# HG changeset patch # User Christian Brabandt # Date 1517752804 -3600 # Node ID dc39ef257b6093bda1172d87344ac6099feb15c5 # Parent 13fed273cfd6414e868ff98eb7adf94bb3331239 patch 8.0.1467: libvterm doesn't handle illegal byte sequence correctly commit https://github.com/vim/vim/commit/fef4ddd5eb8816a6607a624aa401bcfa71a63def Author: Bram Moolenaar Date: Sun Feb 4 14:49:57 2018 +0100 patch 8.0.1467: libvterm doesn't handle illegal byte sequence correctly Problem: Libvterm doesn't handle illegal byte sequence correctly. Solution: After the invalid code check if there is space to store another character. Allocate one more character. (zhykzhykzhyk, closes #2614, closes #2613) diff --git a/src/libvterm/src/encoding.c b/src/libvterm/src/encoding.c --- a/src/libvterm/src/encoding.c +++ b/src/libvterm/src/encoding.c @@ -46,14 +46,16 @@ static void decode_utf8(VTermEncoding *e return; else if(c >= 0x20 && c < 0x7f) { - if(data->bytes_remaining) + if(data->bytes_remaining) { + data->bytes_remaining = 0; cp[(*cpi)++] = UNICODE_INVALID; - + if (*cpi >= cplen) + break; + } cp[(*cpi)++] = c; #ifdef DEBUG_PRINT_UTF8 printf(" UTF-8 char: U+%04x\n", c); #endif - data->bytes_remaining = 0; } else if(c == 0x7f) /* DEL */ diff --git a/src/libvterm/src/state.c b/src/libvterm/src/state.c --- a/src/libvterm/src/state.c +++ b/src/libvterm/src/state.c @@ -248,8 +248,9 @@ static int on_text(const char bytes[], s VTermPos oldpos = state->pos; - /* We'll have at most len codepoints */ - codepoints = vterm_allocator_malloc(state->vt, len * sizeof(uint32_t)); + /* We'll have at most len codepoints, plus one from a previous incomplete + * sequence. */ + codepoints = vterm_allocator_malloc(state->vt, (len + 1) * sizeof(uint32_t)); encoding = state->gsingle_set ? &state->encoding[state->gsingle_set] : diff --git a/src/version.c b/src/version.c --- a/src/version.c +++ b/src/version.c @@ -772,6 +772,8 @@ static char *(features[]) = static int included_patches[] = { /* Add new patch number below this line */ /**/ + 1467, +/**/ 1466, /**/ 1465,