Space Quest III French translation ?

[Threepwang]

Hello everyone!

*I am sorry for my bad English!
I am a new members and "Well, guess what?"
I am a fan of Sierra games!

I think that we should to translate in all languages the Sierra classic games to make them accessible for players.

I want to translate Space Quest 3 in french.
And I am looking (search) for an SCI engine expert.
I am having some difficulty with accented letters French.
Does anyone know if there is a code OEM in SCI Compagnon?
Because SCI Compagnon is not include OEM Sierra.
It is ANSI and my computer is a ANSI (ASCII) Default Encoding.
For example with german game:
"The terminator's invisibility belt is now completely out of power.%GDer Energievorrat des Tarngrtels ist jetzt v”llig ersch”pft."
==> "The terminator's invisibility belt is now completely out of power.%GDer Energievorrat des Tarngrtels ist jetzt völlig erschöpft."
You see, (ö) does not appear in words "völlig" and "erschöpft."
Screenshot:
https://nsa40.casimages.com/img/2020/09/24/200924123901494297.png
It is a big probleme for VOCAB file.
I need to translate this file with accented French letters (OEM code).

"Help me, Obi-Wan Kenobi. You're my only hope !"

My best,
Threepwang

[Kawa]

SCI, targeting DOS and being written mostly in DOS, uses ASCII-7 for American releases and DOS-437 for European releases. The font, if you look closely, is DOS-437 with the line and box characters removed.

But the game and engine don't really care what code page is used.

(Aside: I like to call them by number because "OEM" and "ANSI" are somewhat confusing terms. "ASCII-7" would comprise the first 128 characters, which DOS-437 and Windows-1252 add another 128 to.)

If you want SCI Companion (which uses Windows-1252) and the font (which is anything at all) to match, change the font. I have a bunch of them right here that I manually converted from DOS-437 to Windows-1252. Insert these fonts into a copy of SQ3 German, and you should be able to just type new lines in French, without worrying about encoding.

[Threepwang]

@ Kawa;
Hi,
I modified the Sierra fonts from SQ3 (DOS-437 to Windows-1252).
I put the fonts in the game folder (SQ3).
I cannot edit the VOCAB.900 file.
I have no problem with SCR and TXT files.
Only with VOCAB.990.
https://nsa40.casimages.com/img/2020/09/25/20092504575813357.png
Thank you for your help!

[Threepwang]

Continuation of my investigations ; I finally found the order file in German.
It is called VOCAB.910. I renamed this file to VOCAB.900 to try and transplant it into Space Quest 3.
But, I still have the same problem, I cannot edit the VOCAB.900 file with the accented French letters.

I also extracted the French vocabulary from Leisure Suit Larry 3 to try and transplant it into Space Quest 3.
But unfortunately it is too different. At the moment I'm still blocking 
I took the example of discussions here : http://sciprogramming.com/community/index.php?topic=1781.15

[ZvikaZ]

Hi.
That's a great idea, and I wish you luck!

2 friends of mine, and me, are about to finish Hebrew translation of SQ3.
We've already overcome all the technical problems.

If it's possible to translate to Hebrew, I'm sure it's possible to translate to French, which should be easier.
You might be interested to take a look at https://github.com/ZvikaZ/HebrewAdventure/ , to see the tools that I wrote for the tasks, and the translation files that we've got.
Currently these tools are tailored-made for my requirements, and even my specific environment; hopefully I will have time someday to make a general tool from them. However, they might help you overcome some of your problems.

Just be aware, that translating the texts in the "Texts" tab of SCICompanion isn't sufficient:
- There are text strings scattered in the scripts files (I have written a script that "export"s them all to a csv file; and another one then "import"s them back from the csv file; then you need to manually recompile in SCICompanion).
- There are texts drawn as pictures, in various "view"s
- You you'll probably need to modify the fonts, as you've already discussed
- The vocabulary by default needs to be 7 bits, I've overcome that problem with some hack, that probably works only in ScummVM, and not in Sierra's original interpreter (i.e., no Dosbox)
- There are items that are single word in English, but might be two words in French. I will give a Hebrew example - there is no word as 'anemometer' - it's called 'wind meter'. So, you need to take care of that.
- Furthermore, in Hebrew the grammar rules dictate to say "meter wind" instead of "wind meter", so I had to edit the grammar rules file, to make it aware of the Hebrew grammar.
- I also had to take care of the Right to Left display, instead of Left to Right - but that's no problem for European languages.
- There might be more issues, that I don't remember now...

All these issues can be solved, if you're ready to learn and investigate...
It'll be also easier to have some help from other French speaking Sierra fans who want to collaborate on this.

BTW,
First I built scripts that exported various things to csv file; then we created a Google Docs sheet with all the csvs, and the translation was done over there - it's much easier, allows collaboration, history, and more user-friendly for non-programmers; then I wrote a scripts that "import_all" (it's its name, IIRC), that downloads everythings from the Drive, runs the different "import" scripts, prompts to recompile in SCICompanion, and then copies the compiled files from SCICompanion.

I hope that you'll enjoy your Translation Quest!

[Threepwang]

@ ZvikaZ;
Hi ZvikaZ,
Thank you for your explanations.
"Space Quest 3 is definitely an international game"!
I find difficult to explain my problem, because "I am hopeless when it comes to figuring out computer problems".
In addition my English is so poor.

- I have no problem with the "TEX" files:
https://nsa40.casimages.com/img/2020/10/01/201001023817354205.jpg

- I have no problem with the "SCRIPT" files:
https://nsa40.casimages.com/img/2020/10/01/201001031155843459.jpg

- I have no problem with the "WIEW" or "PIC" files:
https://nsa40.casimages.com/img/2020/10/01/201001023816789655.jpg

- Overall, SCI Compagnon accepts accented letters French fonts in the game :
https://nsa40.casimages.com/img/2020/10/01/2010010421355601.png
- Except in the "vocab.900" :
https://nsa40.casimages.com/img/2020/10/01/20100102381221320.jpg
* SCI Compagnon refuses accented fonts in "vocab.900".
* I don't know how to add accented letters French fonts in "vocab.900".
* Example with "strange". I can write "etrange" but not "étrange".
* Do you know a solution for this problem?
I can not find a solution and I am lost 
If you have created a tool, of course I am more than interrested!

Thank you for your time.
Thank you very much!

[ZvikaZ]

Yes, I think I can help you.
I can't fix SCICompanion (maybe @Kawa can help here?), but I can give you two python scripts. You'll run the first one on the original vocab file, and it will create a csv file (which can be edited with Excel, or even better, with a text editor).
You'll make all your changes to the csv file. When you want, you'll run the other Python script, it will take the csv file, and build a new vocab file.
SCICompanion doesn't need to know about the new Vocab file! (as long as you only translate existing words, not adding new ones)
Just put it in your game directory (or PATCHES directory), and the game will be aware of the new translations.


Does it make sense to you?
(Do you know how to run Python scripts?)

BTW,
I'm assuming you're using the hybrid English German version, from Sierra. Am I right?

[Threepwang]

@ ZvikaZ;
There are 4 versions:
-VERSION (US) 1.018.
-VERSION (DE) 1.052.
-VERSIONS (US) 1.0P.
-VERSIONS (US) 1.0U.
Not all "resource" are compatible with each other.

I am using the German version 1.052. But, I prefer US version 1.018.
because she is more accessible by the gamer's. It's feel better for my futur French patch.

I don't know about Python scripts. But, I'll learn if this is the solution !

[ZvikaZ]

IIRC, the German version has different vocab file format than the American versions.
That's means, that before I will help you, you'll need to choose the version you're going to base your translation on.

[Threepwang]

@ ZvikaZ;
Yes ZvikaZ, you are right. I choose the US version.
I think that's quite a logical answer.
Because the German version is a hybrid rather rare.
"Thank you so much".

[ZvikaZ]

I've modified a little my script, to make it more suitable for you.

1.
Do you know how to use Git?

If yes, you need to clone 'https://github.com/ZvikaZ/HebrewAdventure.git', switch to branch 'sq3_french', and use the two files 'tools/vocab_import.py' and 'tools/vocab.csv'

If not, download these two files to some directory:
https://raw.githubusercontent.com/ZvikaZ/HebrewAdventure/sq3_french/tools/vocab_import.py
https://raw.githubusercontent.com/ZvikaZ/HebrewAdventure/sq3_french/tools/vocab.csv

2.
As you see, I've already created the csv file for you. Note, that it has a column of rooms. That's helpful for few reasons:
- Context. In some words, it's helpful to understand where it's used, to check the exact nuance.
- There are a lot of words, that aren't used in any room! No need to translate them
- The words are sorted according to room usage; I find it nicer, to be able to translate all the words required for the first rooms, check that they are working well, and then continue

3.
Open the csv file, and translate 'look' (or some other simple verb, so you can check right away that it's working).
The translation is just another word(s) in the entry.
For example:

Code: [Select]

"examine | gaze | inspect | look | observe | peer | scan | stare | view | בחן | הבט | צפה | השקף | שוף | התבונן | הסתכל | תביט | תסתכל | בדוק | תבדוק | תשקיף | תבחן | תצפה | תתבונן","IMPERATIVE_VERB | NOUN","1140","in 000, 002, 003, 004, 005, 006, 007, 008, 009, 010, 011, 012, 013, 014, 015, 018, 025, 027, 028, 029, 040, 041, 042, 043, 044, 045, 046, 047, 048, 049, 050, 051, 052, 053, 054, 060, 069, 070, 071, 072, 073, 074, 075, 080, 081, 085, 086, 090, 091, 092, 093, 094, 095, 290, 420, 421, 470, 501, 503, 600, 690, 700, 702, 703","","",""

4.
It's legal to have several '|' in raw.
You can write comments in the columns after the 'rooms' column. It's very helpful.

5.
Now, install Python (Google it).

6.
Open the 'vocab_import.py' script in an editor.
Replace the

Code: [Select]

ENCODING = 'windows-1255'line with whatever encoding you're using.

7.
Run the script (depends on your installation - maybe just click it in Windows Explorer, maybe run in a 'cmd' prompt something like:

Code: [Select]

python vocab_import.py)

8.
It will create two files in the local directory:
'vocab.000' - should be used by ScummVM (you can't use original Sierra SCI in DosBOX anymore, sorry...)
'vocab.900' - if you want SCICompanion to know about your changes.

You can ignore for now the .900 file.
Make a copy of the game aside.
In one directory, you'll continue to work with SCICompanion.
In the other directory, you'll run ScummVM to play the game (after copying patch files from the first directory).

9.
Copy the 'vocab.000' created by the script to the ScummVM game directory.

10.
Play the game and check that it understands your translated word.

11.
Next time, you need to repeat steps 3,7,9 and 10.

12.
Enjoy, and it's OK to ask questions if something isn't clear...

[Threepwang]

@ ZvikaZ;
Hi ZvikaZ,
Yes I'm member of Git, okay, it's in the box.
I'd need to look at all of this pretty carefully.
But I already know that I'll get back to you 
I can not thank you enough!

[Threepwang]

@ ZvikaZ
Hello ZvikaZ,
I am currently working on the scr files. Do you advise me to work directly on the scr files your patch in Hebrew or on my game?
Because I noticed that you have invalidated many words:
https://nsa40.casimages.com/img/2021/01/05/210105021243601760.jpg

Happy New Year to you and to everyone members of sciprogramming !!
My best !

[ZvikaZ]

Hi.

I'm sorry, but I'm not sure that I understood your question, and I'm not sure what do you mean by saying that I've invalidated many words.

Anyway, if I do understand correctly - certainly - don't use my Hebrew patch files.
It's better to base your translation on clean fresh English SQ3.

You indeed might take inspiration from my work, but don't use it directly.
Because of few reasons:
- It's better to have shorter path. Why have an English -> Hebrew -> French path, instead of English -> French? The longer the path you're going, more complications get involved, and more potential bugs.
- English to Hebrew is much more complicated than English to French - it's using a different font (while in French it's the same font, or maybe you'd just need to add few accented letters, but the majority will be the same)
- Even worse - Hebrew is written from Right to Left. Part of my code supports it, and it will be wrong if you'll base French on the Hebrew...

I've linked to my GitHub repo not for the SQ3 directory, but to the TOOLS directory.
Using these tool might help you.

Also note, that I haven't directly created any of these patch files - my tools did it.

Let me repeat the method that I've worked:
- My tools have extracted the strings, with small manual help
- The output of the previous step was an Excel file, uploaded to Google Drive
- The translation team have created the translation in the spreadsheet - not in SciCompanion, or similar tools
- My tools imported the spreadsheet from Google Drive, and created the required patch files (again, with small manual step, because SciCompanion cannot be run in batch mode, and a recompilation was required)

Good luck with your work, and don't hesitate to request further assistance.

[ZvikaZ]

Further thinking about it, you can just take the translation excel https://github.com/ZvikaZ/HebrewAdventure/raw/master/sq3/PATCHES/sq3.xlsx , ignore everything in Hebrew (or better - delete all the Hebrew...), and just translate it to French.
I will help you with creating the required patch files, after that.