Nov5 pre-release: entry ids are unreliable
Brought to you by:
luthien-merilin
Menu
▾
▴
#7 Nov5 pre-release: entry ids are unreliable
The "id" attribute of "entry" elements in the original XML TEI source was apparently used to index the dictionary.
However, it is not always reliable, as it is an internal reference - it could even have been a raw number or anything else provided it is unique - and it's just per chance that I initially used something looking like the entry text itself.
Ideally we should use the firth "orth" element in the first "form" element of an "entry" element to ensure the correct word is indexed.
See attached screenshot for an example of the issue - This one is not critical, but there could be other cases where this might be more serious (for instance, words which had a typo fixed at some time, but without changing the internal id).
Anotated screenshot (png)