Archives/notcurses
2
0
Fork 0
mirror of https://github.com/dankamongmen/notcurses.git synced 2024-11-08 01:10:23 +00:00
Code Issues Packages Projects Releases Wiki Activity

[ncman] parse all five arguments of TH

Browse Source
This commit is contained in:
nick black 2021-12-08 05:44:34 -05:00
parent e23cf59f57
commit f8626e48cb
No known key found for this signature in database
GPG Key ID: 5F43400C21CBFACC
1 changed files with 61 additions and 110 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

171
src/man/main.c
View File

@ -2,6 +2,7 @@
#include <errno.h>
#include <unistd.h>
#include <stdlib.h>
#include <wctype.h>
#include <getopt.h>
#include <inttypes.h>
#include <sys/mman.h>
@ -314,6 +315,8 @@ typedef struct pagedom {
char* title;
char* section;
char* version;
char* footer;
char* header;
} pagedom;
static const char*
@ -321,131 +324,79 @@ dom_get_title(const pagedom* dom){
return dom->title;
}
// get the next token. first, chew whitespace. then match a string of
// iswgraph(), or a quoted string of iswprint(). return the number of
// characters consumed, or -1 on error (no token, unterminated quote).
// heap-copies the utf8 to *token on success.
static int
lex_next_token(const char* s, char** token){
mbstate_t ps = {};
wchar_t w;
size_t b, cur;
cur = 0;
bool inquote = false;
const char* tokstart = NULL;
while((b = mbrtowc(&w, s + cur, MB_CUR_MAX, &ps)) != (size_t)-1 && b != (size_t)-2){
if(tokstart){
if(b == 0 || (inquote && w == L'"') || (!inquote && iswspace(w))){
if(!tokstart || !*tokstart || *tokstart == '"'){
return -1;
}
*token = strndup(tokstart, cur - (tokstart - s));
return cur + b;
}
}else{
if(iswspace(w)){
cur += b;
continue;
}
if(w == '"'){
inquote = true;
tokstart = s + cur + b;
}else{
tokstart = s + cur;
}
}
cur += b;
}
return -1;
}
// take the newly-added title section, and extract the title, section, and
// version (technically footer-middle, footer-inside, and header-middle).
// they ought be quoted, but might not be.
static int
lex_title(pagedom* dom){
const char* tok = dom->root->text;
while(isspace(*tok)){
++tok;
}
bool quoted = false;
if(*tok == '"'){
quoted = true;
++tok;
}
if(!*tok){
int b = lex_next_token(tok, &dom->title);
if(b < 0){
fprintf(stderr, "couldn't extract title [%s]\n", dom->root->text);
return -1;
}
const char* endtok = tok + 1;
while(*endtok){
if(!quoted){
if(isspace(*endtok)){
break;
}else if(*endtok == '"'){
quoted = true;
break;
}
}else{
if(*endtok == '"'){
quoted = false;
break;
}
}
++endtok;
}
if(!*endtok){
fprintf(stderr, "couldn't extract title [%s]\n", dom->root->text);
return -1;
}
dom->title = strndup(tok, endtok - tok);
tok = endtok + 1;
if(!*tok){
tok += b;
b = lex_next_token(tok, &dom->section);
if(b < 0){
fprintf(stderr, "couldn't extract section [%s]\n", dom->root->text);
return -1;
}
if(!quoted){
while(isspace(*tok)){
++tok;
tok += b;
b = lex_next_token(tok, &dom->version);
if(b < 0){
//fprintf(stderr, "couldn't extract version [%s]\n", dom->root->text);
return 0;
}
quoted = false;
if(*tok == '"'){
quoted = true;
++tok;
tok += b;
b = lex_next_token(tok, &dom->footer);
if(b < 0){
//fprintf(stderr, "couldn't extract footer [%s]\n", dom->root->text);
return 0;
}
if(!*tok){
fprintf(stderr, "couldn't extract section [%s]\n", dom->root->text);
return -1;
tok += b;
b = lex_next_token(tok, &dom->header);
if(b < 0){
//fprintf(stderr, "couldn't extract header [%s]\n", dom->root->text);
return 0;
}
}
endtok = tok + 1;
while(*endtok){
if(!quoted){
if(isspace(*endtok)){
break;
}else if(*endtok == '"'){
quoted = true;
break;
}
}else{
if(*endtok == '"'){
quoted = false;
break;
}
}
++endtok;
}
if(!*endtok){
fprintf(stderr, "couldn't extract section [%s]\n", dom->root->text);
return -1;
}
dom->section = strndup(tok, endtok - tok);
tok = endtok + 1;
if(!*tok){
fprintf(stderr, "couldn't extract version [%s]\n", dom->root->text);
dom->version = strdup("");
return 0; // allow empty version
}
if(!quoted){
while(isspace(*tok)){
++tok;
}
quoted = false;
if(*tok == '"'){
quoted = true;
++tok;
}
if(!*tok){
fprintf(stderr, "couldn't extract version [%s]\n", dom->root->text);
dom->version = strdup("");
return 0; // allow empty version
}
}
endtok = tok + 1;
while(*endtok){
if(!quoted){
if(isspace(*endtok)){
break;
}else if(*endtok == '"'){
quoted = true;
break;
}
}else{
if(*endtok == '"'){
quoted = false;
break;
}
}
++endtok;
}
if(!*endtok){
fprintf(stderr, "couldn't extract version [%s]\n", dom->root->text);
dom->version = strdup("");
return 0; // allow empty version
}
dom->version = strndup(tok, endtok - tok);
return 0;
}