How to encode those notes in the edition? Different options came to my mind:

1. <p xml:id="HVVU_05_010">
   naladasya gandhamāṃsyā
   <lb n="2"/>vidalanam udbhedaḥ | ‘atibalaphalavati’ iti ni<metamark function="reference" place="above">1</metamark><note place="footer" type="exegesis">1. ādarśapustake 'pi sadoṣapāṭho nopalabdhaḥ.</note>rdoṣa eṣa pāṭhaḥ | atibalāni mahāprabhā-
   <lb n="3"/>vāṇi phalāni santi yasmin | amarabhuvi svarge vibhur indraḥ || 10 ||
   </p>

2. Or should the <note place="footer" type="exegesis">1. ādarśapustake 'pi sadoṣapāṭho nopalabdhaḥ.</note> rather be placed at the end of the commentary in the xml file as well?

3. Or would it make sense to create special xml:ids for these notes like it is done for the verses (e.g. <p xml:id="HVVU_05_010p"><note place="footer" type="exegesis">1. ādarśapustake 'pi sadoṣapāṭho nopalabdhaḥ.</note></p>?

In the final sentence of canto 28 of the Jaisalmer MS the scribe used an avagraha. It looks different to the avagraha-like symbols he uses to signify a defect of the paper.
This should make us think about a proper encoding of the symbols once more. ATM the defect symbol has been encoded by me as <metamark>ऽ</metamark> and the avagraha for now as the ऽ itself (although I'm not sure that would work).
Any ideas?

Maxi

Hi,
First of all I'm not quite sure if there is something above the "ma" which should be taken into consideration and encoded or if it can be left out completely.
Secondly I'm not sure about the "=" above "kra". I tried the encoding as depicted in the middle but commented it out for now because of my doubts.
Looking forward to your ideas,
Maxi

:%s/\(<l\_.\{-}||\s*\(\d\+\)\s*||<\/l>\)/<lg type="verse" xml:id="HV_XX_\2">\r\1\r<\/lg>/gc did the trick. Ideally one would like to have \2 zero-padded, and the whole thing being aware in which canto (XX) it is, but that can all also be dealt with later.

After line 2609 some more text probably related to the fifth canto needs to be transcribed, before the sixth canto begins.

Make it a separate TEI-File. I should have a look at that. And I should think about how to re-transfer my corrections to raw sanskrit.

Something like perl -pi.bak -e 's/kvaciditīva/kvacid itīva/g' hv-all-k.xml might save time in some cases.

In verses 17:40 to 17:41 I just came across an odd repetition (Folio 32B L3 M/R).

First, we have the last line of verse 40 written as "bhṛṃgālīcaraṇavidhūtapakṣmamālāparyastastabakaparāgaparibhadraṃ|".

After that follows the beginning of the first line of verse 41 with "udgacchannavalavalīlavaṃgapuṣpapratyāgratanutaranāṃ" but at that point the scribe suddenly jumped back to the last line of verse 40 and wrote the following:
"gālīcaraṇavidhūtapakṣmamā<space type="binding-hole"/>lāparyastastabakaparāgapāribhadraṃ|| udgacchannavalavalīlavaṃgapuṣpapratyagrātanunarasaurabhābhirāmaṃ|"

I know we had a somewhat similar case which was discussed at [https://github.com/chchch/upama/issues/27] before but in that case there were whole verses that were written fully and then repeated in a different place with revised verses.

I can't quite figure out if and how the @corresp attribute would work in this case here since we don't have a full verse but rather it's cut out from two different ones.

In case it shouldn't get applied here, how am I to go about this issue otherwise?

Looking forward to your help.

How should we understand and encode those dots (p. 61)? Should they be <metamark/>ed (e.g. <metamark place="inline">......</metamark> although the dots are rather at the top of the line not at the bottom)?

From @MAMMHOTH (the template is a TEI-encoded version of the e-text of the Kāvyamālā edition):

In the MS, Canto 24 (as given in the template --> T) breaks off in the middle of T24_35 pāda c and jumps to T25_49b.

From there it continues 'til the end of T25 (HV_25_71) and ends with the final line for canto 24.

Afterwards the MS continues with the verses until the middle of T26_6c where it then jumps back to T25_15c.

From there on the MS continues with the verses up to the middle of T25_49b.

At that point it jumps back to 24_35c to that place where it broke off initially and ends the T24 (after HV_24_50) with the final line for canto 25.

Here the MS continues until T25_15b after which it jumps forward to T26_6c and continues as usual from there on.

• Kāvyamālā https://ia802806.us.archive.org/25/items/Kavyamala_anthology_series_of_Nirnaya_sagar_press/KavyamalaAnthologyVol.01-MahaganapatistotraOfRaghavachaitanyaEtc1929.pdf

• BNF https://gallica.bnf.fr/ark:/12148/btv1b100826020/f101.item.zoom

• E-Text

26,27,32,43,48,49

Start with k-ocr. Have to adapt the buildscript accordingly.

U+00AF ¯

So far we transcribed it as ꣹ (UA8F9), Devanāgarī gap filler.

This is a variant of the same, preferentially used when space to the right of the text is to be left free, whereas the other one is more used on the left side.

Maxi: <!-- [bow] -->

Einicke: VF (Vermeidung von Fehldeutungen) Randausgleich (oder Worttrenner am Zeilenende?)

So far we transcribed it as ¦ (UA6), probably it would be good to wrap it in a <metamark>.

This looks like we got a correction between the two preceding glyphs.

And apply it throughout.

1. Extract a sequence of <p>s:

for i in seq -w 066; do xmlstarlet sel -N x="http://www.tei-c.org/ns/1.0" -t -c "//x:p[@xml:id=\"HVVU_05_$i\"]" ../all/hv-all-k-ocr.xml ; done > HVVU05-a.xml

2. Open the resulting file with vim, and
   a) get rid of namespace attributes: :%s#xmlns="http://www.tei-c.org/ns/1.0" ##gc
   b) get rid of <lb>s: :%s#<lb n="\d\+"/>##gc
   c) get rid of <pb>s: :%s#<pb n="\d\+"/>##gc
   d) Insert linebreaks between <p>s: :%s#</p>\zs\ze<p#\r#gc
   e) Manually bring every <p> into one line, removing hyphenation, empty lines etc.

3. Insert the resulting file into hvvu-all-p.xml, and other places it might be required, with some necessary adjustments.

It's an anusvāra which sort of looks like one of those floral marks that seperate the cantos from another. I wonder if it could be a fancy replacement of kāṃ for ko. How will I encode this though?

Regards,
Maxi