ESP32 library for displays 1602, 2004, MT-20S4M (I2C interface) with Russian letters (Cyrillic)

[Tesla_X]

Hello. Who can tell you where to get the ESP32 library for displays 1602, 2004 (I2C interface) with Russian letters (Cyrillic). By the link "https://github.com/dhylands/python_lcd " there is no library for ESP32 I2C, but there is for ESP8266. Will it work on ESP32?

[davef]

I would expect that it would run on a ESP32. GPIO assignment will probably need looking at, as some pins are reserved for other functions on each device.

[Tesla_X]

I checked on ESP 32 - it works. Tested the library on the MT-20S4M_en LCD display. The problem is that it does not display Russian letters. Judging by the symbol table (attached), there is an offset. How can I fix it?
https://github.com/dhylands/python_lcd/ ... en.LCD.pdf

[dhylands]

I haven't really dealt with non-ASCII characters, but my understanding is that micropython strings are UTF8 encoded, and the LCD isn't. So you'll need to translate your strings from UTF8 to the codeset needed by the LCD.

Can you provide some examples of what you're trying to do, and elaborate on what exactly you mean by "there is an offset".

[Tesla_X]

Hello. I use the MT-20S4M_en LCD display. I attached the documentation above. In the first photo in the attachment, part of the code for displaying lines on the display. The second photo shows the output result. As you can see from the photo, the characters in the third and fourth lines do not correspond to the characters in the code. If you look at the symbol table on page 9 of the documentation for the display, you can see that there is a certain offset when comparing the symbols according to the table. Or the library processes 8 bits of a character, and you need 16. Tell me how can I fix it?

[dhylands]

What's the output of the following?

Code: Select all

import binascii
print(binascii.hexlify("The string with your characters here"))

[Tesla_X]

Code: Select all

>>> print(binascii.hexlify("Test string"))
b'5465737420737472696e67'
>>> print(binascii.hexlify("/n"))
b'2f6e'
>>> print(binascii.hexlify("Русские символы"))
b'd0a0d183d181d181d0bad0b8d0b520d181d0b8d0bcd0b2d0bed0bbd18b'
>>> 

[dhylands]

The hex data that you printed is the UTF-8 encoding of the string. The Cyrillic alphabet can be seen on this page: https://en.wikipedia.org/wiki/ISO/IEC_8859-5

If you iterate through each of the characters of the string character by character and subtract 864 from the unicode value, you'll see something like this:

Code: Select all

import binascii

s = "Русские символы"
print(binascii.hexlify(s))

for ch in s:
    if ord(ch) >= 864:
        print(ch, hex(ord(ch)), hex(ord(ch) - 864))
    else:
        print(ch, hex(ord(ch)))

which produces this output:

Code: Select all

b'd0a0d183d181d181d0bad0b8d0b520d181d0b8d0bcd0b2d0bed0bbd18b'
Р 0x420 0xc0
у 0x443 0xe3
с 0x441 0xe1
с 0x441 0xe1
к 0x43a 0xda
и 0x438 0xd8
е 0x435 0xd5
  0x20
с 0x441 0xe1
и 0x438 0xd8
м 0x43c 0xdc
в 0x432 0xd2
о 0x43e 0xde
л 0x43b 0xdb
ы 0x44b 0xeb

You'll notice that the rightmost column corresponds to the index into the "Codepage layout"

I took a look at the docs for your LCD and it doesn't look like the characters map directly to the codepage, you'll need to create your own map. I would make a 256 entry lookup table, where you use (ord(ch) - 864) as the index into the table and the value is the value to send to the LCD.

For example, the first character looks like a capital P, and the only thing I see like that is the ASCII P at 0x50. So map[0xc0] would have the value 0x50. The backwards N appears on the LCD with a code of 0xb8, so map[0xd8] would equal 0xb8. Rather than use an actual array, I would probably using a 256 byte bytearray. This will take up less memory than a list.

You could also just treat the values 0x400 thru 0x45f specially, and subtract 0x400 (1024) from the ord(ch) and get a value from 0 thru 95. Then your map would only need the 96 values corresponds to values 0xA0 thru 0xFF on the Codelayout.

[Tesla_X]

The hex data that you printed is the UTF-8 encoding of the string. The Cyrillic alphabet can be seen on this page: https://en.wikipedia.org/wiki/ISO/IEC_8859-5

If you iterate through each of the characters of the string character by character and subtract 864 from the unicode value, you'll see something like this:

Code: Select all

import binascii

s = "Русские символы"
print(binascii.hexlify(s))

for ch in s:
    if ord(ch) >= 864:
        print(ch, hex(ord(ch)), hex(ord(ch) - 864))
    else:
        print(ch, hex(ord(ch)))

which produces this output:

Code: Select all

b'd0a0d183d181d181d0bad0b8d0b520d181d0b8d0bcd0b2d0bed0bbd18b'
Р 0x420 0xc0
у 0x443 0xe3
с 0x441 0xe1
с 0x441 0xe1
к 0x43a 0xda
и 0x438 0xd8
е 0x435 0xd5
  0x20
с 0x441 0xe1
и 0x438 0xd8
м 0x43c 0xdc
в 0x432 0xd2
о 0x43e 0xde
л 0x43b 0xdb
ы 0x44b 0xeb

You'll notice that the rightmost column corresponds to the index into the "Codepage layout"

I took a look at the docs for your LCD and it doesn't look like the characters map directly to the codepage, you'll need to create your own map. I would make a 256 entry lookup table, where you use (ord(ch) - 864) as the index into the table and the value is the value to send to the LCD.

For example, the first character looks like a capital P, and the only thing I see like that is the ASCII P at 0x50. So map[0xc0] would have the value 0x50. The backwards N appears on the LCD with a code of 0xb8, so map[0xd8] would equal 0xb8. Rather than use an actual array, I would probably using a 256 byte bytearray. This will take up less memory than a list.

You could also just treat the values 0x400 thru 0x45f specially, and subtract 0x400 (1024) from the ord(ch) and get a value from 0 thru 95. Then your map would only need the 96 values corresponds to values 0xA0 thru 0xFF on the Codelayout.

Thanks for the detailed explanation. I have no experience in drawing up bypass maps yet, and I can send symbols using the library specified above. You can write an example of a solution to my problem in the code applied to this library. In theory, there is a table of Russian characters sewn into the display, where each letter has a specific code. The library should read this code, find a match in the table and output the symbol.

[scruss]

The mapping for Page 1 (Table 6, p.9 of the datasheet) is fairly straightforward, subtracting 848 from the Unicode character number:

Code: Select all

LCD     Python  Unicode         
Decimal Decimal Entity  Letter  Full Unicode Glyph Name
======= ======= ======= ======  ===============================
192	1040	U+0410	А	CYRILLIC CAPITAL LETTER A
193	1041	U+0411	Б	CYRILLIC CAPITAL LETTER BE
194	1042	U+0412	В	CYRILLIC CAPITAL LETTER VE
195	1043	U+0413	Г	CYRILLIC CAPITAL LETTER GHE
196	1044	U+0414	Д	CYRILLIC CAPITAL LETTER DE
197	1045	U+0415	Е	CYRILLIC CAPITAL LETTER IE
198	1046	U+0416	Ж	CYRILLIC CAPITAL LETTER ZHE
199	1047	U+0417	З	CYRILLIC CAPITAL LETTER ZE
200	1048	U+0418	И	CYRILLIC CAPITAL LETTER I
201	1049	U+0419	Й	CYRILLIC CAPITAL LETTER SHORT I
202	1050	U+041A	К	CYRILLIC CAPITAL LETTER KA
203	1051	U+041B	Л	CYRILLIC CAPITAL LETTER EL
204	1052	U+041C	М	CYRILLIC CAPITAL LETTER EM
205	1053	U+041D	Н	CYRILLIC CAPITAL LETTER EN
206	1054	U+041E	О	CYRILLIC CAPITAL LETTER O
207	1055	U+041F	П	CYRILLIC CAPITAL LETTER PE
208	1056	U+0420	Р	CYRILLIC CAPITAL LETTER ER
209	1057	U+0421	С	CYRILLIC CAPITAL LETTER ES
210	1058	U+0422	Т	CYRILLIC CAPITAL LETTER TE
211	1059	U+0423	У	CYRILLIC CAPITAL LETTER U
212	1060	U+0424	Ф	CYRILLIC CAPITAL LETTER EF
213	1061	U+0425	Х	CYRILLIC CAPITAL LETTER HA
214	1062	U+0426	Ц	CYRILLIC CAPITAL LETTER TSE
215	1063	U+0427	Ч	CYRILLIC CAPITAL LETTER CHE
216	1064	U+0428	Ш	CYRILLIC CAPITAL LETTER SHA
217	1065	U+0429	Щ	CYRILLIC CAPITAL LETTER SHCHA
218	1066	U+042A	Ъ	CYRILLIC CAPITAL LETTER HARD SIGN
219	1067	U+042B	Ы	CYRILLIC CAPITAL LETTER YERU
220	1068	U+042C	Ь	CYRILLIC CAPITAL LETTER SOFT SIGN
221	1069	U+042D	Э	CYRILLIC CAPITAL LETTER E
222	1070	U+042E	Ю	CYRILLIC CAPITAL LETTER YU
223	1071	U+042F	Я	CYRILLIC CAPITAL LETTER YA
224	1072	U+0430	а	CYRILLIC SMALL LETTER A
225	1073	U+0431	б	CYRILLIC SMALL LETTER BE
226	1074	U+0432	в	CYRILLIC SMALL LETTER VE
227	1075	U+0433	г	CYRILLIC SMALL LETTER GHE
228	1076	U+0434	д	CYRILLIC SMALL LETTER DE
229	1077	U+0435	е	CYRILLIC SMALL LETTER IE
230	1078	U+0436	ж	CYRILLIC SMALL LETTER ZHE
231	1079	U+0437	з	CYRILLIC SMALL LETTER ZE
232	1080	U+0438	и	CYRILLIC SMALL LETTER I
233	1081	U+0439	й	CYRILLIC SMALL LETTER SHORT I
234	1082	U+043A	к	CYRILLIC SMALL LETTER KA
235	1083	U+043B	л	CYRILLIC SMALL LETTER EL
236	1084	U+043C	м	CYRILLIC SMALL LETTER EM
237	1085	U+043D	н	CYRILLIC SMALL LETTER EN
238	1086	U+043E	о	CYRILLIC SMALL LETTER O
239	1087	U+043F	п	CYRILLIC SMALL LETTER PE
240	1088	U+0440	р	CYRILLIC SMALL LETTER ER
241	1089	U+0441	с	CYRILLIC SMALL LETTER ES
242	1090	U+0442	т	CYRILLIC SMALL LETTER TE
243	1091	U+0443	у	CYRILLIC SMALL LETTER U
244	1092	U+0444	ф	CYRILLIC SMALL LETTER EF
245	1093	U+0445	х	CYRILLIC SMALL LETTER HA
246	1094	U+0446	ц	CYRILLIC SMALL LETTER TSE
247	1095	U+0447	ч	CYRILLIC SMALL LETTER CHE
248	1096	U+0448	ш	CYRILLIC SMALL LETTER SHA
249	1097	U+0449	щ	CYRILLIC SMALL LETTER SHCHA
250	1098	U+044A	ъ	CYRILLIC SMALL LETTER HARD SIGN
251	1099	U+044B	ы	CYRILLIC SMALL LETTER YERU
252	1100	U+044C	ь	CYRILLIC SMALL LETTER SOFT SIGN
253	1101	U+044D	э	CYRILLIC SMALL LETTER E
254	1102	U+044E	ю	CYRILLIC SMALL LETTER YU
255	1103	U+044F	я	CYRILLIC SMALL LETTER YA

But I don't know how easy it is to switch between character ROM pages. Page 0 is not so easy to work out