a small utility for comparing near-exact text files at the level of individual characters, made by a philologist, for philologists or anyone else
Squinter takes the strain out of comparing near-exact versions of text with long or otherwise difficult-to-read segments, by helping identify what actually needs attention and freeing you to ignore the rest. Its basic unit is the newline-separated lines, so input files should be formatted accordingly.
(English lexicographer Samuel Johnson, expert squinter)
While comparing and merging two digitizations of a particular (Sanskrit) text containing very long words, I became tired of how BBEdit's "compare" function, which I otherwise like for merging files, does not distinguish differences below the level of individual words. I needed to know which individual characters differed, without being distracted by the rest of the line.
Squinter is above, in the browser,
Preview is open in the middle with the PDF image,
and BBEdit is open below.
Modifications are done in BBEdit.
python app.py --use-local-mode \
--local-filepath1 "/Users/tyler/Dropbox/Docs/Projects/Computational/Digitizations/bāṇa_kādambarī/5_merge/2_Tylers_new.txt" \
--local-filepath2 "/Users/tyler/Dropbox/Docs/Projects/Computational/Digitizations/bāṇa_kādambarī/5_merge/1_Andrew_derived.txt" \
--start-percentage 15 \
--end-percentage 50