How to get a single Arabic letter in a string with its Unicode transformation value in DELPHI? -

considering arabic word(جبل) made of 3 letters .

-the first letter جـ, -name (ǧīm), -its unicode value fe9f when in beginning, -its basic value 062c , -its isolated value fe9d last 2 values return same shape drawing ج .

now, whenever try single character -trying many different ways-, delphi returns basic unicode value. well,that makes sense,but happens char transformation? single char too..looks takes transformed value when within string, where? how extract it?when , process decides these values? again main question: how can arabic letter or unicode value within string?

just information: unlike english has tow cases letters(capital , small), arabic has 4 cases(isolated, beginning,middle , end) different rules well.

i'm not sure understand question. if want know how write u+fe9f in delphi source code, in modern unicode version of delphi. so:

char($fe9f) 

if want read individual characters جبل this:

const   myword = 'جبل'; var   c: char; .... c := myword[1];//this u+062c 

note code above fine particular word because each code point can encoded single utf-16 widechar character element. if code point required multiple elements, best transform utf-32 code point level processing.

---

now, let's @ string included in question. downloaded question using wget , file came down wires utf-8 encoded. used notepad++ convert utf16-le , picked out 3 utf-16 characters of string. are:

 u+062c u+0628 u+0644 

you stated:

the first letter جـ, name (ǧīm), unicode value u+fe9f.

but incorrect. can seen above, actual character posted u+062c. reason why attempts read first character yield u+062c u+062c first character of string.

---

the bottom line nothing in delphi code transforming character. when do:

s[1] := char($fe9f); 

the compiler performs simple 2 byte copy. there no context aware transformation occurs. , likewise when reading s[1].

---

let's @ how these characters displayed, using simple code on vcl forms application contains memo control:

memo1.clear; memo1.lines.add(stringofchar(char($fe9f), 2)); memo1.lines.add(stringofchar(char($062c), 2)); 

the output looks this:

as can see, rendering layer knows u+062c character appears @ beginning of string.