leading % stripped

[adamyg]

Any leading %,are stripped from the output, assume related to digraph parsing:

example input "test.txt":

%
%x
%% %%
%%%

result:

#line 1 "test.txt"

x
% %%
%%

[decahedron1]

The problem might be in this code but I'm not sure. This section looks fine though.

[adamyg]

Hi, I did tweak the same, which solved the problem partially yet further test cases uncovered additional digraph issues. } else if (z == S_PCT && ls->ltwnl) { /*XXX*/ put_char(ls, '%'); outc = c; } else if (z == S_LCHAR || z == S_SLASH || (z == S_SHARP && ls->ltwnl) //XXX || (z == S_PCT && ls->ltwnl) || (z == S_BACKSLASH)) { outc = c;

…

On Tue, Mar 23, 2021 at 10:17 PM Carson M. ***@***.***> wrote: The problem might be in this code <https://github.com/lpsantil/ucpp/blob/master/lexer.c#L930-L953> but I'm not sure. This section <https://github.com/lpsantil/ucpp/blob/master/lexer.c#L173-L185> looks fine though. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKF46VU4WTS2BG52BGG65QLTFCPIBANCNFSM4ZO6RANQ> .

[decahedron1]

Could you describe the additional issues you're running into?

Also, that solution isn't really a solution in my opinion; you basically just removed digraph support 😅

[adamyg]

Hi, Agree wasn't a solution, only a hint towards one; the following looks closer, additional testing required. One question, behaviour of -s and stand-alone "%:", as they are removed unconditionally unlike "#" ? thx if (ttSTO(nstat)) { if (S_TOKEN(noMOD(nstat))) { wan(ls->ctok->name, ltok, (unsigned char)0, ls->tknl); } ls->ctok->type = noMOD(nstat); if (outc) { //XXX: flush partial matches if (outc > 0) { put_char(ls, outc); } outc = 0; } break; }

…

On Wed, Mar 24, 2021 at 1:46 AM Carson M. ***@***.***> wrote: Could you describe the additional issues you're running into? Also, that solution isn't *really* a solution in my opinion; you basically just removed digraph support 😅 — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKF46VX53UJR4HOWBLFQGJ3TFDHWBANCNFSM4ZO6RANQ> .