[sword-devel] Strong's search does not return result if Strong's number in text points to different parts of a verse
Hi all, The Strong's search does not return results if the Strong's numbers in the text point to a part of a verse. Example: G5179 This should actually yield 15 results (according to Blue Letter Bible). One example result is John 20:25. The source text in NASB contains this: the imprint Another example is Acts 7:44. The source text in NASB contains this: to the pattern In both examples the Strong's id contains a character (b) at the end and that's what the search is not able to handle. This is how I’m setting up the search flags in this case: // from swmodule.h api docs: // for use with entryAttrib search type to match whole entry to value flags |= SEARCHFLAG_MATCHWHOLEENTRY; searchTerm = "Word//Lemma./" + searchTerm; Is this a bug? Best regards, Tobias ___ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
Re: [sword-devel] Strong's search does not return result if Strong's number in text points to different parts of a verse
On 7/14/20 4:09 PM, Tobias Klein wrote: > > The Strong's search does not return results if the Strong's numbers in > the text point to a part of a verse. > No, it's because the Strong's numbers in NASB are keyed to their own lexicons, the NAS Heb/Grk lexicons, which have multiple entries at 5179, as 5179a and 5179b. It's not a reference to a verse segment. 5179a: τυπικῶς \tupikōs\ Etymology: from G5179b Definition: typically Usage: example(1). 5179b: τύπος \tupos\ Etymology: from G5180 Definition: the mark (of a blow), an impression, stamp (made by a die) Usage: example(3), examples(2), form(2), images(1), imprint(1), model(1), pattern(3), type(1). A literal search for 5179a or 5179b succeeds. Unfortunately there's no way for Joe Average to know that an a/b/c suffix is needed or useful. (There is one entry, 3103, with 3 [a/b/c] entries. There are 182 a/b pairs.) ___ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
Re: [sword-devel] Strong's search does not return result if Strong's number in text points to different parts of a verse
Thanks, Karl. Makes sense now. @Troy: Since we don't have the NAS Heb/Grk lexicons, but just standard Strong's keyed lexicons, would it make sense to "normalize" the way how Sword searches in this case? Would it be correct to simply return all verses that have either 5179a or 5179b? I guess that's what you would get with other translations anyway, right? Best regards, Tobias On 7/14/20 10:30 PM, Karl Kleinpaste wrote: On 7/14/20 4:09 PM, Tobias Klein wrote: The Strong's search does not return results if the Strong's numbers in the text point to a part of a verse. No, it's because the Strong's numbers in NASB are keyed to their own lexicons, the NAS Heb/Grk lexicons, which have multiple entries at 5179, as 5179a and 5179b. It's not a reference to a verse segment. 5179a: τυπικῶς \tupikōs\ Etymology: from G5179b Definition: typically Usage: example(1). 5179b: τύπος \tupos\ Etymology: from G5180 Definition: the mark (of a blow), an impression, stamp (made by a die) Usage: example(3), examples(2), form(2), images(1), imprint(1), model(1), pattern(3), type(1). A literal search for 5179a or 5179b succeeds. Unfortunately there's no way for Joe Average to know that an a/b/c suffix is needed or useful. (There is one entry, 3103, with 3 [a/b/c] entries. There are 182 a/b pairs.) ___ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page ___ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
Re: [sword-devel] Strong's search does not return result if Strong's number in text points to different parts of a verse
On 7/15/20 2:15 AM, Tobias Klein wrote: > Since we don't have the NAS Heb/Grk lexicons Actually, we (well, I) do...which is why I could explain your difficulty, after all. I just moved them from my private repo to the Xiphos repo. Feel free to refresh and install. Troy and I discussed making them available a little while ago, but we never got to a concrete point. At least for now, they're available there. Perhaps we'll decide to move them a different repo sometime, but he may want to regenerate them as, say, TEI rather than the ThML in which they're currently encoded. Normalization of NASB's references might still be a good idea, but Lockman may well have something to say about that. ___ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
Re: [sword-devel] Strong's search does not return result if Strong's number in text points to different parts of a verse
On 7/15/20 1:23 PM, Karl Kleinpaste wrote: On 7/15/20 2:15 AM, Tobias Klein wrote: Since we don't have the NAS Heb/Grk lexicons Actually, we (well, I) do...which is why I could explain your difficulty, after all. I just moved them from my private repo to the Xiphos repo. Feel free to refresh and install. Troy and I discussed making them available a little while ago, but we never got to a concrete point. At least for now, they're available there. Perhaps we'll decide to move them a different repo sometime, but he may want to regenerate them as, say, TEI rather than the ThML in which they're currently encoded. Normalization of NASB's references might still be a good idea, but Lockman may well have something to say about that. Thank you! :) I didn't expect to get additional dictionary modules when starting my original post on this. I just installed them both and started looking around a bit. Normalization of the actual references in the dictionary module would be one thing, but the search is another thing I think. So for me the question is still whether a search for "G5179" could yield results for both "G5179a" and "G5179b" in case of the NASB. Independent of the specifics of a translation a user probably just thinks "Strong's" and doesn't even know about this feature of the NASB Strong's. Is there any other translation out there that uses these special Strong's keys? I'm asking all these questions because I'm currently working on a Strong's based dictionary view that combines the display for multiple Strong's based modules in one view. The "standard Strong's keys" are the common denominator here (even though some use a variant with a 0 prefix in front of the actual Strong's number). So in case of using the NASB it would still be nice for the user to have the same kind of search behavior as with any other of the Strong's enabled bible modules. Best regards, Tobias ___ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
Re: [sword-devel] Strong's search does not return result if Strong's number in text points to different parts of a verse
On 7/15/20 5:08 PM, Tobias Klein wrote: > Is there any other translation out there that uses these special > Strong's keys? I don't think any other uses the NAS lexicons. However, those are not the only "not-Strong's Strong's". For example, there is the Berean Bible (http://berean.bible) who have produced their own content of this sort. I am (over)due to re-work my existing BIBdraft, BSBdraft, and lexicon modules with the full, finalized content. This sort of thing is going to induce me to make per-module Strong's dictionaries possible in Xiphos. ___ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
Re: [sword-devel] Strong's search does not return result if Strong's number in text points to different parts of a verse
One or more of the modules hosted by Tyndale House for STEP Bible makes use of the TH augmented Strong’s numbers. Some of these modules can be manually installed in other front-end apps. AFAIK, TH has still not set up a proper repository that the SWORD module manager could access, were the connection details available. Correct me if I’m wrong. David Sent from ProtonMail Mobile On Thu, Jul 16, 2020 at 03:02, Karl Kleinpaste wrote: > On 7/15/20 5:08 PM, Tobias Klein wrote: > >> Is there any other translation out there that uses these special Strong's >> keys? > > I don't think any other uses the NAS lexicons. However, those are not the > only "not-Strong's Strong's". For example, there is the Berean Bible > (http://berean.bible) who have produced their own content of this sort. I am > (over)due to re-work my existing BIBdraft, BSBdraft, and lexicon modules with > the full, finalized content. This sort of thing is going to induce me to make > per-module Strong's dictionaries possible in Xiphos.___ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
Re: [sword-devel] Strong's search does not return result if Strong's number in text points to different parts of a verse
On 7/16/20 4:02 AM, Karl Kleinpaste wrote: On 7/15/20 5:08 PM, Tobias Klein wrote: Is there any other translation out there that uses these special Strong's keys? I don't think any other uses the NAS lexicons. However, those are not the only "not-Strong's Strong's". For example, there is the Berean Bible (http://berean.bible) who have produced their own content of this sort. I am (over)due to re-work my existing BIBdraft, BSBdraft, and lexicon modules with the full, finalized content. This sort of thing is going to induce me to make per-module Strong's dictionaries possible in Xiphos. Thanks once more! I've found a way to handle these special Strong's while still being able to use them for dictionaries with the standard keys. This is how: 1) When looking up an entry in a dictionary, first try to use the "raw key" (like G5179b) 2) If the dictionary does not have that key, try with a "normalized key" (in this case G5179) This way the special Strong's from NASB for example can still be used to lookup "normalized entries" from other Strong's keyed dictionaries. This is how my dictionary info box looks like now: https://pasteboard.co/JibMDcW.png Best regards, Tobias ___ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
Re: [sword-devel] Strong's search does not return result if Strong's number in text points to different parts of a verse
That looks nice, Tobias. Thanks to everyone who commented on this. I think we do similar normalization in the engine for the lookup, but not in the search. You'd know if the NASB on the the live swordweb site allows a word click at one of these locations with a letter suffix and it still produces a result. I'm on my phone right now and can't easily try. Great work everyone! On July 18, 2020 10:00:46 AM GMT+02:00, Tobias Klein wrote: >On 7/16/20 4:02 AM, Karl Kleinpaste wrote: >> On 7/15/20 5:08 PM, Tobias Klein wrote: >>> Is there any other translation out there that uses these special >>> Strong's keys? >> >> I don't think any other uses the NAS lexicons. However, those are not > >> the only "not-Strong's Strong's". For example, there is the Berean >> Bible (http://berean.bible) who have produced their own content of >> this sort. I am (over)due to re-work my existing BIBdraft, BSBdraft, > >> and lexicon modules with the full, finalized content. This sort of >> thing is going to induce me to make per-module Strong's dictionaries >> possible in Xiphos. > > >Thanks once more! I've found a way to handle these special Strong's >while still being able to use them for dictionaries with the standard >keys. This is how: > >1) When looking up an entry in a dictionary, first try to use the "raw >key" (like G5179b) > >2) If the dictionary does not have that key, try with a "normalized >key" >(in this case G5179) > >This way the special Strong's from NASB for example can still be used >to >lookup "normalized entries" from other Strong's keyed dictionaries. > >This is how my dictionary info box looks like now: >https://pasteboard.co/JibMDcW.png > >Best regards, >Tobias -- Sent from my Android device with K-9 Mail. Please excuse my brevity.___ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page
