JoakimSoderberg
/
catcierge
/
371
/
3
89%
master: 89%
master: 89%
|
Ran
|
Files
14
|
Run time
3s
|
Badge
Embed ▾
README BADGES
|
coverage: 87.508% (-0.1%) from 87.647%
push
travis-ci
JoakimSoderbergBugfixes in template output and more Turns out template output was broken in more than one way. - When generating a template, we have to realloc also in the case when expanding a variable. Because otherwise if one has several variables with a lot of input, that themselves don't cause a realloc, if those then are followed by a bunch of normal text in the template, the buffer will run out. - Because how the match_count was being used in inconsistent ways, and an old "bugfix" of mine in 470a49fb2 which reset it to 0 on"match_group_end". Only the first "match_*" variables could be used in any template. Since there is a check that the template variable doesn't specify an index that is greater than match_count... And we had just set it to 0 before calling the template generator. - Also the recursion check for template variables was kind of broken and could've possibly had some edge cases where it would go nuts. Now the error messages will also make sure, and differentiate between missing variables and a recursion failure.
2683 of 3066 relevant lines covered (87.51%)
252587.42 hits per line
