I think the deduplication script ignores some sentences. for example: #383500 & #1089048 and #2206464 & #2128669, which were exactly the same before the script was used, so they were ignored by the script, apparently. Those two pairs are the only ones I noticed, I don't know if there are other sentences that have been ignored by the script

Concerning #4233047, where the sentence has been automatically truncated because it's a long sentence, I wonder if it would be possible to broaden the characters limits, otherwise it won't be possible to have complete translations if the original sentence is quite long

> You mean https://tatoeba.org/sentences/s...eng&to=ita ?
yes, actually

+1, i would find it better to have the opportunity of making exact searches instead of using "=word" each time i want to see the exact occurrences of something

would it also be possible to search for given words/expressions that are not translated in a given language? for example: I want to search for "once in a blue moon" (or any other expression in any other language) and I want to see all the sentences containing that expression that are not translated in Italian (or any other language). I would also find it useful if i could see all the sentences with a given expression/word that are translated in a given language. for example: i search for "apple pie" and i want to see only the sentences containing "apple pie" that have translations in Italian

thanks for the fix, gillux. everything seems to be perfectly working for me now

I found an issue with accents. First query: https://dev.tatoeba.org/ita/sen...rom=ita&to=und

It becomes this query when I search for my sentences corresponding to that query: https://dev.tatoeba.org/ita/sen...ser=Guybrush88

As you can see, no results are shown because the accent is changed by the query, while sentences I own are shown without specifying my username

probably 29984 is correct, i just noticed that yesterday morning there has been a deduplication, and probably the number in the progress bar hasn't been updated yet

Thanks for your answer, Trang

*Bug report*
I noticed that the date in sentences, wall posts and comments older than around one month isn't localized, even if the interface is.
An example: http://tatoeba.org/ita/sentences/show/1120890
The dates in the sentence history aren't localized at all, even if the rest of the interface is

well, maybe you can rephrase the idiom when translating, if there aren't equivalent idioms in the second language. In the translation you could simply explicit the meaning of the idiom, or maybe you can translate like in the examples you posted, if you think there aren't exact equivalents and the translation you came up with can convey the meaning of the original sentence you want to translate

The best solution, imho, would be to find the equivalent ones in your target language. For example, if you're translating an English idiom into Vietnamese, find a Vietnamese equivalent of that idiom

I agree for a more frequent deduplication, in order to have more frequently a cleaner corpus. I'm already aware of this discussion: https://github.com/Tatoeba/tatoeba2/issues/569
and I hope it will actually be implemented

I just saw another issue with translations, this time concerning Italian.
This is how this string has been translated on Transifex: http://i.imgur.com/d9LqFFs.png
This is how it actually appears on the website: http://i.imgur.com/fb0SYWr.png

Congrats!

A recent example: https://tatoeba.org/ita/sentences/show/4162900
I manually changed the flag, but it was automatically recognized as italian

yes, it was before that restriction. I had just disowned all my sentences in foreign languages (I had just sentences in Italian and in Latin), and anytime i added a sentence in English it was recognized as Italian

i stumbled upon the same issue some time ago when adding some sentences in English. They were recognized as Italian, even if they didn't even have any italian words in them, so i thought the auto-detection was depending on one's sentences rather than the whole corpus

A more urgent change before deduplicating is, imho, to make apostrophes and spaces before punctuation (especially for french) the same. There are sentences that are exactly the same (same wording, same punctuation) but can't be merged because there are different apostrophes or different spaces before punctuation.

for example: #50984 and #2176373
#1850008 and #11578

and i'm sure there are many more

The script works for any language. Generally it's used once a month. Even if you add the same sentence a million times, this script will merge all the sentences so you'll have just one when it's done. You don't have to worry if you translate different sentences and they happen to have the same translation. With this script, you'll have just one sentence linked to all the sentences you translated with the same sentence