-
Notifications
You must be signed in to change notification settings - Fork 1.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
ro-RO: Update romanian translation #2256
base: main
Are you sure you want to change the base?
Conversation
Hi! I just finished the review (sorry for waiting so long, it took quite a bit to go through all the changes). I'm not super familiar with the .po files, so please take everything that I mention with a grain of salt and feel free to ignore it if it doesn't make sense :D I couldn't find a way to leave comments easily (since the diff is so large), so I'll mention the line (from the newest version) for each note:
|
I also apologize for the super slow response. This is my first attempt to review something like this and I'm still struggling to find a convenient way to review this. This is a partial review, up to line 721 (ID 174). What I tried was to split each message in separate files and then diff them assuming that the IDs are the same between the base of the PR and the PR. That doesn't seem to be the case though because from ID 78 things are diverging (the base has
As @AlexandraImbrisca also mentioned, I find the term "Jurnarlier" a bit confusing in this technical setting.
I believe
It seems that the term "owned" is translated as "deținut" or "propriu" in various places. Is there a rule we are following?
I think "Actualizare implementărilor" might be better because both the client and server code needs to be updated (ref) .
It seems that we use the terms "vector" in a bunch or places already but I'm not sure that's quite the right one because Rust seem to also have a std::vec::Vec which the docs claims it's a short name for vector. The mathematical explanation in the English dictionary seems to be "an arrangement of quantities or symbols in rows and columns; a matrix" so maybe the term "matrice/matrici" might be more suitable. I'll try to find time to make progress in a more timely fashion. 😊 |
Thank you for your review. You have caught some very important issues.
Translation was initially performed with DeepL and messages are marked as
`needs work`. This means that afterwards I've inspected each message and
only for those that I have corrected I removed the 'fuzzy' mark. Of course,
some mistakes happened. Only corrected the observations in the not-fuzzy
strings.
Regards,
Răzvan
…On Mon, 19 Aug 2024 at 00:43, Alexandra Imbrisca ***@***.***> wrote:
Hi! I just finished the review (sorry for waiting so long, it took quite a
bit to go through all the changes). I'm not super familiar with the .po
files, so please take everything that I mention with a grain of salt and
feel free to ignore it if it doesn't make sense :D
I couldn't find a way to leave comments easily (since the diff is so
large), so I'll mention the line (from the newest version) for each note:
- Line 302 (and other occurrences later): when I read "Jurnalier," I
don't connect it with "Logger." It might be my limited experience reading
Romanian technical content, but I thought it was worth mentioning that I
find it slightly confusing
Fixed.
- Line 721: "array types" is translated to "tipuri de tip vector".
Could it be simplified to "tipuri de vectori" instead?
Fixed
- Line 1507: "gap" is translated to "interval". "Interval" is a bit
generic, maybe "interval de pauza" instead?
Fixed. 'gap' -> 'pauză'
- Line 3298: "n>1" is translated to "n>2". Shouldn't the condition
remain the same?
Fixed.
- Line 4439: "Hello" is not translated. I'm not sure if it's a
problem, but I saw it translated in almost all the other places
Not sure it affects tests. Marked as needs work.
- Line 4553: the translation "tuturorcoordonatelor" doesn't contain a
space between the 2 words
Fixed.
- Line 4671: "Tuple" is translated to "cuplu" even though it's
referenced as "tuplu" in most of the other places
Fixed.
- Line 5070, 5074, 5078, 5120, 5124, 5128: from the context, the
sentences seem to be referring to an elevator (correctly referenced in the
translations), but the msgids reference a car instead
Fixed.
- Line 5991: "greet is provided" is translated to "green este
furnizat". Greet seem to be translated incorrectly
Fixed.
- Line 6567: "virtual method table" is translated to "tablele de
metode virtuale". Tablele is incorrect (maybe "tabel" instead?)
Fixed.
- Line 7639: the translation contains double "cu" ("casting cu cu")
which sounds incorrect
Fixed.
- Line 10806: "crates" is translated to "crates-uri" instead of
"crate-uri"
Fixed.
- Line 11356, 22737, 25429, 25564: commented msgids. Should they be
removed?
Will fix this when I update messages.pot.
- Line 14125, 14159, 14187, 17229: the translation of the msgids don't
match with the msgstrs. Should the msgids contain the additional details
included in the msgstrs?
Auto-translated strings, not reviewed, marked as needs work.
- Line 21284, 21342: "crate" is translated to "crăiasa". Should it
remain "crate" as in the other places?
Fixed. Auto-translated strings that were marked as needs work.
—
Reply to this email directly, view it on GitHub
<#2256 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AA3WDVNHBD6DMRG2NZN4TC3ZSEIRTAVCNFSM6AAAAABLWMSVI2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEOJVGQYDKMJQHE>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
Thank you for the feedback.
Same comment as in Alexandra's review: translation was initially performed
with DeepL and messages are marked as `needs work`. This means that
afterwards I've inspected each message and only for those that I have
corrected I removed the 'fuzzy' mark. Of course, some mistakes happened.
Only corrected the observations in the not-fuzzy strings.
On Mon, 19 Aug 2024 at 03:38, Razvan Musaloiu-E. ***@***.***> wrote:
I also apologize for the super slow response. This is my first attempt to
review something like this and I'm still struggling to find a convenient
way to review this. This is a partial review, up to line 721 (ID 174).
What I tried was to split each message in separate files and then diff
them assuming that the IDs are the same between the base of the PR and the
PR. That doesn't seem to be the case though because from ID 78 things are
diverging (the base has msgid "Exercise: Generic `min`" for that ID while
the PR has msgid "`dyn Trait`"). What I did for now is use git diff -u
HEAD^..HEAD po/ro.po and look at the chunks. That's what I'm using below.
------------------------------
@@ -302,9 +299,8 @@ msgid "Deriving"
msgstr "Derivarea"
#: src/SUMMARY.md
-#, fuzzy
msgid "Exercise: Generic Logger"
-msgstr "Exercițiu: Logger generic"
+msgstr "Exercițiu: Jurnalier generic"
#: src/SUMMARY.md
msgid "Day 2: Afternoon"
As @AlexandraImbrisca <https://github.com/AlexandraImbrisca> also
mentioned, I find the term "Jurnarlier" a bit confusing in this technical
setting.
Fixed.
------------------------------
@@ -334,6 +330,10 @@ msgstr "Limite de trăsături"
msgid "`impl Trait`"
msgstr "`impl Trait`"
+#: src/SUMMARY.md src/generics/dyn-trait.md
+msgid "`dyn Trait`"
+msgstr "`din Trait`"
+
#: src/SUMMARY.md src/generics/exercise.md
msgid "Exercise: Generic `min`"
msgstr "Exercițiu: 'min' generic"
I believe dyn is a keyword in Rust so we should not translate it.
Fixed.
------------------------------
@@ -466,8 +465,8 @@ msgid "`Rc`"
msgstr "`Rc`"
#: src/SUMMARY.md src/smart-pointers/trait-objects.md
-msgid "Trait Objects"
-msgstr "Obiecte cu trăsături"
+msgid "Owned Trait Objects"
+msgstr "Obiecte cu trăsături proprii"
#: src/SUMMARY.md src/smart-pointers/exercise.md
msgid "Exercise: Binary Tree"
It seems that the term "owned" is translated as "deținut" or "propriu" in
various places. Is there a rule we are following?
------------------------------
@@@ -705,30 -709,37 +705,30 @@@ msgid "Changing API
msgstr "Schimbarea API-ului"
#: src/SUMMARY.md
-#, fuzzy
msgid "Updating Implementations"
-msgstr "Implementare"
+msgstr "Actualizare implementare"
I think "Actualizare implementărilor" might be better because both the
client and server code needs to be updated (ref
<https://google.github.io/comprehensive-rust/android/aidl/example-service/changing-implementation.html>)
.
Fixed. "Updating Implementations" -> "Actualizarea implementărilor"
------------------------------
#: src/SUMMARY.md src/android/aidl/types/arrays.md
-#, fuzzy
msgid "Array Types"
-msgstr "Array-uri"
+msgstr "Tipuri de tip vector"
It seems that we use the terms "vector" in a bunch or places already but
I'm not sure that's quite the right one because Rust seem to also have a
std::vec::Vec <https://doc.rust-lang.org/std/vec/struct.Vec.html> which
the docs claims it's a short name for vector. The mathematical explanation
in the English dictionary seems to be "an arrangement of quantities or
symbols in rows and columns; a matrix" so maybe the term "matrice/matrici"
might be more suitable.
Yeah, this is something I was not very fond of. Your suggestion sounds
better. Fixed.
… ------------------------------
I'll try to find time to make progress in a more timely fashion. 😊
—
Reply to this email directly, view it on GitHub
<#2256 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AA3WDVKPZFNYSBZON45ZY3TZSE5BLAVCNFSM6AAAAABLWMSVI2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEOJVGQ3DKNZYGM>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
Do you have a script for this? If not, https://crates.io/crates/cloud-translate might be useful for you. This is a tiny crate I wrote which takes a PO file and sends the content through Google Translate (we could add other backends). The result is then written back to a well-formed PO file, with all new entries marked as fuzzy. If you have an automated approach for this, then we could mention it in our documentation! It would be cool to give people an easy way to bootstrap a new translation. |
Yes, the huge PO files break the normal GitHub workflow. I would like to see us use a translation platform instead, but that requires work in finding one which we can integrate into our workflow (#1305). What I do myself is to use command line tools. Specifically, I like gh pr diff 2256 | bat -l patch which very quickly shows me the diff, regardless of the size. |
I just used the POedit included feature, so no automation.
Thank you for pointing to the cloud-translate crate.
…On Wed, 21 Aug 2024 at 10:20, Martin Geisler ***@***.***> wrote:
Translation was initially performed with DeepL and messages are marked as needs
work.
Do you have a script for this? If not,
https://crates.io/crates/cloud-translate might be useful for you. This is
a tiny crate I wrote which takes a PO file and sends the content through
Google Translate (we could add other backends). The result is then written
back to a well-formed PO file, with all new entries marked as fuzzy.
If you have an automated approach for this, then we could mention it in
our documentation! It would be cool to give people an easy way to bootstrap
a new translation.
—
Reply to this email directly, view it on GitHub
<#2256 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AA3WDVI5J4KOYF6PUN74BN3ZSQ5SBAVCNFSM6AAAAABLWMSVI2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMBRGMYTOMRSHA>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
Another random tip, since I ran into this recently: see #1948 (comment) for a multi-step procedure to hand-merging out-of-date |
Hi @razvanm, does this look okay to merge now? |
It's a long weekend in the US due to Labor Day and I hope I'll finish maybe a complete pass over it. Sorry for the delay! 😊 |
I figured out a way to diff that using This is a review for c0447d3 and it only covers the IDs from 175 to 770. I believe there are 450 more diffs I haven't looked at yet. Once again, I apologize for the slow speed. Note, unlike before, the diffs from below only show the diff between the translations. I hope indicating the ID would help to locate the right entry. ID: 176 msgid:
diff:
I'm not sure this should be translated because the term Parcel seems to be a particular type of Java object in Android (reference). Maybe "Obiecte Parcel"? ID: 351 msgid:
diff:
I think "vectori" should be "matrici" as discussed earlier. ID: 487 msgid:
diff:
I think "Incluzând" sounds odd. How about "Împreună cu"? ID: 553 msgid:
diff:
I think "Inferența tipului" or "Inferența tipurilor" would sound a bit more natural. ID: 604 msgid:
diff:
I'm not sure "înfășoară" is the right term for "wraps". I haven't researched this much but one translation that seems more suitable is "a o lua de la capăt" that is used here. -- ID: 617 msgid:
diff:
I think it would be good to use backticks around "if" like we do in ID 619. ID: 620 msgid:
diff:
We don't seem to be very consistent in the way the term "scope" is translated. ID 30 for example uses the term "scopuri". I think "domeniu de aplicare" is long but a reasonable choice. ID: 724 msgid:
diff:
I think "vectori" should be "matrici" as discussed earlier. ID: 727 msgid:
diff:
Like for ID 487, I think "Împreună cu" would sound a bit better than "Incluzând". ID: 728 msgid:
diff:
Same as for ID 724, I think this should be "Matrici". ID: 731 msgid:
diff:
We use the term "șabloane" (and not "modele") for "patterns" in other places (ID 57 for example). ID: 748 msgid:
diff:
We use the term "șabloane" (and not "tipare") for "patterns" in other places (ID 57 for example). ID: 770 msgid:
diff:
I'm not sure the "în suspensie" translation for "dangling references" is quite right because that particular form is typically used in chemistry (dexonline). The term "referințe suspendate" seems to be mentioned in one slide deck I found on the web (slideshare). A term that doesn't seem to exist that I would have proposed is "referințe orfane". |
0f9a45b
to
3afa569
Compare
Thank you @razvanm for taking the time to look over the commit. I've included your observations. Hopefully I didn't missed anything. Some guidance on translating terms:
|
This is a review for 3754e09 and it only covers the IDs from 770 to 1054. I think there are 363 more diffs to go. Once again, I apologize for the slow speed. 😔 ID: 771 msgid:
diff:
In other parts se seem to use the word "posesie" for "ownership". In here the work "proprietate" is used. ID: 783 msgid:
diff:
The "o imagine a unei colecții mai mari" doesn't seem to capture the well the meaning of "view into a larger collection". Maybe we can say "o porțiune dintr-o colecție mai mare"? ID: 789 msgid:
diff:
As discussed earlier, the preferred term for "array" is "matrice/matrici". ID: 795 msgid:
diff:
The "octeți codificați UTF-8" sounds incorrect because the buffer is what is encoded in the UTF-8 format (and not the bytes). Maybe we can say "un buffer propriu de octeți, în format UTF-8"? ID: 802 msgid:
diff:
I see the term "wrapper" translated as "înveliș" in a bunch of places that this PR is not touching. I find the term "înveliș" unsual and I think somethink like "adaptor" or "container" would sound better. ID: 805 msgid:
diff:
I think "prin selectarea intervalului" would sound better as "prin selectarea unui interval". I also think the "aliniat la limitele caracterelor" would sound better as "aliniat cu frontierele caracterelor". For "în locul încercării de a obține limite de caractere corecte" I would suggest "în locul încercării de a obține corect frontierele caracterelor". ID: 806 msgid:
diff:
Similar with ID 795, for "poate conține numai octeți codificați UTF-8" I would suggest "folosește întotdeauna formatul UTF-8". ID: 841 msgid:
diff:
This is one of the places that uses the term "inveliș" I mentioned above for ID 802. ID: 876 msgid:
diff:
I think something simpler like " ID: 878 msgid:
diff:
The third sentence doesn't seem to be present in the English text. ID: 900 msgid:
diff:
Note that in the other places we use the term "lift" and not "ascensor". ID: 901 msgid:
diff:
Same note about the term "lift" vs "ascensor". ID: 903 msgid:
diff:
Same note about the term "lift" vs "ascensor". ID: 904 msgid:
diff:
Same note about the term "lift" vs "ascensor". ID: 906 msgid:
diff:
Same note about the term "lift" vs "ascensor". ID: 907 msgid:
diff:
Same note about the term "lift" vs "ascensor". ID: 908 msgid:
diff:
Same note about the term "lift" vs "ascensor". ID: 912 msgid:
diff:
Same note about the term "lift" vs "ascensor". ID: 915 msgid:
diff:
I mentioned earlier that we use the term "șabloane" (and not "tipare") for "patterns" in other places (ID 57 for example). ID: 923 msgid:
diff:
Similar with ID 487 and ID 727, I think "Împreună cu" would sound a bit better than "Incluzând". ID: 934 msgid:
diff:
Based on https://github.com/google/comprehensive-rust/blob/main/src/pattern-matching/match.md this string shows up as the explanation for matching "q". Did we also update the code to match the new string? ID: 973 msgid:
diff:
We use the term "șablon" in other places where we talk about pattern matching. ID: 981 msgid:
diff:
This is used for a ID: 989 msgid:
diff:
Shouldn't "returns" be "returnează"? ID: 995 msgid:
diff:
Should we use "cursanți" instead of "elevi" like we do in a few other places in this PR? Also, I think the last sentence would sound a bit better as "Rescrieți versiunea complicată pentru a ajuta cursanții să vadă transformarea." ID: 999 msgid:
diff:
Note that the translation for "syntactic sugar" also needs be fixed in ID 3500. ID: 1021 msgid:
diff:
As far as I can tell ID: 1025 msgid:
diff:
I'm not sure the "receptori posibili" translation for "common receivers" is correct. Maybe we can say "receptori populari" sau "receptori frecvenți"? ID: 1028 msgid:
diff:
Nit: the term "dealocat" sounds a bit better than "dezalocat". The term doesn't seem to apear in any Romanian dictionary yet (at least not the ones I was able to search) and a blog post from 2012 (post) reaches the same conclusion as me. For the last sentence I would suggest "Proprietatea completă a unui obiect nu înseamnă că obiectul este mutabil". ID: 1041 msgid:
diff:
Typo: "companiel" should be "companie". ID: 1048 msgid:
diff:
Should "block" be translated to "bloc"? ID: 1054 msgid:
diff:
We use the term "cursanți" instead of "studenți". The last sentence might sound a bit better if we clearly state the subject: "Moștenirea de trăsături specifică [...]". |
Thank you @razvanm for the comments. I've updated the translation. I revised some terms on the whole translation and made replacements for "șabloane" and "cursanți". I'll wait for the rest of the review. |
1a4d563
to
7de5fb5
Compare
Signed-off-by: Razvan Stefanescu <rstefanescu@google.com>
@AlexandraImbrisca and @razvanm, do you think we could merge this? It's been out for review for quite a while and I see lots of small updates since then. Basically, if this is a step forward, then let's merge it. It's easier to do small incremental updates afterwards than to juggle a huge PR like this over many months. |
As I indicated in my previous comment, I only reviewed the changes up to ID 1054 and I think there were 363 more diffs to go. My preference would be to split this in two: keep in this PR the part reviewed and create another PR for the rest of the changes. If we get everything in as is it will make it harder at least for me to look at the changes made here. From what I can tell so far, it will be valuable to look carefully at the rest of the changes. 😊 Thoughts? |
I think that is a good idea! |
More strings translated and checked for Romanian translation