Confusion of Haggai and Revelation

[michalkitta]

In the Czech version of Xiphos the search for Zephaniah skips to Revelation instead of Haggai. Same when browsing from the top of Zechariah. The links in the search do not show Haggai. Also, Haggai is missing in Bibletime.

[mcepl]

Maintainer of Czech Bible modules here. Sorry, I don’t follow Xiphos (using AndBible most of the time), and even less I follow this issue tracker. Could you tell me, please, which Czech module you are using? I am maintaing four of them CzeCSP, CzeBKR, CzeCEP, and CzeB21. Could we continue this discussion there?

[mcepl]

Actually, after emailing with the reporter in private, I was told that the error is the same in all modules with Xiphos and not for example in AndBible.

However, I have to admit, I have not been able to reproduce it neither with Xiphos nor with BibleTime (both the latest packages on FlatHub).

[karlkleinpaste]

I don't understand the complaint. Is Zephaniah seen as missing in some circumstance?

I can navigate to Zephaniah off the book pulldown, or I can just type Zeph.2.3 in the location bar, and it navigates there just fine. And similarly I have no problem getting to Haggai.

What exactly is the problem? An exact repeat-by seems necessary.

[michalkitta]

Hallo!

I use debian and Xiphos. 4.2.1 (gtk3 webkit2) Haggai is missing. The link to Haggai refers to Revelation. I tried Czech, German, English, Polish, Ukrainian bibles. Haggai in WLC refers to Nehemiah. Haggai is missing in the LXX. After installing Xiphos from Flathub (gtk3 webkit2) everything is fine.

Thanks.

Michal Kitta

---------- Původní e-mail ---------- Od: karl kleinpaste @.> Komu: crosswire/xiphos @.> Kopie: michalkitta @.>, Author @.> Datum: 8. 12. 2023 17:28:25 Předmět: Re: [crosswire/xiphos] Confusion of Haggai and Revelation (Issue # 1137) "

I don't understand the complaint. Is Zephaniah seen as missing in some circumstance?

I can navigate to Zephaniah off the book pulldown, or I can just type Zeph. 2.3 in the location bar, and it navigates there just fine. And similarly I have no problem getting to Haggai.

What exactly is the problem? An exact repeat-by seems necessary.

— Reply to this email directly, view it on GitHub (https://github.com/crosswire/xiphos/issues/1137#issuecomment-1847482464), or unsubscribe (https://github.com/notifications/unsubscribe-auth/BERDAML63TMBHHJ5TRHV54DYIM52JAVCNFSM6AAAAABALXAXHSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNBXGQ4DENBWGQ) . You are receiving this because you authored the thread. Message ID: < @.***> "

[karlkleinpaste]

OK, let's get precise here, because what's seen so far is in such generalities that I cannot even guess what you mean.

"The link to Haggai". What link? Where exactly are you seeing a link? As I said previously, an exact repeat-by is in order.

Everywhere that I see Haggai in Xiphos, it opens Haggai just fine. No reference to Revelation.

Haggai is present in module LXX:

I don't have anything at all to say about flatpak or appimage builds of Xiphos. We've seen those before and they've generally been nightmarish in having subtly different behaviors that just caused us grief to track down. I build and use Xiphos as built directly out of the github tree.

[karlkleinpaste]

I found myself wondering whether something is broken in the Czech locale. So I used Preferences to set locale to "Čeština (cs_CZ)" and restarted. I looked in /usr/share/sword/locales.d to find that Zephaniah in Czech is Sofoniáš, moved to it and screenshot it, complete with the pulldown. So far so good.

Now, when I went looking for Haggai, things got weird. I found myself with "Zjeveni 1:1" which is Revelation, but with content of Haggai 1:1 showing, again from a previous run and having exited with that as current verse.

What this tells me is that the Czech locale in the Sword library is broken somehow; Xiphos has nothing to say about this specifically, because it depends on the Sword library to do book name conversions appropriately. I'm surprised that Matěj has never noticed or reported this. In any event, this isn't a Xiphos error, it's a Sword library error, and will have to be fixed there. Please file a report at https://tracker.crosswire.org.

[LAfricain]

What this tells me is that the Czech locale in the Sword library is broken somehow; Xiphos has nothing to say about this specifically, because it depends on the Sword library to do book name conversions appropriately

You are right, it's a sword issue. Probably a translation issue. Matej, can you have a look to the locale file?

[michalkitta]

Yes, you can't access the text of Haggai with the Czech interface. I am attaching a gif file.

Michal Kitta

---------- Původní e-mail ---------- Od: karl kleinpaste @.> Komu: crosswire/xiphos @.> Kopie: michalkitta @.>, Author @.> Datum: 8. 12. 2023 21:12:06 Předmět: Re: [crosswire/xiphos] Confusion of Haggai and Revelation (Issue # 1137) "

I found myself wondering whether something is broken in the Czech locale. So I used Preferences to set locale to "Čeština (cs_CZ)" and restarted. I looked in /usr/share/sword/locales.d to find that Zephaniah in Czech is Sofoniáš, moved to it and screenshot it, complete with the pulldown. So far so good.

sofonias.png (view on web) (https://github.com/crosswire/xiphos/assets/7548920/58e46f32-fd0e-41d7-8baf-3fb80fe79e48)

Now, when I went looking for Haggai, things got weird. I found myself with "Zjeveni 1:1" which is Revelation, but with content of Haggai 1:1 showing, again from a previous run and having exited with that as current verse.

What this tells me is that the Czech locale in the Sword library is broken somehow; Xiphos has nothing to say about this specifically, because it depends on the Sword library to do book name conversions appropriately. I'm surprised that Matěj has never noticed or reported this. In any event, this isn't a Xiphos error, it's a Sword library error, and will have to be fixed there. Please file a report at https://tracker.crosswire.org (https://tracker.crosswire.org).

— Reply to this email directly, view it on GitHub (https://github.com/crosswire/xiphos/issues/1137#issuecomment-1847786030), or unsubscribe (https://github.com/notifications/unsubscribe-auth/BERDAMLVONNLG5U2LXKFNSDYINYBFAVCNFSM6AAAAABALXAXHSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNBXG44DMMBTGA) . You are receiving this because you authored the thread. Message ID: < @.***> "

[mcepl]

I don't have anything at all to say about flatpak or appimage builds of Xiphos. We've seen those before and they've generally been nightmarish in having subtly different behaviors that just caused us grief to track down. I build and use Xiphos as built directly out of the github tree.

This is completely en passant this ticket, but I have switched to a Flatpak-oriented distro recently (MicroOS-based container oriented one) and I honestly believe that Flatpaks are the future. I know that you don’t provide Linux binaries (and it is probably a right thing), but if you ever did, then I think Flatpak is the best way to do it.

[mcepl]

Hmm, I still cannot reproduce: search for “Aggeus 1:1” (there are two g in Aggeus in Czech; display of Revelation 1:1 usually happens when the verse is not found) leads to the correct display of the beginning of the book of Haggai:

[karlkleinpaste]

there are two g in Aggeus in Czech

Aaaaaaaand there's the problem in the Sword library.

$ grep Ag /usr/share/sword/locales.d/cs-utf8.conf 
Haggai=Ageus

No wonder Sword can't translate book names properly. The badly translated names will never match a versekey in a Czech module. A simple typo.

I had no reason to know that the Sword library Czech locale had a plain misspelling. That's what will need to be fixed.

[mcepl]

Actually, no there are not. There are no two g in Czech, but because I am maintainer of all Czech modules it is consistently wrong in all Czech modules. I'll fix it.

[mcepl]

Filed as https://tracker.crosswire.org/projects/API/issues/API-256