I think a great future enhancement would be "Search for <string1> from <fromLanguage> to <toLanguage> where results contain <string2>". For example,

Search string: "wonder if" {acaba}
From: English To: Turkish

Results: Sentences in Turkish containing "acaba" where the corresponding English sentence contains "wonder if".

🇬🇧 I wonder if I should trust Tom.
🇹🇷 Acaba Tom'a güvensem mi?

I inadvertantly typed in a hyphen (-) instead of the equals sign (=) in my search string:

-personally (instead of =personally)

The URL is: https://tatoeba.org/en/sentence...sonally&to=tur

Obviously this means I wouldn't expect to get any results because a search string like "-personally" is effectively equivalent to searching for an empty string. Instead of getting a more user-friendly, explanatory message, I got:

An error occurred while performing the search. If the problem persists, please let us know and include the error code "65fb71bbcc781" in your message.

I suggest showing the user a more meaningful message including a link to the Tatoeba-specific regexp help page.

Thanks!

What is the regexp-ish expression to search for sentences containing the string "sizin" (without the quotes) but only if it does NOT occur by itself? In other words, I want to see sentences containing qsizin, aabbccsizin, zzzsizin, pqrssizin, but NOT sentences like "aaa bbb sizin ccc".

This linked sentences stuff seems to be something that the developers of the site care about. As a regular user, all I care is that search results return sentences that are consistently accurate translations of each other. I don't want to see "indirect sentences" and what not. Those should be suppressed by DEFAULT.

OK, but when I raised a similar issue before, defenders of the "feature" thought it was fine to leave the sentences as they were:

https://tatoeba.org/en/wall/sho...3message_38960

Query:

https://tatoeba.org/en/sentence...+one%22&to=tur

One result:

eng
How did you know that was my favorite one?
tur
Onun en sevdiğim film olduğunu nasıl bildin?

Once again, I see a translation that, due to linking or what not, connected two sentences that aren't accurate translations of each other. The English sentence has "my favorite one", but the Turkish translation has "my favorite film" (en sevdiğim film)!

Honestly this feature/bug sucks.

The minus sign before the word works only if the unwanted word is in the same language as the "from" language. I should revise my question:

Is there a way to do a search for X in language P and bring back results in language Q that do not contain Y in language Q?

In my example, I am searching for "if you need" (quotes included) from English to Turkish, but I don't want any results that contain the word "ihtiyacın" which is a Turkish word.

Is there a way to do a search for X and bring back results that do NOT contain Y?

Example:

https://tatoeba.org/en/sentence...need%22&to=tur

I don't want results that contain the sub-string "ihtiyacın". Is that possible?

All Turkish sentences with the word "Pazar" (P in uppercase) that is NOT in the context of a date, and NOT the first word in the sentence, should be changed to "pazar" (p in lowercase). This is based on rule 26 in:

https://www.tdk.gov.tr/icerik/y...ildigi-yerler/

This applies to other days of the week (pazartesi, salı, çarşamba, perşembe, cuma, cumartesi) and months of the year (ocak, şubat, mart, nisan, mayıs, haziran, temmuz, ağustos, eylül, ekim, kasım, aralık).

What about the use of '\' as an escape character just like in true regexp?

Even if all punctuation is thrown away before indexing, it would be nice if an escaped hyphen in the search pattern is recognized as a literal hyphen and the returned results further filtered, based on the search pattern ("*\-rich"), to return only those that DO contain a hyphen immediately before "rich".

Either there are no English sentences containing "culturally-rich" or the pseudo-regexp "*-rich" (double quotes included) is able to extract such a text pattern.

"Hover" implies a mouse pointer over an area on the screen. There is no mouse pointer when accessing a web page in a browser on a smart phone or tablet. A long click or long press might work.

@AlanF_US

You're right. Language code "ron" is missing in the comma-delimited list.

Why is it "ron" and not "rom" for Romanian, by the way?

@Shishir
@CK

OK, I think there's a bug. My profile has English, Malay, Turkish, Romanian shown in the Languages section.

My search URL is:
https://tatoeba.org/en/sentence...3DGoing&to=ron

I see the translations in Romanian. In the "Show translations in" drop-down, I select "None" and then hit "Search". The results are sentences in English without any translations. That's good. That's what I want.

I then tap on "#" of one of the English sentences, and the next page shows only Turkish translations—even though Romanian is one of the Languages in my profile.

@Shishir
Thanks. I was able to only show the English sentences without any translations.

https://tatoeba.org/en/sentence...roved=no&user=

What do you mean by "If you want to see the translation you open the sentence"? How do I "open the sentence"?

It would be a good learning and self-testing feature if a parameter can be added to the URL to blank out the target translation with a checkbox displayed at the end of the hidden sentence. When you check the checkbox, the translation is revealed. When you uncheck it, the translation is hidden again. Add to feature list? Please? 😊

For the purpose of testing myself, is there a way to force the results of a search to be displayed WITHOUT the target translation. Example:

https://tatoeba.org/en/sentence...=%3Dsaw&to=ron

I want the results to show only the English sentences with the Romanian translations [temporarily] suppressed. Is there a parameter I can add to the URL?

Why is the Tatoeba search textbox shortened, leaving an empty green space to the left of it?