When copying and pasting an Arabic text from a PDF file generated by context, the characters are not correctly joint.
Consider the following MWE:
\definefontfamily [font] [serif] [Amiri] [features=arabic] \setupbodyfont [font] \setcharacterstripping \setupdirections[bidi=global] \setupalign[r2l] \starttext جملة عربية قصيرة \stoptext
Copying and pasting the short sentence produce:
ة ر ي ص ق ة ي ب ر ع ة ل م ج. The letters are disjointed and have extra spaces between them. By looking at following string:
"\u202bة\u202c \u202bر\u202c \u202bي\u202c \u202bص\u202c \u202bق\u202c \u202bة\u202c \u202bي\u202c \u202bب\u202c \u202bر\u202c \u202bع\u202c ة\u202c \u202bل\u202c \u202bم\u202c \u202bج\u202c"
we can see the DIRECTIONAL FORMATTING chars. My question is how do I get proper PDF output which is copy-paste-able and searchable.
By looking at similar questions, I suppose that font encoding is not the problem here.
I am using
context version: 2015.05.18 12:26, and
luatex version beta-0.80.0 (TeX Live 2015).