Importing Hebrew Bible text

[TYA]

Shalom all,

 

I'm sure this has been asked before, but I couldn't find any existing thread for it (at least not for this specific question).  How do you import a Hebrew Bible text into Accordance?  I successfully imported an English Bible, but how do you import a Hebrew one?

 

Specifically, I have a text file that I created for BibleWorks (I'm a Windows user), which is in ANSI formatting, with the Hebrew (technically Aramaic language, but in Hebrew letters), reading as follows for Gen 1:1.

 

Gen 1:1 BR$YT BR) )LH) YT $MY) WYT )R()

 

How do I import this to read in Hebrew letters in Accordance, and will it then be searchable along with other Hebrew texts, like the BHS?  That is mainly what I'm going for, and it is very important to me, since Syriac texts aren't cross-searchable in Accordance with Hebrew texts.

 

Thanks

[Λύχνις Δαν]

This is likely not the answer you want but can you record the input file to Unicode ?

 

Once imported it will be searchable like an untagged text. But I am not sure that Research queries can currently include User Bibles or not. A quick experiment with my Sinaiticus says not. I'm pretty sure it's been requested somewhere though.

 

Thx

D

[TYA]

D,

 

How would I record the input file to unicode?

 

And though I haven't yet imported any Hebrew texts, I have imported English texts, and they do pull up in the "Research" -> "All" when I search for the word "God."  That's a good start.  Hopefully, the Hebrew will be found as well--at the very least, if I do an "inflected" search with the quotation marks, or something like that.

Edited September 9, 2018 by TYA

[Λύχνις Δαν]

It seems it works with English but I wasn't able to get a hit in Greek or Hebrew.

 

As to recoding (sorry I meant recode it rather than record it - typo on my part) I would probably do it in Python myself. The problem is that it is using completely different chars to represent the Hebrew characters. There has to be a mapping of what they mean somewhere. I've seen this sort of coding elsewhere too. The problem isn't whether it's ANSI or ASCII as the chars are not in the extended range - these are incidentally the same characters in UTF-8 encoding of Unicode too. You really need them to represent the Hebrew characters directly as Unicode codepoints. So you cannot just use an editor and load the file and save as UTF-8. I have never used BW but I guess they must have known how to understand this sort of file directly. Maybe they defined the format you are using ? If the format is more generally available then perhaps someone has a converter around.

 

Thx

D

[A.D. Riddle]

Can you use find/replace-all in a text editor? I just tried it with Gen 1:1 and it was pretty easy. Getting the final forms right will take a little fiddling, but since the text is unpointed, it should not take too long. 

 

A.D.

[TYA]

You really need them to represent the Hebrew characters directly as Unicode codepoints. So you cannot just use an editor and load the file and save as UTF-8. I have never used BW but I guess they must have known how to understand this sort of file directly. Maybe they defined the format you are using ? If the format is more generally available then perhaps someone has a converter around.

 

BW just gave a simple ASCI conversion chart, showing how to type the Hebrew letters in English.  22 letters in Hebrew.  Pretty simple.  Hence, Gen 1:1 reads BR$YT BR) )LH) YT $MY) WYT )R().  I then imported the text file (.txt) into BW, and in the compiler apparatus, it asks what language this is supposed to be.  I said "Hebrew," and it caused it to appear properly in Hebrew.  Fully searchable, along with all other Hebrew / Aramaic texts.  Easy.

 

Again, I'm a Windows user, and I don't know almost anything about coding.  If you, or someone knows how to do this, I would greatly appreciate that knowledge.  Thanks.

[Λύχνις Δαν]

Can you use find/replace-all in a text editor? I just tried it with Gen 1:1 and it was pretty easy. Getting the final forms right will take a little fiddling, but since the text is unpointed, it should not take too long. 

 

A.D.

 

Yep for a one-off this makes a lot of sense. Just have to watch not to convert letters in the book names. Is it just the one you have to do ?

 

Thx

D

[Fabian]

With every new feature in Accordance – OakTree fear the amount of additional work to support user.

 

This  means BW has features which I believe would not (never) come in Accordance. One is the import of Texts in beta code. So even if you import your text as unicode, it won't be the same as the texts from Accordance. Except Accordance is in 10 years in full unicode. Because I guess Accordance use also still beta code for their own Greek and Hebrew texts. One reason is the speed. So BW and Accordance are faster as L. Hopefully one time the computers are so fast this would be an issue anymore. But then Logos is also fast as the other two.

 

The second is fine tuning for verses. I don't know why, because the system now has the same difficulties to manage, but with less superior results. Only the KJV verse mapping is working fine, if the text you import is also in the KJV verse mapping.

 

 

BW just gave a simple ASCI conversion chart, showing how to type the Hebrew letters in English.  22 letters in Hebrew.  Pretty simple.  Hence, Gen 1:1 reads BR$YT BR) )LH) YT $MY) WYT )R().  I then imported the text file (.txt) into BW, and in the compiler apparatus, it asks what language this is supposed to be.  I said "Hebrew," and it caused it to appear properly in Hebrew.  Fully searchable, along with all other Hebrew / Aramaic texts.  Easy.

 

Again, I'm a Windows user, and I don't know almost anything about coding.  If you, or someone knows how to do this, I would greatly appreciate that knowledge.  Thanks.

 

This would work if it's a Greek text, but it doesn't work for the RTL texts.

 

The best what you can do is: 

 

Export it from BW, not by ->Tools, Importing/Exporting Information, Export Database.

 

Just by selecting the text in the 2 field. Sorry I don't know how it is called. I use 

 

Be sure you export it as unicode https://www.youtube.com/watch?v=pbZRB8yzLjk

 

Paste it in Notepad++

 

Then delete all information you don't need. Like the reference on the end and all file informations, like WTT.  

 

‎^WTTGenesis 1:1 בְּרֵאשִׁ֖ית בָּרָ֣א אֱלֹהִ֑ים אֵ֥ת הַשָּׁמַ֖יִם וְאֵ֥ת הָאָֽרֶץ׃

 ² וְהָאָ֗רֶץ הָיְתָ֥ה תֹ֙הוּ֙ וָבֹ֔הוּ וְחֹ֖שֶׁךְ עַל־פְּנֵ֣י תְה֑וֹם וְר֣וּחַ אֱלֹהִ֔ים מְרַחֶ֖פֶת עַל־פְּנֵ֥י הַמָּֽיִם׃

 ³

 

 

 (Gen. 1:1-3 WTT)

 

Also all empty lines. I think it is search as \r\r and replace $1 or \r. 

 

 

 

Then you ask best Michael G. Distefano (also here in the forums) how to prepare the text. Because I have sent him the text which is downloadable from http://www.accordancefiles1.com/exchange/bibles.htm, but he has changed something which makes the text better than mine.

 

It is allowed to support me under:

https://www.accordancebible.com/forums/topic/21829-recognizing-of-verses-and-chapter-0/?hl=import

https://www.accordancebible.com/forums/topic/13348-user-bibles-import-options/?hl=import

 

 

 

How do I import this to read in Hebrew letters in Accordance, 

I answered above

and will it then be searchable along with other Hebrew texts, like the BHS?  That is mainly what I'm going for, and it is very important to me, since Syriac texts aren't cross-searchable in Accordance with Hebrew texts.

Rather No, because Accordance is not yet able to compare and search for unicode and beta code as the same time.

 

 

Greetings

 

Fabian

Edited September 10, 2018 by Fabian

[TYA]

Thank you all for the advice and help here.

 

I have now successfully uploaded the Hebrew version I wanted.  However, to my great disappointment, as was already said above, it is not cross-searchable with other Hebrew-letter versions in Accordance (hence my other feature request for the Peshitta Tanakh in a cross-searchable format).

 

I wish there was an ability to import any text that we wanted, and have it instantly cross-searchable.  The latter is, in my opinion, where the greatest power lies in this wonderful program.

 

Thanks again

[Fabian]

Perhaps you want to add the Bible from Michael G. Distefano, then you would be able to compare the texts (yours and the WLC) with the compare function, because both are in unicode then.

 

Greetings

 

Fabian 

Edited September 11, 2018 by Fabian

[TYA]

Thanks Fabian,

 

I've already thought about building a "Unicode library" of texts, but it just doesn't make sense.  First of all, I've already tested searching with unicode extensively in Accordance.  In fact, I imported two different Hebrew versions already, and found that they do cross-search with each other.  But the problem is, wildcards ( * ? ) aren't able to be used in the search like they are in the regular Hebrew texts.

 

For instance, with regular Hebrew resources, I can use an "Inflected Search" using wildcards, and find all instances of Hebrew word "amar" (to say, speak), including with prefixes (bet, lamed, mem, etc.).  But with my unicode experiment, those wildcards didn't work, and so if I wanted to find all instances of "amar" with prefixes, I would manually have to search *exactly* each spelling, making 5X as much work.  No way.

 

It was a good idea, but simply doesn't work.  And I could mention other problems with it.  Bottom line, easiest solution is for Accordance to make the resources cross-searchable themselves.  Or develop a way to make unicode cross-searchable with beta, or whatever the other thing was.  Thanks.

[Helen Brown]

The wildcard search with Unicode is a bug that has been reported and fixed for the imminent update.

[TYA]

Thanks Helen.  That is good news, and I appreciate you sharing it.  However, when I said "other problems" with searching unicode above, here are some of things I meant, which makes it virtually impossible.

 

#1) When typing in the search bar to do the unicode search, the keyboard was set to English, and I couldn't get the keyboard to Hebrew.  (Of course, I'm only interested in unicode for the sake of Hebrew cross-searching, so if there is no way to type in Hebrew, what is the point?)

 

The searches I performed during my experiment were all based on copying and pasting text into the search bar from the Hebrew text I had imported.  But that is totally unviable, as I shouldn't have to see the word on the screen in order to be able to search it.  I should be able to type it into the search bar, in Hebrew, of course.

 

#2) Once I copied and pasted Hebrew into the search bar (again, set to "unicode" search), when I tried to add wildcards, they were always placed at the opposite end of the word from where the cursor was.  (This is a familiar problem I've dealt with for years with word processors when it comes to Hebrew text.  Others who work with Hebrew will have experienced this too).

 

Bottom line, even if wildcards *did* work for unicode search, getting them in the right place is a chore.  You have to position the cursor at some random location, and whatever you type appears in the wrong place.  (I could upload screenshots if anyone really cared to see, but I trust the basic concept will be understood).

 

This leads to a technically separate question: how can I change the language of the keyboard when typing in the command / search line?  Can I do this whenever I want?  I've already become frustrated with the bar showing Hebrew letters when what I really want (not related to discussion above) is to type and search English words.

 

Helen, any idea when that imminent bug fix will be rolled out?  Thank you.

Edited September 11, 2018 by TYA

[Joel Brown]

TYA,  entering unicode characters is a system specific choice.  If you don't want to type english characters, you must use a different unicode keyboard.  I believe these instructions work on adding that keyboard to Windows:

 

https://support.microsoft.com/en-us/help/4027670/windows-10-add-and-switch-input-and-display-language-preferences

[TYA]

Joel: so even though I'm typing into Accordance' program, I would have to specify a different keyboard from Windows?  That seems odd to me (though if it works, it's at least a workaround).  But I'm trying not to add unnecessary steps to everything I do.

 

If, however, you are suggesting that installing a unicode keyboard in Windows would allow me to manually type Hebrew letters into Accordance' search bar for the sake of doing unicode searches, **and** Accordance rolls out the bug fix soon which is keeping wildcards like (* ?) from working (again, very important to me), then I'm potentially interested.

 

Can you confirm if I have the correct understanding here, or do you need more information on what I'm trying to accomplish?  Thank you much!

[Λύχνις Δαν]

That's what I do. I have a number of keyboard mappings enabled for use on Windows and switch between them to enter text in the appropriate language. There was a post recently with some links on how to set this up. Let me see if I can find it.

 

EDIT: Here it is https://www.accordancebible.com/forums/topic/24374-using-accordance-font-in-windows10-msword-2016/?hl=%2Bwindows+%2Bkeyboard&do=findComment&comment=119696

 

Thx

D

Edited September 11, 2018 by דָנִיאֶל

[Joel Brown]

For an actual Hebrew or Greek text, Accordance knows and will convert input on the fly. However, a Unicode text could have anything, mixing Hebrew, Greek, Ethiopic, Korean, etc. So, Accordance doesn’t try to assume what you mean, you need to type the characters you desire directly.

[TYA]

Thanks D. for the helpful links.  That seems like the direction to go in for installing a unicode keyboard.

 

Not to be bothersome, but Joel or Helen, could you give a rough estimate on when the bug fix would be released, allowing unicode wildcards to be used for search just like regular Hebrew/Greek?  (Or, could you give an estimate on when the Peshitta OT could be made cross-searchable with Hebrew texts?)  

 

Thank you.

[Fabian]

Thanks D. for the helpful links.  That seems like the direction to go in for installing a unicode keyboard.

 

Not to be bothersome, but Joel or Helen, could you give a rough estimate on when the bug fix would be released, allowing unicode wildcards to be used for search just like regular Hebrew/Greek?  (Or, could you give an estimate on when the Peshitta OT could be made cross-searchable with Hebrew texts?)  

 

Thank you.

 

12.2.9 is out

 

From the Release Notes:

• Now the search character wild card (?) works properly with Unicode text.

 

Greetings

 

Fabian

[TYA]

Thanks Fabian.  I see the 12.2.9 update available.  Since I'm new, I just want to know if updating to this from 12.2.8 will wipe out any settings, imported modules, or anything like that, since I've spent a ton of time trying to customize the program.  Do you happen to know?

[Fabian]

I had never problems by updating. But I'm on Mac and iPhone.

On Windows I don't know.

 

Greetings

 

Fabian

[Helen Brown]

Updating the program will not wipe out any of your settings or modules on any platform. Only beta testers need to be careful on the mobile platforms. I find on Android that if I switch between the beta and the public release I have to reinstall everything.

[TYA]

Thanks!

[TYA]

The wildcard search with Unicode is a bug that has been reported and fixed for the imminent update.

 

Thanks so much for fixing this in 12.2.9.  I just updated, and the unicode search works perfectly with my experimental Hebrew text, whereas wildcards were not available before.

 

Helen, it was so kind of you to step in and inform me (us all) that this fix would be coming soon.  Otherwise, I would have had no idea.  Thanks so much.  What amazing support here in Accordance!

[Fabian]

TYA

 

Import it again with 12.3!

 

 

• Import of a UTF8 or UTF16 file which mainly uses a standard Accordance font, such as Greek, Hebrew, Syriac, etc., is now converted to be a regular Text module with Accordance fonts instead of Unicode fonts. (This gives better compatibility between other Accordance texts that use the same font.)

!Maybe you want first delete your module from the library and then do not delete it. So you have the old one in unicode too! You can then after the import with 12.3 double click on the unicode one and have both in your library.

 

 

Greetings

 

Fabian

Edited November 1, 2018 by Fabian