
the 'Everything32' of Zap 1.47) and I put the !ZapSpell program into
my Zap 1.46) there are only 4 buttons, Replace, Add, Ignore and
Cancel. I see no 'Previous' button.
So I am confused.
Have I got the right and latest version of ZapSpell?
Merry Xmas.
--
Cheers
Roger
The only thing that moves is the wind.
Dr Peter Young
wrote:
> Peter Naulls <pe...@chocky.org> wrote:
>> You'd do well to name specific examples of applications here. In any
>> case, we're really talking about two different problems here. One is
>> the actual dictionary itself - there should be a central one (other
>> rather system, allowing for different languages) - and the other is
>> the behaviour of applications which might interact with it - Zap/SE,
>> Impression, Ovation, etc.
> It's been years since I played with ImpressionSpell, but it seems to me that
> improving that module is part of the way forward. Or writing a new version
> from scratch with the same API. As I recall the API is fairly simple.
> That would presumably handle the question of user dictionaries, and might be
> extended to multiple language support.
That sounds interesting, and is it something that might get done?
However, are you talking about just Impression (which I don't use) or
such as StrongED (which I do use)?
> I have to say, I do prefer the MS Word idea of red squiggly underlines,
> rather than having to go through word by word. Though a Zap-style 'list of
> found' (maybe even an editable one) might be useful too.
StrongED does a "list of found", or are you wishing that Impression
did?
Russell Hafter News
<mailin...@my-postbox.eu> wrote:
> In message <50ce5aed...@walkingingermany.invalid>
> Russell Hafter News
> <see...@walkingingermany.invalid> wrote:
> > In article <d4c54ace5...@pnyoung.ormail.co.uk>,
> > Dr Peter Young <pny...@ormail.co.uk> wrote:
> [snip]
> > By the way, I noticed this time a possible error in
> > EW's original German dictionary. Does anyone know if
> > there is a way to remove this? (It is a proper name,
> > and even if correct, it is not one that I need. But it
> > is very similar to one that is important for me.) Or
> > maybe I should just ask Martin Wurthner directly.
> You can remove words from the dictionary by clicking
> menue on the EW/TW icon on the icon bar. Then select
> "exceptions -> spell checker"*. There you can add new
> words or remove unwanted/incorrect ones.
As far as I can see, that only allows me to modify *the
words that I have added* to the dictionary.
There seems to be no way that I can edit the original
dictionary, which seems to hold both Linderhof and
Linderhoff. I would like to remove the latter, as it is not
an unusual mistake in correspondent's documents that they
want checked.
Hauke Wegner
>> [snip]
Yes, i realize this now too. Also copying or renaming the original
dictionary file to the user dictionary file does not work as both are
the same filetype but different internal format.
Sadly i have currently no other idea right now, sorry for that.
BTW: Both Linderhof and Linderhoff are part of the original german
dictionary file. However both are no common words (there is a castle
Linderhof somewhere in Bavaria and perhaps same families got this
name. Seems to be a more personal decision of the
developer/translater.
Russell Hafter News
<mailin...@my-postbox.eu> wrote:
> BTW: Both Linderhof and Linderhoff are part of the
> original german dictionary file. However both are no
> common words (there is a castle Linderhof somewhere in
> Bavaria
Schloss Linderhof is near Ettal, a little further from, but
still within walking distance of, Oberammergau.
This is the one that I *need* regularly, as a tour operator
to Germany.
> and perhaps same families got this name. Seems to
> be a more personal decision of the developer/translater.
Googling Linderhoff on google.de and restricting the listing
to pages from Germany gives a number of listings for family
and business names, usually in the north of Germany.
When I found that Linderhoff was not flagged as a spelling
error, I had assumed that it was something that had got
added to my user dictionary, but no. (My German user
dictionary must be about 50% palce names, at least!)
Thanks for trying - I shall send something to the EW list to
see if there is any way of editing the original dictionary.
Steve Drain
> It's been years since I played with ImpressionSpell, but it seems to me that
> improving that module is part of the way forward. Or writing a new version
> from scratch with the same API. As I recall the API is fairly simple.
I produced a StrongHelp manual for SpellMod a few years back; it is
still available on my site:
http://www.kappa.me.uk/spellmod.htm
IMHO the API is awkward - it retains a some quirks from the original
SpellMaster ROM. I am not sure that it is the best option for a common
spell checker, but that is unlikely to be adopted for apps such as
OvationPro and Easiwriter anyway.
--
; ,', * Basalt * - gives RO 3.10+ versions of BASIC V new and alternative
;,' keywords, dynamic memory for arrays and blocks, new variable types.
;', Legal, fast and simple to use. Freeware - version 0.98� - 19 Aug 03
,; ',, Steve Drain, Kappa : http://www.kappa.me.uk/basalt.htm
Martin Wuerthner
Russell Hafter News <see...@walkingingermany.invalid>
wrote:
> In article <daec72d050.mailinglists@risc001>, Hauke Wegner
> <mailin...@my-postbox.eu> wrote:
>> BTW: Both Linderhof and Linderhoff are part of the
>> original german dictionary file. [...]
> When I found that Linderhoff was not flagged as a spelling
> error, I had assumed that it was something that had got
> added to my user dictionary, but no. (My German user
> dictionary must be about 50% palce names, at least!)
> Thanks for trying - I shall send something to the EW list to
> see if there is any way of editing the original dictionary.
You can save the effort: no, there isn't.
Martin
--
---------------------------------------------------------------------
Martin Wuerthner MW Software http://www.mw-software.com/
ArtWorks 2 -- Designing stunning graphics has never been easier
spam...@mw-software.com [replace "spamtrap" by "info" to reply]
Hauke Wegner
Martin Wuerthner <spam...@mw-software.com> wrote:
> In message <50d0a7a6...@walkingingermany.invalid>
> Russell Hafter News <see...@walkingingermany.invalid>
> wrote:
>> In article <daec72d050.mailinglists@risc001>, Hauke Wegner
>> <mailin...@my-postbox.eu> wrote:
>>> BTW: Both Linderhof and Linderhoff are part of the
>>> original german dictionary file. [...]
>> When I found that Linderhoff was not flagged as a spelling
>> error, I had assumed that it was something that had got
>> added to my user dictionary, but no. (My German user
>> dictionary must be about 50% palce names, at least!)
>> Thanks for trying - I shall send something to the EW list to
>> see if there is any way of editing the original dictionary.
> You can save the effort: no, there isn't.
> Martin
Well Martin if you need any suggestions for a further EW/TW release...
This might be one :o)
Martin Wuerthner
Hauke Wegner <mailin...@my-postbox.eu> wrote:
> In message <ec5ccfd0...@bach.planiverse.com>
> Martin Wuerthner <spam...@mw-software.com> wrote:
>> In message <50d0a7a6...@walkingingermany.invalid>
>> Russell Hafter News <see...@walkingingermany.invalid>
>> wrote:
>>> [...]
>>> Thanks for trying - I shall send something to the EW list to
>>> see if there is any way of editing the original dictionary.
>> You can save the effort: no, there isn't.
> Well Martin if you need any suggestions for a further EW/TW release...
> This might be one :o)
There is not too much point in suggesting features that only a tiny
minority of users would be interested in.
In any case, even if I wanted to do it I could not since the EW/TW
spell-checker is a licensed third-party product and the dictionaries
are in a proprietory compressed format.
Dr Peter Young
> In message <ec5ccfd0...@bach.planiverse.com>
> Martin Wuerthner <spam...@mw-software.com> wrote:
>> In message <50d0a7a6...@walkingingermany.invalid>
>> Russell Hafter News <see...@walkingingermany.invalid>
[snip]
>>> Thanks for trying - I shall send something to the EW list to
>>> see if there is any way of editing the original dictionary.
>> You can save the effort: no, there isn't.
>> Martin
> Well Martin if you need any suggestions for a further EW/TW release...
> This might be one :o)
FWIW with OPro you can edit the supplied dictionary (and I happily use
both apps where appropriate). Just as well, as our daughter's name was
mis-spelled in the supplied OPro dictionary.
Steffen Huber
> In message <e561e2d050.mailinglists@risc001>
> Hauke Wegner <mailin...@my-postbox.eu> wrote:
>
>> In message <ec5ccfd0...@bach.planiverse.com>
>> Martin Wuerthner <spam...@mw-software.com> wrote:
>
>>> In message <50d0a7a6...@walkingingermany.invalid>
>>> Russell Hafter News <see...@walkingingermany.invalid>
>>> wrote:
>
>>>> [...]
>>>> Thanks for trying - I shall send something to the EW list to
>>>> see if there is any way of editing the original dictionary.
>
>>> You can save the effort: no, there isn't.
>
>> Well Martin if you need any suggestions for a further EW/TW release...
>> This might be one :o)
>
> There is not too much point in suggesting features that only a tiny
> minority of users would be interested in.
>
> In any case, even if I wanted to do it I could not since the EW/TW
> spell-checker is a licensed third-party product and the dictionaries
> are in a proprietory compressed format.
An alternative might be the OpenOffice dictionaries. For most languages,
the licences (along with the library code to access it) are LGPL or
BSD.
We recently integrated OO spellchecking, thesaurus and grammar checking
into one of our products, and the components are really high quality.
Maybe a generic RISC OS dictionary manager based on the OO stuff would
be a good (albeit a bit late) idea.
Steffen
--
Steffen Huber
hubersn Software - http://www.hubersn-software.com/
Martin Wuerthner
Steffen Huber <sp...@huber-net.de> wrote:
> Martin Wuerthner wrote:
>> In any case, even if I wanted to do it I could not since the EW/TW
>> spell-checker is a licensed third-party product and the dictionaries
>> are in a proprietory compressed format.
> An alternative might be the OpenOffice dictionaries. For most languages,
> the licences (along with the library code to access it) are LGPL or
> BSD.
> We recently integrated OO spellchecking, thesaurus and grammar checking
> into one of our products, and the components are really high quality.
Yes, indeed. I know them very well because I recently integrated the
OO spellchecking engine (Hunspell) into a commercial product, too. ;-)
> Maybe a generic RISC OS dictionary manager based on the OO stuff would
> be a good (albeit a bit late) idea.
These are really two separate things. Creating a common generic spell
checking interface that could hide the differences between the
individual implementations has been proposed before and would
certainly be a good idea. Actually, would have been, because, as you
say, this would be rather late now. Most, if not all, applications
that would benefit from this development would probably not be updated
to use it.
Porting the OO stuff would probably only make sense if such a common
interface existed. Or, if one of the existing interfaces was bolted on
to it, so there would at least be some applications that could make
use of it out of the box.
Martin Bazley
> Porting the OO stuff would probably only make sense if such a common
> interface existed. Or, if one of the existing interfaces was bolted on
> to it, so there would at least be some applications that could make
> use of it out of the box.
>
How about a measure such as Jeffrey Lee's QTeM, which mapped the API for
an old module (TrackerModule) to the equivalent commands in a newer one
(QTM)?
--
__<^>__
/ _ _ \ You always find something in the last place you look.
( ( |_| ) )
Martin Wuerthner
Martin Bazley <martin...@blueyonder.co.uk> wrote:
> The following bytes were arranged on 29 Dec 2009 by Martin Wuerthner :
>> Porting the OO stuff would probably only make sense if such a common
>> interface existed. Or, if one of the existing interfaces was bolted on
>> to it, so there would at least be some applications that could make
>> use of it out of the box.
>>
> How about a measure such as Jeffrey Lee's QTeM, which mapped the API for
> an old module (TrackerModule) to the equivalent commands in a newer one
> (QTM)?
Yes, that would be one (and probably the cleanest) way of achieving
the effect of bolting on one of the existing interfaces.