On Mon, 8 Jun 2026 22:17:16 +0000, Robert Prins wrote:
>>
>> Does IND$FILE even document what it does?
>>
>No, but is that necessary? I've zapped it to use SDB (thanks to the late
>GSF) and it would be equally easy(ish) to zap the translate table it uses
>to make them equivalent to those use by FTP.
>
No, translation to/from UTF-8 can not be done
with a table. I could code it from memory, but not
a table. Does GSF'S code handle UTF-8?
>>
>> Transfer the UTF-8 file again in BINARY.
>>
>No, because the data is in ASCII, except for the locations
>
ASCII is a subset of BINARY.
"except for"??? The sun always shines except for at night.
>>
>> >by either, but the program uses data embedded in the original white box
>> >file to actually build the first FB(121) record, which looks like
>> >
>> > | B | Brussel (Metro Delta) |
>> > | D | Raststätte Gütersloh Süd |
>> > | F | Calais |
>> > | GB | Brentwood M25 (W) |
>> > | L | Aire de Capellen Nord |
>> > | LT | A13 exit Palanga (K) |
>> > | LV | exit LiepÄ ja (Bus Dienvida robeža) |
>> > | NL | VZP De Kroon |
>> > | PL | MOP Sosna |
>> >
>> That's implausible. Are the "¼" silent?
>>
>They're the codepage-437/850 representation of "Raststätte Gütersloh Süd"
>and "exit Liepāja (Bus Dienvida robeža)".
>
Here's CP437:
Host: UTF-8 output: IBM437
0 16 32 48 64 80 96 112 128 144 160 176 192 208 224 240
0 10 20 30 40 50 60 70 80 90 A0 B0 C0 D0 E0 F0
0 0 0 @ P ` p á ░ └ ╨ α ≡
1 1 ! 1 A Q a q í ▒ ┴ ╤ ß ±
2 2 " 2 B R b r ó ▓ ┬ ╥ Γ ≥
3 3 # 3 C S c s ú │ ├ ╙ π ≤
4 4 $ 4 D T d t ñ ┤ ─ ╘ Σ ⌠
5 5 % 5 E U e u Ñ ╡ ┼ ╒ σ ⌡
6 6 & 6 F V f v ª ╢ ╞ ╓ µ ÷
7 7 ' 7 G W g w º ╖ ╟ ╫ τ ≈
8 8 ( 8 H X h x ¿ ╕ ╚ ╪ Φ °
9 9 ) 9 I Y i y ⌐ ╣ ╔ ┘ Θ ∙
10 A * : J Z j z ¬ ║ ╩ ┌ Ω ·
11 B + ; K [ k { ½ ╗ ╦ █ δ √
12 C , < L \ l | ¼ ╝ ╠ ▄ ∞ ⁿ
13 D - = M ] m } ╜ ═ ▌ φ ²
14 E . > N ^ n ~ « ╛ ╬ ▐ ε ■
15 F / ? O _ o » ┐ ╧ ▀ ∩
And CP850:
Host: UTF-8 output: IBM850
0 16 32 48 64 80 96 112 128 144 160 176 192 208 224 240
0 10 20 30 40 50 60 70 80 90 A0 B0 C0 D0 E0 F0
0 0 0 @ P ` p á ░ └ ð Ó
1 1 ! 1 A Q a q í ▒ ┴ Ð ß ±
2 2 " 2 B R b r ó ▓ ┬ Ê Ô ‗
3 3 # 3 C S c s ú │ ├ Ë Ò ¾
4 4 $ 4 D T d t ñ ┤ ─ È õ ¶
5 5 % 5 E U e u Ñ Á ┼ ı Õ §
6 6 & 6 F V f v ª Â ã Í µ ÷
7 7 ' 7 G W g w º À Ã Î þ ¸
8 8 ( 8 H X h x ¿ © ╚ Ï Þ °
9 9 ) 9 I Y i y ® ╣ ╔ ┘ Ú ¨
10 A * : J Z j z ¬ ║ ╩ ┌ Û ·
11 B + ; K [ k { ½ ╗ ╦ █ Ù ¹
12 C , < L \ l | ¼ ╝ ╠ ▄ ý ³
13 D - = M ] m } ¢ ═ ¦ Ý ²
14 E . > N ^ n ~ « ¥ ╬ Ì ¯ ■
15 F / ? O _ o » ┐ ¤ ▀ ´
Does either of them contain the troublesome characters?
>How many times has it already been stepped on?
>>
>> Start over; use BINARY.
>>
>Already told you, the file is ASCII with just locations in UTF-8.
>
ASCII is a subset of UTF-8; the first 128 code points.
>No, it's transferred correctly, I cut the text from my non-UTF-8 aware
>editor.
>
Get a better terminal (emulator). There are many that grok UTF-8,
and some are very cheap.
--
gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
