home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Whiteline: delta
/
whiteline CD Series - delta.iso
/
english
/
xlator
/
read.me
< prev
next >
Wrap
Text File
|
1995-11-25
|
11KB
|
276 lines
----------------------------------------------------------------------
XLATOR V2.01 8/6/94 (Even better value at £5)
----------------------------------------------------------------------
* Searches for German or French resource files (XLATOR_G.RSC and
XLATOR_F.RSC) if running on a German or French machine.
(If not found, the standard resource file is loaded)
* All utilities are now contained in one program.. XUTILS.PRG
* All have been completely re-written in 'C' and are much faster.
* A new utility function, 'Twist dictionary' has been added.
The twist function allows a dictionary to be reversed.
e.g. The German dictionary entry for 'wieder' is:
wieder again/against/back/recurrence/render
after twisting, it becomes:
again wieder
All other translations for this word are saved in a separate
file called UPDATE.DIC. The UPDATE.DIC file can then be sorted
in exactly the same way as for an UNKNOWN.DIC file. After sorting
you can update the twisted dictionary with the other translation
words if required (see example).
The dictionary updating routine has been substantialy altered to
provide support for dictionary twisting. The changes are understood
best by an example, so here goes:
OK, let's say you want to convert the German to English dictionary
into an English to German dictionary.
(1) Select 'Twist dictionary' from menu, select GER_ENG.DIC as the
source file then have a coffee while you're waiting. When it's
finished, save the output file as say TWISTED.DIC (Ouch!).
(2) Select 'Process UNKNOWN.DIC' from menu and choose UPDATE.DIC as
the source file. This will take just a few seconds to sort.
(3) Select 'Update dictionary' from menu, choosing TWISTED.DIC as
the dictionary and then UPDATE.DIC as the update file. When done
the fileselector appears again and this time enter the proper
name for the dictionary, ENG_GER.DIC. The program will now start
a second pass on the new dictionary. Pass 2 re-formats the file
such that multiple entries appear on a single line.
e.g. for the English word 'assign', after steps 1,2 & 3 above:
The TWISTED.DIC file will contain
assign uberweisen
The UPDATE.DIC file will contain
assign berufen
assign zuteilen
assign zuweisen
And ENG_GER.DIC file will contain (this is done during pass 2)
assign berufen/uberweisen/zuteilen/zuweisen
Note that the German words after 'assign' are in alphabetical order.
Don't be too alarmed if the final twisted dictionary is much smaller
than the original. The reason is that many translations which would
be unsuitable are automatically deleted. For example many translations
are of the form:
bildschirmgradationen grey_scale_monitor
angeboten offer_to(?)
The above translations would be deleted since they make no sense when
switched round. No one writes English in this way. Also, any validity
characters present in the original dictionary will be deleted from the
twisted dictionary.
I apologise if it all seems complicated, but I wanted to be able to
create a minimal dictionary with only one translation. This can be
done by performing only step 1 above and ignoring steps 2 and 3. I'm
always open to suggestions, so don't hesitate to drop me a line if
you have any ideas (or money) you would like to share with me.
P.S.(I have changed the numbering system for the version numbers,
hence the jump from 1.03 to 2.01, I have my reasons!)
----------------------------------------------------------------------
XLATOR V1.03 12/3/94 (unreleased)
----------------------------------------------------------------------
Two versions in just two days!
* Sped things up a bit in MultiTOS mode, its much better now.
* Source files can now be dropped on the XLATOR icon on the
desktop (TOS >= 2.xx). You need only then select the language
and click the Translate button.
* Minor bug fixes.
I've decided the translation speed is fast enough, so I'm not
going to bother changing it. Will work on the utilities instead.
----------------------------------------------------------------------
XLATOR V1.02 10/3/94 (Still Shareware, still £5)
----------------------------------------------------------------------
* Translation routines are now even faster!!
* Dictionaries can be stored anywhere on disk since
the XLATOR.INF file now saves the dictionary path.
The new routines are based on the mathematical bi-section method
of convergence, if I can find a way to implement Reguli Falsi,
further speed improvements could be made.
* A new utility has been added to this release: GUESS.PRG
After you've sorted you're UNKNOWN.DIC file (see below), you may
run this utility to guess at words which were not found in the
dictionary. It produces a file called GUESSED.DIC with some
possible translations for you to try.
Be warned, the program is written in BASIC and is very slow.
----------------------------------------------------------------------
XLATOR V1.01 5/2/94 (Shareware!)
----------------------------------------------------------------------
XLATOR is now shareware, fee is £5.00 sterling (very cheap).
* GER_ENG.DIC contains over 27,000 translations.
* Translation routines are now case sensitive.
* Translation speed is approximately 10 times faster!!
* Bug which sometimes caused the last line in a file to be
skipped, has been fixed.
* Optimized for 68030 processor (still works on all ST's).
* Resource file supports 3D GEM.
* Utilities can now handle 50,000 line dictionaries.
* Progress box has been incorporated into main window display.
Permission is hereby granted for the distribution of this software
in any form. Including, but not restricted to, Magazine cover disks,
Public Domain libraries, electronic traffic, internal combustion
engined traffic (2 & 4 stroke), rotary Wankel engine traffic, FM and
AM radio, UHF, microwave satellite link, Patriot missile (but NOT SCUD),
disks may also be glued to carrier pigeons/lemmings etc.
**************** SHORT MANUAL FOR VERSION 1.00 ****************
Welcome to XLATOR V1.00 8/6/93 (Public Domain)
by Gair Shields.
XLATOR is a universal language translation program. It is not
clever, just a word for word translator, examining a file and
translating words which have an entry in the dictionary. The
translated text will be completely out of context but still a
whole lot easier to understand than a foreign language.
As of 8/6/93 there is only one dictionary available, GER_ENG.DIC,
the German to English dictionary. If the program proves popular
and people send me new dictionaries, they will be rewarded with
regular updates to the software.
The German language has a tendency to combine several words to
make a single word. For example:
Bildschirmgradationen = Grey scale monitor
This causes problems for a word for word translator. Future
versions of the program may address these problems. I would
therefore be grateful to anyone who knows a bit about the German
language, contacting me and helping out with this.
When you run the program you will see that two options in the
'FILE TYPE' menu are disabled. These are intended for a future
version of XLATOR.
MultiTOS (or any other multitasking OS)
If you run the program in MultiTOS mode, the first thing you
notice is that it is incredibly slow. This is because the program
is continually checking for GEM events while it is translating.
It does have the advantage that you can move the window around
the screen and the window is redrawn if it becomes obscured. It
is not necessary to switch to MultiTOS mode when running under
MultiTOS, if you don't, it just means the program will ignore all
window events while it is translating and will run much quicker.
SUBSTITUTE
The SUBSTITUTE menu allows you to specify the number of words to
insert, in place of the foreign word. For example, the
translation of the word 'wieder' shown below, has five equivalent
english words. If SUBSTITUTE is set to 1, only the first
translation word will be inserted, if set at 2, the first two
words are inserted etc..
Dictionaries
XLATOR dictionaries are plain ASCII files in the following
format:
XLAT This is always on line 1, it identifies a valid
dictionary file. Nothing else should be on this
line.
Line 2 contains 'validity characters' these are alphabetic
characters which are valid for the particular language being
translated. For example the German language has the following
validity characters on this line:
üäÄöÖÜß
From line 3 to the end of the file, are the translation words.
(Foreign_word)(Single_TAB)(Translated_word(s))
e.g.
wieder again/against/back/recurrence/render
NOTE: The tab MUST be a tab, not a space.
It is recommended that dictionaries are updated using the
UPDT_DIC.PRG utility and not by hand.
Utility programs
A few utilities are supplied with the main program, they may one
day be a part of it.
UNKN_SRT.PRG
When a file is translated, an additional file is created on
disk called UNKNOWN.DIC, this file contains all the words
the program was unable to translate, because they were not
in the dictionary. Use this utility to remove duplicate
entries in this file, and sort the file into alphabetical
order. You can then go though the file, armed with your
pocket dictionary, find suitable translations for these
words and add them to UNKNOWN.DIC. Delete words which you
cannot find a translation for. The file should be in the
same format as the dictionary, except you should have no
special entries for lines 1 and 2.
UPDT_DIC.PRG
When you have sorted your UNKNOWN.DIC file and added new
translations, this utility is used to update the dictionary
with the new words, making sure they are sorted correctly.
The program does not check to see if the word is already in
the dictionary, since it assumes it is being updated by the
UNKNOWN.DIC file. You can use it to merge two dictionaries
together, but you must run the DIC_SORT program on the
output file, to remove identical lines.
DIC_SORT.PRG
This utility is used to process a new word list and turn it
into a dictionary. The file is sorted alphabetically,
duplicate entries are removed and blank lines are deleted.
You must run this program on an merged dictionary!
N.B.
The utilities are written in GFA BASIC and therefore may not work
under MultiTOS. XLATOR itself is written in Lattice C 5.6 and
should have no problems with compatibility.
Send your dictionaries/suggestions/cash? to..
Gair Shields
73 Old Castle Road
Cathcart
Glasgow
G44 5TG
Scotland