I have a set of plain text files with a mix of Hebrew and English. These files are from the late 90s to early 2000s and were apparently written in NisusWriter.
When I open the text files the english lines render correctly but the Hebrew lines are jumbled up mojibake like this:
Â∆˙ŸÙœÏ∆À˙À̆††ÂŸ‰À÷Õ·†††‡Œ˙†††‰ÀÚ⁄·«„À‰††††
I wrote a loop to run iconv with every encoding it supports but none of the outputs were fixed
Running hexdump -C on the first 3 lines (one English, one jumbled Hebrew, and one transliteration in Latin characters) gave the following. It seems the jumbled Hebrew is written as just . chars
00000010 50 2e 20 31 31 30 20 20 2d 20 41 56 4f 44 41 48 |P. 110 - AVODAH|
00000020 0d 0d 20 f8 d9 f6 cd e4 a0 ac a0 a0 a0 e9 d9 e9 |.. .............|
00000030 cb a0 a0 a0 e0 db ec dd e4 cd d8 e9 f0 e5 c6 a0 |................|
00000040 ac a0 a0 a0 e1 c6 d9 f2 cc ee c6 d9 ea cb a0 a0 |................|
00000050 a0 e9 cf f9 dd d9 f8 cb e0 cd ec a0 ac 0d 0d 52 |...............R|
00000060 65 2d 74 7a 65 68 d5 2c 20 20 20 20 41 64 6f 6e |e-tzeh., Adon|
00000070 61 69 20 20 20 20 20 20 45 2d 6c 6f 2d 68 65 69 |ai E-lo-hei|
00000080 d5 2d 6e 75 20 2c 20 20 20 20 20 20 20 62 65 2d |.-nu , be-|
00000090 61 6d 2d 63 68 61 d5 20 20 20 20 20 20 20 20 79 |am-cha. y|
000000a0 69 73 2d 72 61 2d 65 6c d5 20 0d 62 65 20 70 6c |is-ra-el. .be pl|
000000b0 65 61 73 65 64 2c 20 20 20 20 41 64 6f 6e 61 69 |eased, Adonai|