Create translation_RO.json

[discip]

as requested by @cipyyy here #1026

@sandmanRO
Could you please have a look at this and translate the bits, which were not translated yet?

[discip]

@Ralim
Good morning Sir,
could you please help me finding the cause of this failing?

thanks in advance

[Ralim]

The current OLED font does not have the symbols {'„', 'Î', '”', 'ş', 'ț', 'â', 'Ă', 'ș', 'ă', 'î'} defined

[discip]

@Ralim
Thank you for the spontaneous reaction! 😃👍

1. Where does this 'result' pop up, what am I missing?
   I do not want to bother you every time, so I would like to be able to find it myself. 😁
2. Is there a way to fix this, or do I have to remove those symbols?

[sandmanRO]

Oh, it is Romanian. It needs Latin extended in order to work.

[sandmanRO]

I believe you also recently added some new entries not included with this translation.

[discip]

@sandmanRO
Yes, I thought about that too, and changed that bit, but without any luck.
If I got @Ralim right, there are no such symbols in any defined tables used here yet.

[Ralim]

Ah you can see it by clicking on the red X->details then going to the logs of one of the builds.
Then in the logs you can see KeyError: "Symbols not found in specified fonts: {'ț', '”', 'ș', '„'}"
Which is its way of saying that the letters ț ” ș „ dont exist in here

If you draw them using the bitmap editor (html file in that folder) or any other tool, export the hex and add them to that file then it will work.

[sandmanRO]

I thought I saw those symbols in font_tables.py but I might be wrong.

[sandmanRO]

by the way discip, are you building under Windows?

[Ralim]

I thought I saw those symbols in font_tables.py but I might be wrong.

Could be identical looking but different codepoint symbols. Thats happened before 🙃

[discip]

by the way discip, are you building under Windows?

Yes I do.

[sandmanRO]

I see...the reason I was asking about Windows...under Windows the default java / python encoding is not UTF-8 but iso-8859-1

[sandmanRO]

so you have to modify some of the files you are using (I will check in a second their name) to explicitly specify the UTF-8 encoding

[discip]

I see...the reason I was asking about Windows...under Windows the default java / python encoding is not UTF-8 but iso-8859-1

I downloaded the file provided by @cipyyy and edited it in notepad++.

[sandmanRO]

Yes, but I believe the job of adding the language to your firmware is done inside make_translation.py (my apologies if I'm wrong)

[sandmanRO]

...and that is python code which would try to load a UTF-8 encoded file as iso-8859-1

[discip]

@sandmanRO
Sorry, but I am not building anything on my machine, since I change the files here and the rest is done automatically. 😁

[sandmanRO]

I see...I'm not familiar with the way GitHub works

[discip]

@sandmanRO
Could you please tell me, if there are alternatives for these letters?

ș ț

[sandmanRO]

s and t

[Ralim]

So for the CI here, these are all built under the latest LTS ubuntu release, and all production firmware / testing etc uses Linux for the build pipeline.
So no matter how @discip edits the file, it will be interpreted under linux.

[sandmanRO]

The use of diacritics is not critical... I personally don't use them when I write in Romanian on computer

[discip]

@sandmanRO
Would it be a great deal, if I would use s & t instead of going the way @Ralim proposed to do it by drawing and adding the missing characters to the actual map?

[sandmanRO]

Thank you @Ralim. I knew nothing about GitHub until not so long ago (not that know more now)

[sandmanRO]

@discip, 100% it's ok to replace the special characters with their plain ASCII "counterpart".

[Ralim]

Github is a massive thing that has grown massively.
The Continious Integration (CI) is the checks, and they are there so that you cant merge in changes that break the build for any device on any language. (I cant actually click the merge button unless they all pass).

It does make things like this annoying, but it keeps the main codebase "working" (roughly, it cant spot code issues that still compile).

I would go for just subbing in closest characters for now and then if desired someone can loop back and add the better ones later on

[sandmanRO]

line 210: "text2": ["Inverseaza", "culoarea"],
line 211. "desc": "Inverseaza culoarea ecranului"

Sorry for my slow reaction (in my defence, here is about 4:30AM)

[discip]

line 210: "text2": ["Inverseaza", "culoarea"],
line 211. "desc": "Inverseaza culoarea ecranului"

Sorry for my slow reaction (in my defence, here is about 4:30AM)

No worries, did already change it.
Thank you. 👍

[discip]

@sandmanRO
Please have a final look over this once more, before asking @Ralim to merge this in. 😊

And take your time please, you have not to do it now. 😃

thanks in advance

[Ralim]

I'll wait for Sandman to wake up and confirm this looks good, then will merge this in :)

[cipyyy]

Hello,
Thank you very for the interest and for the effort made to correct the translation.
Good luck!

[sandmanRO]

Hello Gentlemen, I would do few small changes:

Line 171: "desc": "Intensitatea puterii la păstrati-pulsul-treaz (watt)"
I would replace with:
"desc": "Puterea pulsului de mentinere activa a blocului de alimentare (watt)"

Line 195: "desc": "Întârziere înainte de declansarea pulsului trezit (x 2.5s)"
with
"desc": "Perioada pulsului de mentinere (x 2.5s)"

Line 199: "desc": "Păstrati-pulsul-treaz durată (x 250ms)"
with
"desc": "Durata pulsului de mentinere (x 250ms)"

There is no "good" translation for "keep-awake". Obviously you can translate each word, as originally done, but it would not have too much sense (as technical term)

[discip]

Will change this later.

[discip]

@Ralim
Should be good to go now. 😊

[cipyyy]

It is very good.
Thanks!

[discip]

@sandmanRO

Thank you for sorting this out! 😃👍