Please, insert the period before those signs “.#}”. Just make sure that the period is needed there at all and the sentence doesn’t continue in the next segment.
What’s concerned quotes. If you see that there are single quotes and they are only in the target text – it’s ok. If you consider they are extra somewhere you can remove them.
okay, forget what I said about Gerald or Sebastian having to proofread. I did it myself and it went quite fast.
Regarding the glossary:
My thinking was to first run the list on my latest glossary, the German column, I will send it to you. That way, we have a reference for the people spotchecking, even if it’s not proofed. Whoever does the spotchecking should of course be able to use some of their own judgment to tell if the glossary terms that have been S&R-ed are good (not blindly apply them). That’s because I don’t think we have the resources to fully proofread the glossary after insertion but still need a final reference. It should be run without the $$ or anything, too.
I will now integrate Sebastian’s stuff and after that, I’ll just call it final. Perhaps we can apply stuff that might still come up later.
Another issue: The term “Arcane” like in column 47, 48 “1 pair of Arcane Boots” seemst to have been interpreted by some translators as being a adjective in the sense of ‘mysterious’, others interpreted it as a noun describing a people or something (accordingly they translated it in German with a hyphen, like “Arcaner’s-Boots”, Arcaner-Stiefel)
Which version is the correct one, or is this to be ignored?
I run the S&R list through the glossary.
I skipped the previous S&R list (the one found in SandR.zip) and assumed that that list are obsolete.
I then copied the relevant data from the glossary. I ran the script using the “no spotting required” data. Then ran the script again using the first pass (Â§Â§Â§) data.
I can’t fathom what you are trying with the second pass ($$) data. As far as I can tell, they are identical with the first pass data. The whole column is basically the same Concatenation formula as the first pass column. You just used different flags ($$ as opposed to Â§Â§Â§).
I compared the values from both the first pass and second pass columns by pasting them as text, and they are identical.
Since they are identical, there’s no point in running a second pass.
You can find the glossary after the first S&R pass here:
Please have a look and let me know whether I can start the first pass with all the other files.
Also please let me know what your plan for the 2nd pass is, because with the current formula being identical, I don’t think that will accomplish what you are trying to do.
You mean row 47-48? Because the file is just 2 columns.
Ignore it. That’s not spellchecking. There’ll be another step after spellchecking to take care of those.
” Yes, I meant row. Ok, I’ll ignore it. “I asked Sebastian for a list of terms whose translations were conflicting (e.g., Minions vs. Slave). Below is his reply, which is kinda off-topic, but still needs careful attention.
no, I don’t have a real list, but here’s what I have from the notes I took and what I remember (some of these are context dependent!). I also mentioned some of these below.
Einfluss was changed to K$$raftschlag; Engl. Impact. When the skill is meant, this is correct, but in other contexts it should remain translated as Einfluss (I guess this is a case for proofreading)
Einsatz was changed to A$$usfall. However, it should only be Ausfall for English Sortie (which is actually french – a type of attack).
Zauber was changed to G$$lÃ¼cksbringer, should only happen for English Charm, but NOT Charming, and also not for Spell and other German words that could be translated as Zauber.
Engl. “Charming” is “GlÃ¼cksbringer” in the glossary, but needs to be translated as “Charmant”.
Ware was changed to Warenauftrag. (Waren became Warenauftragn). Actually, most cases need to be “Ware”/”Waren”, Warenauftrag is only used when the type of quest is referred to (an Auftrag to get Waren), english “Consignment” (but some consignments are better of as Waren — a complicated case which really needs case by case checking.)
A$$tlantis-Teleporter -> A$$tlantis-Teleporterin produces Atlantis-Teleporterinin -> I don’t remember if I suggested a rule to shorten it back, because it occured when I first checked 121-150.
Maybe there are not that many cases and we should really review them manually case by case, as some depend on context.
Also, you should know Christina pointed out some conflicts that I didn’t find critical, eg. Gorge – Schlucht and Canyon – Schlucht. I said such can be ignored because we only have Kings Gorge and Undead Canyon – so KÃ¶nigs-Schlucht and Untoten-Schlucht is not a real conflict. The same is true for fighter, combatant, etc. distinctions: Seahorse Combatant is Seepferdchen-KÃ¤mpfer (Christina suggests: Seepferdchen-Soldat); Tribe Fighter is Stammes-KÃ¤mpfer — again, no two monsters with the same name when we translate both as KÃ¤mpfer.
Just mentioning this to make sure you know what strategy I followed and to discuss whether you agree — when there was danger that two things are really the same name, I did mention it in my comments.
Another thing I forgot: I left a comment REVIEW DESCRIPTION with some skills which I forgot to resolve. Here, how they are named depends on what they do and I remember some files that had their descriptions, so I wanted to go back to them and check. I can do that later and tell you, it’s only 5 terms or so. Sorry about that.
I’ll get back to you later about this, now I have an appointment and need to get going…
> > Another question: Some sentences are missing their period, but
> > end on programmer signs, like : “…gelangt#}”
> > Ok to insert the period after these signs (like: “…gelangt#}.”
> Both will work. The difference is that if you put it before the end
> the programming signs the period will also be
> highlighted/underline/different color while if it’s afterward it
> be the standard font/color.
This isn’t important enough for you to go back and change things. I prefer the period after the programming signs but eh, either way works.
the second pass would of course only make sense after the files have been proofed and Â§Â§Â§ has been removed.
So the logic is:
You stated that the script gets stumped because of compound terms being filled with one-word entries.
Anything the script doesn’t pick up on the first run could be replaced in the second pass. The second run is only if we have enough time and to be extra sure.
Attached you find the final S&R with Sebastian’s comments included. What I’ve seen so far of the glossary seems like the method is working pretty well.
I also took a look at Sebastians last comments and integrated/removed a few terms from the list according to his suggestions. We should have the most critical cases covered, but a few issues may still trickle through to the final product.
Okay, I’ll run the first pass through everything.
Rereading your post above, are you saying that the current 2nd pass list is irrelevant? Because it is. It’s identical to the 1st pass list. Example:
German wrong: Xyyzy. German correct: Xyzzy
1st pass looks for Xyyzy and replaces it with XÂ§Â§Â§yzzy.
2nd pass looks for either Xyyzy or XÂ§Â§Â§yzzy or Xyzzy (I can set it to any of these three, pick one) and replaces it with X$$yzzy.
But other than the flag characters (Â§Â§Â§ vs $$), the result is identical.
This means that any run of the 2nd pass will give the exact same result as not running it in the first place.
Anyway, the first pass will be run tonight on all 314 files.
Do you have time to do some proofreading on some of those files tonight?
PS: See new entries at https://spreadsheets.google.com/ccc?key=0AmlfEgfxF8r3dEFHVmcxVE5lYTBaWVZfYTA3WnpmWlE&authkey=CLm0jrUG&hl=en#gid=0
is “Void Biters” meant to be translated into German as “Leeren-BeiÃŸer” or “Leere-BeiÃŸer”…? Both versions appear in my document.” Leere-BeiÃŸer.
about the passes:
You’re right, since we are proofreading, we can expect proofers to take care of things that the S&R list misses and we don’t need the second pass.
The original thought was to add terms to the second pass after the first was run, but I didn’t fully think that through. A second pass would make sense if we ran the longer words first, proof them and then run the one-word words.
Anyway, perhaps a second pass will be necessary if we come across other issues.
Yeah, I can do some proofreading. What is your estimate regarding hours and starting/due time?
Not sure yet.
Current I’m discovering that I can’t simply stack the no-spotting required with the proofing required and I’m trying to resolve that.
Here’s an example:
Incorrect text: Aktions-Buch Spott
Correct text: Aktionsbuch Spotten
The list for no spotting required: Aktions-Buch -> Aktionsbuch
The list for spotting required: Aktions-Buch Spott -> A$$ktions-Buch Spotten.
If I run the no-spotting-required list first followed by spotting required, the following happens.
The script looks for Aktions-Buch and replaces it with Aktionsbuch. It then looks for Aktions-Buch Spott and does not find it, because the data is now Aktionsbuch Spott. We end up with Aktionsbuch Spott.
If I run the spotting-required list first followed by spotting required, the following happens.
The script looks for Aktions-Buch Spott and replaces it with A$$ktions-Buch Spotten. It then looks for Aktions-Buch. But the data is now A$$ktions-Buch Spotten. So we end up with A$$ktions-Buch Spotten.
In both cases we do not end up with the desired A$$ktionsbuch Spotten.
Currently trying to run the no-spotting-required list against the spotting-required list to see if that will fix the problem.
Any comments, ideas, etc. will be appreciated.
are we reverting Stufe back to Level? I thought the client explicitly prefers Stufe over Level.
I don’t mind, and in fact it’s consistent with the Monster LV70 and so on that is all over the place. I just want to make sure it’s intentional.