punctuation signs interrupting words

[arlogriffiths]

@danbalogh @michaelnmmeyer —

in cases like

<l n="a" enjamb="yes">d<choice><sic>i</sic><corr>ī</corr></choice>kṣā-nimittaṁ kila sarvva-bhāvā<g type="dotMid">.</g></l>
<l n="b">nāṁ svāvaśeṣaṁ vara-bhasma tatra <g type="dotMid">.</g></l>

or

<l n="a">saṁpady aśeṣāṁ pr̥thivīṁ niyogā<g type="dotMid">.</g>d</l>
<l n="b">yo bhasma-de<lb n="A9" break="yes"/>śām anuśāsti viṣṇum· <g type="dotMid">.</g></l>

(both examples taken from DHARMA_INSCIC00141 — https://dharmalekha.info/texts/INSCIC00141)

we find <g> interrupting words. Is any special measure required to ensure indexability of such words, or can Michaël's machine be taught to ignore any <g> altogether?

does EGD address such cases at all, Dan?

(cc to @salomepichon and @chhomkunthea for their information)

[arlogriffiths]

I don't understand why a space is inserted after ° inside sarvva-bhāvā°nāṁ in physical display

[michaelnmmeyer]

I propose we address both these questions later on.

[danbalogh]

On the encoding side, what you have done is sufficient, but there's the option of flagging it as <orig>, which I would recommend, and for the second item, also the option of normalising it, which I would also recommend.
The applicable EGD bits are at the end of §4.2.1 (before the example with paṁcaviṁśati<choice><orig>|n</orig><reg>M|</reg></choice> tat-putro). The situation is also essentially the same as with space in your <l n="a">jayatīndrādidevāsya<space/>ś</l><l n="b">śrīmān yajñapatīśvaraḥ</l>, which is in EGD §4.3.3, though flagging/normalisation is not mentioned there (though perhaps it should; I've made a note of that for myself).
The EGD does not specifically address cases like this in verse, but here is one from my VengiCalukya0078:

<lg n="11" met="anuṣṭubh">
...
<l n="c">sarvva-lokāśraye yasmi<lb n="51" break="no"/><choice><orig>|n</orig><reg>N|</reg></choice></l>
<l n="d">saty<unclear>a</unclear>-rāj<unclear>e</unclear> sthitaṁ jagaT|</l>
</lg>

For word indexing, I'm sure it should not be a problem to ignore g elements.