Arabic Numerals within `<ex>` in DDB

[samosafuz]

It has come to my attention that Arabic numerals appear not infrequently within <ex> in DDB, no doubt a remnant of some past conversion. A good example is (χαλκοῦς 2), in O.Amst. 41.5:

<lb n="5"/><expan>ὁμο<ex>ίως</ex></expan> <gap reason="illegible" quantity="3" unit="character"/> <num value="1"><unclear>α</unclear></num> <expan><ex>δραχμὰς</ex></expan> <num value="4">δ</num> <expan><ex>χαλκοῦς 2</ex></expan><num value="2"/> <gap reason="lost" extent="unknown" unit="character"/> </ab></div>

I searched DDB with the regex <ex>[^<]+\s\d+</ex> and retrieved 7270 occurrences across 557 files. Those have been entered in a Google spreadsheet.

It is desirable to analyze, organize, and update these via XSLT.

[samosafuz]

This issue is not unrelated to #441.

[samosafuz]

A different regex (i.e., [0-9]+</ex>) produced 7793 occurrences in 679 files. Those have been entered in a spreadsheet here.

In the majority cases, it appears that the value is subsequently defined explicitly using num/@value: the regex ([0-9]+)</ex></expan><num value="\1" produces 7605 occurrences in 629 files>.