[sword-devel] LXX Strongs & Morphology

Thu Feb 7 14:39:49 MST 2019

Snippets of knowledge about how JSword does things differently than SWORD ought really to be documented in the developers’ wiki.

Is there a volunteer?

David

Sent from ProtonMail Mobile

On Thu, Feb 7, 2019 at 21:27, DM Smith <dmsmith at crosswire.org> wrote:

> JSword allows the setting of a Greek Mophology module. It doesn’t use the work given in the morph or lemma attributes. Forked XST can provide a different mechanism.
>
> — DM Smith
> From my phone. Brief. Weird autocorrections.
>
>> On Feb 7, 2019, at 3:26 PM, pierre amadio <amadio.pierre at gmail.com> wrote:
>>
>> Hi there !
>>
>> I find it strange that the LXX module tells it is using packard's
>> morphological code.
>>
>> Earlier in 2018, I started to use the LXX module (Version 2.5) and
>> installed the Packard module as well.
>> Then, i realised that the tool I was using gave me a lot of wrong information.
>>
>> The error were visible with diatheke as well, so it was not a bug in
>> the frontend i was using:
>>
>> $ diatheke -b LXX -o avm -f PLAIN -k Gen 1:1
>> Genesis 1:1: εν (PREP) αρχη (N-DSF) εποιησεν (V-AAI-3S) ο (T-NSM) θεος
>> (N-NSM) τον (T-ASM) ουρανον (N-ASM) και (CONJ) την (T-ASF) γην (N-ASF)
>> (LXX)
>>
>> θεος (N-NSM) is obviously nominative, however asking for the packard
>> key gives vocative as a result.
>>
>> $ diatheke -b Packard -f PLAIN -k N-NSM
>> : Part of speech: noun, indeclinable proper noun Case: vocative
>> Number: singular Gender: neuter
>>
>> Looking at εποιησεν (V-AAI-3S), it tells me it is a passive pluperfect 3s...
>>
>> Now, for the same codes, the robinson module gives result thats looks correct:
>>
>> $ diatheke -b Robinson -f PLAIN -k N-NSM
>> : * Part of Speech *: Noun
>> * Case *: Nominative
>> * Number *: Singular
>> * Gender *: Masculine
>> (Robinson)
>>
>> $ diatheke -b Robinson -f PLAIN -k V-AAI-3S
>> : * Part of Speech *: Verb
>> * Tense *: Aorist
>> * Voice *: Active
>> * Mood *: Indicative
>> * Person *: third
>> * Number *: Singular
>> (Robinson)
>>
>> I do not know if this is related to your problem, nor why LXX is
>> mentionning packard in the morph code tag, but it could be interesting
>> to see what happen if you try to use the Robinson module instead of
>> the Packard one.
>>
>>> On Thu, 7 Feb 2019 at 20:41, Dudeck, John <John.Dudeck at sim.org> wrote:
>>>
>>> Further info:
>>>
>>> I dumped the LXX and TR modules using mod2imp.
>>>
>>> 1) I see that the LXX morphology uses the module packard. So I downloaded packard from Crosswire.
>>>
>>> 2) I see that those words that do not have a Strongs number have the morphology duplicated in the Strongs field (!!!) So in Malachi 1:1 you can see for the first word it shows:
>>>
>>> <w lemma="strong:GN-NSN" morph="packard:N-NSN">λημμα</w>
>>>
>>> which is wrong!
>>>
>>> $$$Malachi 1:1
>>> <w lemma="strong:GN-NSN" morph="packard:N-NSN">λημμα</w> <w lemma="strong:G3056" morph="packard:N-GSM">λογου</w> <w lemma="strong:G2962" morph="packard:N-GSM">κυριου</w> <w lemma="strong:G1909" morph="packard:PREP">επι</w> <w lemma="strong:G3588" morph="packard:T-ASM">τον</w> <w lemma="strong:G2474" morph="packard:N-PRI">ισραηλ</w> <w lemma="strong:G1722" morph="packard:PREP">εν</w> <w lemma="strong:G5495" morph="packard:N-DSF">χειρι</w> <w lemma="strong:G32" morph="packard:N-GSM">αγγελου</w> <w lemma="strong:G846" morph="packard:D-GSM">αυτου</w> <w lemma="strong:G5087" morph="packard:V-AMM-2P">θεσθε</w> <w lemma="strong:G1211" morph="packard:PRT">δη</w> <w lemma="strong:G1909" morph="packard:PREP">επι</w> <w lemma="strong:G3588" morph="packard:T-APF">τας</w> <w lemma="strong:G2588" morph="packard:N-APF">καρδιας</w> <w lemma="strong:G5216" morph="packard:P-GP">υμων</w>
>>>
>>> A quick regular expression shows that there are 30512 words where the strongs number has morphology in the field out of 475348 total words.
>>>
>>> 3) The morphology field is not being displayed by AndBible. Only Strongs is being displayed. I see no differences in the tagging syntax used.
>>>
>>>
>>> The mods.d/lxx.conf and tr.conf files both contain:
>>>
>>> GlobalOptionFilter=OSISStrongs
>>> GlobalOptionFilter=OSISMorph
>>> Feature=StrongsNumbers
>>>
>>> So I don't understand why one displays the Morphology codes and the other does not.
>>>
>>> Any ideas?
>>>
>>> John
>>>
>>>> The behavior in AndBible for the LXX is rather odd. It doesn't matter where you look in the text,
>>>> the symptoms appear the same everywhere.
>>>>
>>>> When you tap 'nnnn' to display Strongs numbers, it displays the Strongs numbers, BUT for those
>>>> words where there is no Strongs, it displays a Morphology code. As David mentions, in Mal 1:1, it
>>>> shows N-NSN for the first word, and Strongs numbers for the rest of the words in the verse.
>>>>
>>>> If you open the display menu (three dots menu), Strongs and Morphology are both checked. If you
>>>> uncheck Morphology, nothing changes in the display. If you check Morphology and uncheck
>>>> Strongs, all of the Strongs disappear, and no Morphology codes are displayed. If you then touch
>>>> 'nnnn' again, it displays Strongs, but in the display menu Strongs is unchecked and Morphology is
>>>> checked but grayed out.
>>>>
>>>> If you touch a Strongs number, it displays the entry as expected. If you touch a Morphology code
>>>> (eg. N-NSN as above), it looks up the word 'N-NSN' in Strongs Hebrew and displays the entry
>>>> 3476 Naasson. If you touch 'V-PAPRP', it displays entry 3578 xenia from Strong's Greek.
>>>>
>>>> By comparison with the TR Textus Receptus, when you tap 'nnnn' to display Strongs numbers, it
>>>> displays the Strongs number followed by the Morphology (with no space between). You can touch
>>>> the number and it brings up the Strongs entry from Strongs Greek. You can touch the Morphology
>>>> and it brings up the Robinson's codes. If you open the display menu and uncheck Morphology, it
>>>> shows only the Strongs numbers. However you cannot display Morphology if Strongs is not
>>>> displayed.
>>>>
>>>> I would assume that the behavior of the display menu in AndBible is an AndBible issue, and it's
>>>> not really a problem, just a minor logic bug. But the inability to display Morphology in LXX is an
>>>> issue.
>>>>
>>>> I imagine this could either be addressed in AndBible, or in the LXX source. Is Tuomas reading
>>>> this, or do I need to write him separately?
>>>>
>>>> Where do I find the LXX source? Maybe this is something that can be corrected fairly easily.
>>>>
>>>> John
>>>>
>>>>> LXX installed in PocketSword can display both or either.
>>>>>
>>>>> There are some tagging errors such as the one I spotted in the first word of Malachi 1:1
>>>>>
>>>>> I’ve not looked at more than one location before responding on whether the 2 features work.
>>>>>
>>>>> Could it be AndBible or did you perhaps sample a location with many more tagging issues?
>>>>>
>>>>> David
>>>>>
>>>>> Sent from ProtonMail Mobile
>>>>>
>>>>>
>>>>> On Wed, Feb 6, 2019 at 01:19, Dudeck, John <John.Dudeck at sim.org> wrote:
>>>>> Greetings.
>>>>>
>>>>> I've just tried loading the LXX in AndBible. Supposedly it has both Strongs numbers and
>>>>> Morphology tagging. But in AndBible it only displays Morphology when there is no Strongs
>>>>> number on a given word. And in that case it tries to look up the Morphology in Strongs. In
>>>>> Xiphos you can see where there is no Strongs number in those cases.
>>>>>
>>>>> By the way I am using our own French version of Strongs. Is the morphology lookup
>>>>> supposed to be in Strongs for LXX? With the Textus Receptus it needs Robinson's module.
>>>>>
>>>>> In contrast the Textus Receptus displays both Strongs and Morphology.
>>>>>
>>>>> Does this mean there is a tagging problem in the LXX? Or is it an AndBible/Jsword
>>>>> problem?
>>>
>>> John Dudeck
>>> Programmer at Editions Cle Lyon, France
>>> john.dudeck at sim.org john at editionscle.com
>>> --
>>> "Good things come to those who wait, / Not to those who hesitate." -- Petra
>>>
>>> _______________________________________________
>>> sword-devel mailing list: sword-devel at 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 at 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 at crosswire.org
> http://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.crosswire.org/pipermail/sword-devel/attachments/20190207/3ca99445/attachment-0001.html>