Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Cleanup parse loop

Browse Source
This commit is contained in:
Kovid Goyal 2024-09-13 09:07:06 +05:30
parent 84f371805a
commit 1f003e392b
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
2 changed files with 6 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
src/calibre/ebooks/html_entities.c
View File

@ -114,17 +114,15 @@ bad_entity:
static size_t
process_entity(const char *input, size_t input_sz, char *output, size_t *output_pos) {
size_t input_pos = 0;
size_t input_pos = 1; // ignore leading &
while (input_pos < input_sz) {
char ch = input[input_pos++];
if (('a' <= ch && ch <= 'z') || ('A' <= ch && ch <= 'Z') || ('0' <= ch && ch <= '9') || (ch == '#' && input_pos == 1));
else if (ch == ';') *output_pos += add_entity(input, input_pos-1, output + *output_pos);
else {
output[(*output_pos)++] = '&';
else if (ch == ';') { *output_pos += add_entity(input, input_pos-1, output + *output_pos); return input_pos; }
else break;
}
memcpy(output + *output_pos, input, input_pos);
*output_pos += input_pos;
}
}
return input_pos;
}
@ -134,12 +132,6 @@ replace(const char *input, size_t input_sz, char *output, int keep_xml_entities)
while (input_pos < input_sz) {
const char *p = (const char*)memchr(input + input_pos, '&', input_sz - input_pos);
if (p) {
if (p > input + input_pos) {
size_t sz = p - (input + input_pos);
memcpy(output + output_pos, input + input_pos, sz);
output_pos += sz;
input_pos += sz;
}
input_pos += process_entity(p, input_sz - (p - input), output, &output_pos);
} else {
memcpy(output + output_pos, input + input_pos, input_sz - input_pos);

1
src/calibre/ebooks/html_entities.py
View File

@ -2142,6 +2142,7 @@ def find_tests():
from calibre_extensions.fast_html_entities import replace_entities
def t(inp, exp):
self.assertEqual(exp, replace_entities(inp), f'Failed for input: {inp!r}')
t('&amp', '&amp')
t('', '')
t('a', 'a')