Re: [modules] New Beta Module: StatResGNT

Introducing an appropriate rendering for ‘name’ could be done very fast.

I need to sort out my source access again as on new computer but it is a simple thing.

Sent from Outlook for iOS

----------

From: sword-devel <[email protected]> on behalf of David Haslam <[email protected]>
Sent: Saturday, March 15, 2025 11:58 am
To: SWORD Developers' Collaboration Forum <[email protected]>
Subject: Re: [sword-devel] [modules] New Beta Module: StatResGNT

Thanks Arnaud,

The name element might indeed be appropriate, and it would be good to see this used more in OSIS Bibles, although to identify & cover all names would require a lot of effort - even for our flagship module KJV; (DM & I began some eXperimental work on that a few years ago).

Each of the nomina sacra would be a singular name, so if not abbreviated, the attribute value should be "x-nomen-sacra" - which might even be shortened as "x-nom-sac".

Best regards,

David

Sent with Proton Mail secure email.

On Saturday, March 15th, 2025 at 10:58 AM, Arnaud Vié <[email protected]> wrote:

Ah, so I was indeed missing something ! Thanks David for the clarification.

I still
think using <hi> is not a satisfying solution though, as it loses
semantic information that OSIS is supposed to preserve.

OSIS
does provide a more generic <name type="..."> tag that could be
used. The currently supported name types are "geographic", "holiday",
"nonhuman", "person", "ritual", but x-prefixed extensions are also
allowed.

So a <name type="x-nominasacra"> for example
(with proper handling in sword) would be more appropriate - though of
course the rendering for it would have to be implemented.

Le sam. 15 mars 2025 à 11:46, David Haslam <[email protected] > a écrit :

Arnaud,

The divineName is used only for the Tetragrammaton and its conjugates from the Hebrew OT. The SWORD engine renders such using small-caps.

Nomina Sacra are a quite different matter. Refer to the Wikipedia article.

Even the name Jerusalem has a Nomen Sacrum in some MSS.

David

Sent from
Proton Mail for iOS

On Sat, Mar 15, 2025 at 10:31, Arnaud Vié <
[email protected] < rel=>> wrote:

Maybe I'm missing something, but for the "nomina sacra" use case, why are you not using the dedicated semantic <divineName> tag ?

OSIS is designed to store semantic markup when possible, rather than formatting information.

The OSIS manual clearly specifies that "The hi element is reserved for cases where the reason for the highlighting or other typographic distinction is unknown." Here, you know what the reason is, and OSIS provides a dedicated tag for your use case. So using <hi> here is actually a misusage as far as OSIS is concerned.

The more correct course of action would be using <divineName>, and ensuring that the rendering code in SWORD is applying a proper formatting to this divineName tag.

(Or even better, allow bibles to carry "stylesheets" to customise the rendering of tags per-bible without sacrificing the semantic information - but that's a matter for another day).

Regards,

Arnaud

Le sam. 15 mars 2025 à 04:53, David Haslam <
[email protected] > a écrit :

I've now documented some of this in the developers' wiki, under a new section for
Text styles.

OSIS Bibles - CrossWire Bible Society

Best regards,

David

Sent with
Proton Mail secure email.

On Saturday, March 15th, 2025 at 12:48 AM, DM Smith <
[email protected] > wrote:

Your question is really whether it is buggy. it looks good to me,

Really, Troy has a module for which it works. There may be a future latex, html, xhtml, tei, …. that might find a problem and find a bug.

There is an incomplete test suite (all test suites are incomplete!) that could be improved with fuller input. Maybe this too.

So much of our software development is, “works for me” I’ll go on to the next thing. That’s what I do. If someone finds a problem, then I’ll fix it.

DM

On Mar 14, 2025, at 12:45 PM, David Haslam <
[email protected] > wrote:

Thanks Troy, 😀

Your replies were clearly intended to be read by coders! 🥸

Has anyone tested this bit of code yet?

Best regards,

David

Sent with
Proton Mail secure email.

On Friday, March 14th, 2025 at 4:38 PM, Troy A. Griffitts <
[email protected] > wrote:

I believe we support this in SWORD:

```

[tgriffitts@fedora filters]$ grep overline *
osishtmlhref.cpp: else if (type == "ol" || type == "overline" || type == "x-overline") {
osishtmlhref.cpp: outText("<span style=\"text-decoration:overline\">", buf, u);
osislatex.cpp: // there is no officially supported OSIS overline attribute,
osislatex.cpp: // thus either TEI overline or OSIS x-overline would be best,
osislatex.cpp: // OSIS overline attribute is made available, these should all
osislatex.cpp: else if (type == "ol" || type == "overline" || type == "x-overline") {
osislatex.cpp: outText("\\textoverline{", buf, u);
osisplain.cpp: // there is no officially supported OSIS overline attribute,
osisplain.cpp: // thus either TEI overline or OSIS x-overline would be best,
osisplain.cpp: // OSIS overline attribute is made available, these should all
osisplain.cpp: if (strstr(token, "rend=\"ol\"") || strstr(token, "rend=\"x-overline\"") || strstr(token, "rend=\"overline\"")
osisplain.cpp: || strstr(token, "type=\"ol\"") || strstr(token, "type=\"x-overline\"") || strstr(token, "type=\"overline\"")) {
osisplain.cpp: u->hiType = "overline";
osisplain.cpp: if (u->hiType == "overline") {
osisxhtml.cpp: .overline { text-decoration: overline; }\n\
osisxhtml.cpp: // there is no officially supported OSIS overline attribute,
osisxhtml.cpp: // thus either TEI overline or OSIS x-overline would be best,
osisxhtml.cpp: // OSIS overline attribute is made available, these should all
osisxhtml.cpp: else if (type == "ol" || type == "overline" || type == "x-overline") {
osisxhtml.cpp: outText("<span class=\"overline\">", buf, u);
teihtmlhref.cpp: else if (rend == "overline")
teihtmlhref.cpp: buf += "<span style=\"text-decoration:overline\">";
teihtmlhref.cpp: else if (rend == "overline")
teilatex.cpp: else if (rend == "overline")
teilatex.cpp: buf += "\\overline{";
teixhtml.cpp: else if (rend == "overline")
teixhtml.cpp: buf += "<span style=\"text-decoration:overline\">";
teixhtml.cpp: else if (rend == "overline")
```

On 3/14/25 3:31 PM, David Haslam wrote:

Thanks DM,

As and when we get a fix on this for SWORD as well as JSword, we should add a suitable description in
OSIS Bibles - CrossWire Bible Society

We might even use the
nomen-sacrum
case as the illustrated example.

Can anyone familiar with how SWORD works tell us whether

<hi
rend="overline"
>
θς
</hi>

would be correctly rendered by SWORD?

And would we need to specify this within an eXtended
type attribute value (one with an "x-" prefix) for the
abbr element?

<abbr type="x-nomSac">

<hi
rend="overline"
>
θς
</hi>

</abbr>

Best regards,

David

Sent with
Proton Mail secure email.

On Friday, March 14th, 2025 at 1:58 PM, DM Smith
<[email protected]> wrote:

Regarding JSword, it supports some TEI within OSIS. Specifically it supports hi with TEI’s attribute rend in addition to OSIS’s attribute type. Chris Little directed that this should be so.

It has been a long time since we discussed this, so my memory might be faulty that OSIS was based on TEI and SWORD/JSword would allow some TEI constructs with the OSIS filter.

JSword needs a small, easy change to support overline. At this time, it will show the text without any decoration. It should be added to our edition of OSIS.

On Mar 14, 2025, at 4:55 AM, David Haslam
<[email protected]> wrote:

Hi Troy,

Does SWORD support an OSIS equivalent to this TEI snippet?

<abbr type="nomSac">

<hi rend="overline">
θς
</hi>

</abbr>

Not asking about the
abbr element, but the
rend="overline" attribute for the
hi element.

Rendering text with
overline is not documented in the
OSIS User Manual.

- Has this ever been tested in SWORD front-ends?

- How about JSword-based apps?

Best regards,

David

Sent with
Proton Mail secure email.

On Friday, March 14th, 2025 at 8:48 AM, David Haslam
<[email protected]> wrote:

Follow-up...

The screenshot image in my
previous reply is what I saw in
Safari when I clicked Troy's link while viewing messages from my
iPad Mini 4.

Now that I've done likewise while viewing messages from my
Windows 11 PC, I see a proper XML file via the
Edge browser!

<w>

<abbr
type="nomSac"
>

<hi
rend="overline"
>
θς
</hi>

</abbr>

</w>

<w>
τον
</w>

<w>
κοσμον
</w>

<pc>
·
</pc>

<lb/>

<w>
ωστε
</w>

<w>
τον
</w>

Just shows that WYSINAWYG ! 🥸

Best regards,

David

Sent with
Proton Mail secure email.

On Friday, March 14th, 2025 at 6:40 AM, David Haslam
<[email protected]> wrote:

<106d2d31-021b-436b-80ed-0686abfbff3a_image.png>

Hi Troy,

Above is what I see after I clicked your link! Not sure if that’s what you intended.

David

On Thu, Mar 13, 2025 at 23:52, Troy A. Griffitts <
[email protected] < rel=>> wrote:

This is how we record nomen sacrum in TEI and should transfer to OSIS just fine, e.g.,

https://ntvmr.uni-muenster.de/community/vmr/api/transcript/get/?docID=20001&indexContent=John.3.16&format=teiraw

On 3/13/25 4:28 PM, Peter von Kaehne wrote:

I think the best and correct way is to call this ‘display’ and encode the exact look you want to achieve via style sheet.

Sent from
Outlook for iOS

----------

From: sword-devel <[email protected]> on behalf of David Haslam <[email protected]>
Sent: Thursday, March 13, 2025 2:34 pm
To: Alan Bunning <[email protected]>
Cc: sword-devel mailing list <[email protected]>
Subject: Re: [sword-devel] Fwd: [modules] New Beta Module: StatResGNT

It would also require that we devise a means to encode in the OSIS the equivalent to what was used in the underlying Wiki format !

{{overline|ΘΥ}}

I doubt if the SWORD engine has a means to implement that!

Best regards,

David

Sent with
Proton Mail secure email.

On Thursday, March 13th, 2025 at 2:22 PM, David Haslam
<[email protected]> wrote:

Thanks, Alan, for the explanation.

This then prompts a suggestion as to the potential use of the OSIS
abbr element for each instance. This would require that each type of
nomina sacrum would then need to be retained as the abbreviation, and then expanded using the XML attribute, e.g.

<abbr expansion="Θεοῦ">
ΘΥ
</abbr>

Please refer to the
OSIS User Manual section 13.2

NB. I don't know whether the SWORD engine has any code to display the expansions!

The coders might enlighten us on this point. I have therefore cc: to
sword-devel
in this reply.

Best regards,

David

Sent with
Proton Mail secure email.

On Thursday, March 13th, 2025 at 1:45 PM, Alan Bunning
<[email protected]> wrote:

Yes, that symbol is used to mark nomina sacra.

Alan Bunning, D.Litt.
Executive Director
Center for New Testament Restoration
https://greekcntr.org

On 3/13/2025 9:42 AM, Kovács Zoltán wrote:

Dear Alan,

I received the following emails. I haven't looked into them very much, but maybe you want to join the conversation.

Blessings, Zoltan

---------- Forwarded message ---------

Feladó:
David Haslam <
[email protected] >

Date: 2025. márc. 13., Cs, 14:37

Subject: Re: [modules] New Beta Module: StatResGNT

To: pierre amadio <
[email protected] >

Cc: Kovács Zoltán <
[email protected] >, <
[email protected] >

Hi Pierre,

That's a very plausible explanation!

Here are the relevant lines from my word count.

Word Count

˚Θεέ 2

˚Θεοῦ 690

˚Θεόν 73

˚Θεός 68

˚Θεὸν 72

˚Θεὸς 239

˚Θεῷ 158

˚Κυρίου 231

˚Κυρίῳ 95

˚Κύριε 89

˚Κύριον 63

˚Κύριος 139

˚Κύριόν 1

˚Κύριός 10

˚Πνεύματι 59

˚Πνεύματος 79

˚Πνεύματός 5

˚Πνεῦμά 7

˚Πνεῦμα 99

˚Χριστέ 1

˚Χριστοῦ 252

˚Χριστόν 18

˚Χριστός 41

˚Χριστὸν 48

˚Χριστὸς 71

˚Χριστῷ 104

˚θεὸς 1

˚Ἰησοῦ 329

˚Ἰησοῦν 126

˚Ἰησοῦς 456

NB. If this is a technical means to mark the use of
nomina sacra, then it ought to be described in the conf file!

Best regards,

David

Sent with
Proton Mail secure email.

On Thursday, March 13th, 2025 at 1:11 PM, pierre amadio
[email protected] wrote:

Hi there.

As far as I know, there is no such greek accent as a circular mark
such as Ring Above (U+02DA).
Looking at the source text, they seem to always appears next to words such as
https://github.com/Center-for-New-Testament-Restoration/SR/blob/main/SR.txt
˚Ἰησοῦ ˚Χριστοῦ,
˚Πνεύματος
˚Κυρίου

Could it be a way to mark the use of a nomina sacra in manuscripts ?
https://en.wikipedia.org/wiki/Nomina_sacra

On Thu, 13 Mar 2025 at 13:38, Kovács Zoltán [email protected] wrote:

Thanks, David, I will look into it and get back to you.
Best, Zoltan

David Haslam [email protected] ezt írta (időpont: 2025. márc. 13., Cs, 13:33):

Arising from the grapheme count, I have a further technical query.

There are 3626 instances of U+02DA RING ABOVE.

I am wondering whether these would be better being all replaced by U+030A COMBINING RING ABOVE, and if so, whether the mark should follow the Greek letter rather than precede it.

NB. My knowledge of Ancient Greek accents rules is fairly limited.

Q. What is the most suitable Unicode font for viewing a Greek NT text?
Might it be sensible to include such an element in the .conf file?

Best regards,

David

Sent with Proton Mail secure email.

On Thursday, March 13th, 2025 at 12:13 PM, David Haslam [email protected] wrote:

Hi Kovács,

Using BabelPad Tools, I have just done a grapheme count on the full Greek text with accents.
i.e. After first stripping out the references from the diatheke output. See inside attached 7-Zip file.

There are two left/right count mismatches that ought to be looked at.

2018 ‘ 483
2019 ’ 484
201C “ 1,904
201D ” 1,907

cf. The counts of left/right parentheses were OK.

0028 ( 62
0029 ) 62

Best regards,

David

Sent with Proton Mail secure email.

On Thursday, March 13th, 2025 at 11:59 AM, David Haslam [email protected] wrote:

This module has the wrong Versification.

It's a Greek NT and therefore LXX is simply the wrong choice!

Please fix this ASAP.

Dom, Please also fix any bug in your script that led to this mistake!

Best regards,

David

Sent with Proton Mail secure email.

On Wednesday, March 12th, 2025 at 5:52 PM, [email protected] [email protected] wrote:

This is to announce that we have just now uploaded StatResGNT
in the CrossWire beta repository for testing purposes.

If no raised concern nor a quality alert has been sent on the list,
StatResGNT will be published in a week.

This is an update.
Language=Ancient Greek (to 1453)
Version=1.3
History_1.3=(2025-03-12) Update to newest release (last database change: 2024-07-16)
TextSource= https://github.com/Center-for-New-Testament-Restoration/SR/blob/master/SR.txt
Versification=LXX

Many thanks to Kovács_Zoltán for the hard work.

yours

P.S.: This email is sent automatically.

_______________________________________________
modules mailing list
[email protected]
http://www.crosswire.org/mailman/listinfo/modules

_______________________________________________
sword-devel mailing list: [email protected]
http://crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

_______________________________________________

sword-devel mailing list:
[email protected]

http://crosswire.org/mailman/listinfo/sword-devel

Instructions to unsubscribe/change your settings at above page

_______________________________________________
sword-devel mailing list: [email protected]
http://crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

_______________________________________________

sword-devel mailing list:
[email protected]

http://crosswire.org/mailman/listinfo/sword-devel

Instructions to unsubscribe/change your settings at above page

_______________________________________________

sword-devel mailing list:
[email protected]

http://crosswire.org/mailman/listinfo/sword-devel

Instructions to unsubscribe/change your settings at above page

_______________________________________________

sword-devel mailing list: [email protected]

http://crosswire.org/mailman/listinfo/sword-devel

Instructions to unsubscribe/change your settings at above page

_______________________________________________
sword-devel mailing list: [email protected]
http://crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page