Hi Akira, On 30/05/2017, at 19:39, "Akira Kakuto" wrote:

...

Here is a minimal working example that shows the issue.

In the original test.tex, I encountered an assertion error in tounicode.c. If I change as

\pdfcompresslevel=0 \pdfgentounicode=1 \input glyphtounicode.tex \pdfglyphtounicode{Z}{1D481} % MATH bold-italic-Z U+1D481 (U+D835 U+DC81)

Yes, I discovered that this works, shortly after posting. But it doesn't explain why what should be valid input is changed to something invalid. And what should happen with longer strings? For example this works fine: \pdfglyphtounicode{t_t}{00770077} % break ligature into separate letters as does adding multiple combining accents (e.g. 0300 etc.) after the Hex for other characters. But what if it was required to map into multiple higher-plane glyphs? Is the only documentation on this by reading the source file? tounicode.c

Z $Z$

\bye

I obtained an attached test.pdf, in which I can find two strings <5A> <D835DC81>.

Best, Akira

Cheers, Ross

Hi Karl, Akira and others. On May 31, 2017, at 7:19 AM, Karl Berry mailto:karl@freefriends.org> wrote: Is the only documentation on this by reading the source file? tounicode.c As far as I know yes. -k OK. Reading it led me to try with spaces separating the groups of 4 Hex digits. \pdfglyphtounicode{Z}{D835 DC81}% MATH bold-italic-Z U+1D481 (U+D835 U+DC81) This worked! So it really was just an instance of needing some documentation to tell what is the supported syntax. Cheers Ross Dr Ross Moore Mathematics Dept | 12 Wally’s Walk, 734 Macquarie University, NSW 2109, Australia T: +61 2 9850 8955 | F: +61 2 9850 8114 M:+61 407 288 255 | E: ross.moore@mq.edu.aumailto:ross.moore@mq.edu.au http://www.maths.mq.edu.au [cid:image001.png@01D030BE.D37A46F0]http://mq.edu.au/ CRICOS Provider Number 00002J. Think before you print. Please consider the environment before printing this email. This message is intended for the addressee named and may contain confidential information. If you are not the intended recipient, please delete it and notify the sender. Views expressed in this message are those of the individual sender, and are not necessarily the views of Macquarie University.

Hi Akira, On May 31, 2017, at 2:59 PM, Akira Kakuto mailto:kakuto@fuk.kindai.ac.jp> wrote: Hi Karl, I realize you're reporting a separate bug, that the value gets misinterpreted \pdfglyphtounicode{Z}{D835DC81} <5A> <36E537DC81> I confirmed that Ross's \pdfglyphtounicode{Z}{D835 DC81} with a space works ok. In the case of \pdfglyphtounicode{Z}{D835DC81} I encountered an assertion error because long code = 0XD835DC81 < 0 in my case, where sizeof(long) = 4. Ross obtained erroneously vh = 0X36E537, vl = 0XDC81 because long code = 0XD835DC81 > 0, if sizeof(long) = 8. Is assert(code >= 0 && code <= 0X10FFFF) OK or not OK? I'm thinking this is OK, *provided* spaces are used to separate the codes, when multiple glyphs are required. Otherwise there should be just a single Unicode point, and the allowable range for this is <= 0X10FFFF . Indeed the top end of this ( 0X100000 upwards ) is for “Private Use” only. My understanding of these pieces of code: for (i = 0; i < l; i++) { if (p[i] == ' ') valid_unistr = 2; /* if a space occurs we treat this entry as a string */ if (valid_unistr == 2) { /* a string with space(s) */ /* copy p to buf2, ignoring spaces */ for (q = buf2; *p != 0; p++) if (*p != ' ') *q++ = *p; *q = 0; gu->code = UNI_STRING; gu->unicode_seq = xstrdup(buf2); … is that blocks of 4-6 hex digits are just copied verbatim, without calling check_unicode_value so that assert is never actually encountered. Do you agree with this interpretation? (from tounicode.c) static char *utf16be_str(long code) { static char buf[SMALL_BUF_SIZE]; long v; unsigned vh, vl; assert(code >= 0); if (code <= 0xFFFF) sprintf(buf, "%04lX", code); else { v = code - 0x10000; vh = v / 0x400 + 0xD800; vl = v % 0x400 + 0xDC00; sprintf(buf, "%04X%04X", vh, vl); } return buf; } Best, Akira Cheers, Ross Dr Ross Moore Mathematics Dept | 12 Wally’s Walk, 734 Macquarie University, NSW 2109, Australia T: +61 2 9850 8955 | F: +61 2 9850 8114 M:+61 407 288 255 | E: ross.moore@mq.edu.aumailto:ross.moore@mq.edu.au http://www.maths.mq.edu.au [cid:image001.png@01D030BE.D37A46F0]http://mq.edu.au/ CRICOS Provider Number 00002J. Think before you print. Please consider the environment before printing this email. This message is intended for the addressee named and may contain confidential information. If you are not the intended recipient, please delete it and notify the sender. Views expressed in this message are those of the individual sender, and are not necessarily the views of Macquarie University.