Archives
/
notcurses
mirror of https://github.com/dankamongmen/notcurses
2
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

LayoutNewlinesAtBorders unit test #829

Browse Source
pull/938/head
nick black 4 years ago committed by Nick Black
parent adab5ba382
commit df3dc7f8e7
2 changed files with 38 additions and 18 deletions
Show Stats Download Patch File Download Diff File

34
src/lib/notcurses.c
Unescape Escape View File

@ -1643,18 +1643,20 @@ int ncplane_hline_interp(ncplane* n, const cell* c, int len,
return ret;
}
static bool
islinebreak(wchar_t wchar){
// UC_LINE_SEPARATOR + UC_PARAGRAPH_SEPARATOR
if(wchar == '\n'){
return true;
}
const uint32_t mask = UC_CATEGORY_MASK_Zl | UC_CATEGORY_MASK_Zp;
return uc_is_general_category_withtable(wchar, mask);
}
static bool
iswordbreak(wchar_t wchar){
/*
int w = uc_wordbreak_property(wchar);
uc_general_category_t c = uc_general_category(wchar);
fprintf(stderr, "wordbreak property: %d general cat: %08ld (%lc)\n", w, c, wchar);
return (w == WBP_OTHER || w == WBP_NEWLINE || w == WBP_CR || w == WBP_LF);
*/
const uint32_t mask = UC_CATEGORY_MASK_Z |
UC_CATEGORY_MASK_Zs |
UC_CATEGORY_MASK_Zl |
UC_CATEGORY_MASK_Zp;
UC_CATEGORY_MASK_Zs;
return uc_is_general_category_withtable(wchar, mask);
}
@ -1725,7 +1727,7 @@ puttext_premove(ncplane* n, const char* text){
}
}
if(*text && breaker == NULL){
//fprintf(stderr, "ADVANCING DA FOKKER, JA\n");
fprintf(stderr, "ADVANCING DA FOKKER, JA\n");
if(n->scrolling){
scroll_down(n);
}else{
@ -1768,13 +1770,13 @@ int ncplane_puttext(ncplane* n, int y, ncalign_e align, const char* text, size_t
// figure how much text to output on this line
mbstate_t mbstate = {};
int width;
wchar_t w;
// let it go all the way through to dimx. on that last hit of dimx, we
// might catch a space, in which case we want breaker updated. if it's
// not a space, it won't be printed, and we carry the word forward.
// FIXME what ought be done with \n or multiple spaces?
// FIXME what ought be done with multiple/leading spaces?
while(*text && x <= dimx){
//fprintf(stderr, "laying out [%s] at %d <= %d, %zu\n", linestart, x, dimx, text - linestart);
wchar_t w;
size_t consumed = mbrtowc(&w, text, MB_CUR_MAX, &mbstate);
if(consumed == (size_t)-2 || consumed == (size_t)-1){
logerror(n->nc, "Invalid UTF-8 after %zu bytes\n", text - beginning);
@ -1805,7 +1807,7 @@ int ncplane_puttext(ncplane* n, int y, ncalign_e align, const char* text, size_t
x += width;
text += consumed;
}
//fprintf(stderr, "%d/%d OUT! %s %zu %d\n", n->y, n->x, linestart, text - linestart, x);
fprintf(stderr, "oury: %d cursor: %d/%d(%d) OUT! %s %zu %d\n", y, n->y, n->x, n->lenx, linestart, text - linestart, x);
bool overlong = false; // ugh
// if we have no breaker, we got a single word that was longer than our
// line. print what we can and move along. if *text is nul, we're done.
@ -1829,12 +1831,12 @@ int ncplane_puttext(ncplane* n, int y, ncalign_e align, const char* text, size_t
breakercol = dimx;
++breaker; // FIXME need to advance # of bytes in the UTF8 breaker, not 1
}
//fprintf(stderr, "exited at %d (%d) %zu looking at [%.*s]\n", x, dimx, breaker - linestart, (int)(breaker - linestart), linestart);
fprintf(stderr, "exited at %d (%d) %zu looking at [%.*s]\n", x, dimx, breaker - linestart, (int)(breaker - linestart), linestart);
if(breaker != linestart){
totalcols += breakercol;
const int xpos = (align == NCALIGN_LEFT) ? -1 : ncplane_align(n, align, breakercol);
// blows out if we supply a y beyond leny
//fprintf(stderr, "y: %d %ld %.*s\n", y, breaker - linestart, (int)(breaker - linestart), linestart);
fprintf(stderr, "y: %d %ld %.*s\n", y, breaker - linestart, (int)(breaker - linestart), linestart);
if(ncplane_putnstr_yx(n, y, xpos, breaker - linestart, linestart) <= 0){
if(bytes){
*bytes = linestart - beginning;
@ -2580,7 +2582,7 @@ int ncplane_putnstr_aligned(struct ncplane* n, int y, ncalign_e align, size_t s,
int ncplane_putnstr_yx(struct ncplane* n, int y, int x, size_t s, const char* gclusters){
int ret = 0;
//fprintf(stderr, "PUT %zu at %d/%d [%.*s]\n", s, y, x, (int)s, gclusters);
fprintf(stderr, "PUT %zu at %d/%d [%.*s]\n", s, y, x, (int)s, gclusters);
// FIXME speed up this blissfully naive solution
while((size_t)ret < s && *gclusters){
int wcs;

22
tests/layout.cpp
Unescape Escape View File

@ -89,20 +89,38 @@ TEST_CASE("TextLayout") {
SUBCASE("LayoutNewlines") {
auto sp = ncplane_new(nc_, 5, 5, 0, 0, nullptr);
REQUIRE(sp);
const char boundstr[] = "a\nb\nc\nd\ne";
size_t bytes;
const char boundstr[] = "a\nb\nc\nd\ne";
CHECK(0 < ncplane_puttext(sp, 0, NCALIGN_LEFT, boundstr, &bytes));
CHECK(0 == notcurses_render(nc_));
CHECK(bytes == strlen(boundstr));
char* line = ncplane_contents(sp, 0, 0, -1, -1);
REQUIRE(line);
fprintf(stderr, "LINE: [%s]\n", line);
sleep(5);
sleep(3);
CHECK(0 == strcmp(line, "a b c d e"));
free(line);
ncplane_destroy(sp);
}
// ensure we're honoring newlines at the start/end of rows
SUBCASE("LayoutNewlinesAtBorders") {
auto sp = ncplane_new(nc_, 4, 3, 0, 0, nullptr);
REQUIRE(sp);
const char boundstr[] = "ab\ncde\nfgh";
size_t bytes;
CHECK(0 < ncplane_puttext(sp, 0, NCALIGN_LEFT, boundstr, &bytes));
CHECK(0 == notcurses_render(nc_));
CHECK(bytes == strlen(boundstr));
char* line = ncplane_contents(sp, 0, 0, -1, -1);
REQUIRE(line);
fprintf(stderr, "LINE: [%s]\n", line);
sleep(3);
CHECK(0 == strcmp(line, "ab cde fgh"));
free(line);
ncplane_destroy(sp);
}
// lay out text where a wide word crosses the boundary
SUBCASE("LayoutCrossBoundaryWide") {
if(enforce_utf8()){

Write Preview
Loading…
Cancel
Save