fix limitation for size of data:*:base64 images and max length of the html input line #454
Conversation
| @@ -420,18 +420,22 @@ file_find_check(const char *filename) /* I - File or URL */ | |||
|
|
|||
| if ((data = strstr(filename, ";base64,")) != NULL) | |||
| { | |||
| len = sizeof(buffer); | |||
| len = strlen(filename); | |||
There was a problem hiding this comment.
I'd prefer to have some limits in place to avoid out-of-control memory allocations.
There was a problem hiding this comment.
agree... actually it must be limited in htmllib.cxx line(s) reader
10240 is definitely too small, limit it to something x100 times bigger ? is not big problem for server application ?
There was a problem hiding this comment.
That might work, will run the numbers...
| @@ -15,7 +15,7 @@ | |||
| #include "htmldoc.h" | |||
| #include "http.h" | |||
| #include <ctype.h> | |||
|
|
|||
| #include <string> | |||
There was a problem hiding this comment.
... and while I do use C++ in HTMLDOC, I've avoided pulling in the standard C++ library since it seems to change with every major OS release on multiple platforms...
There was a problem hiding this comment.
i thought this file was renamed to .cxx keeping in mind idea write c++ code :)
imho std::string is very very stable, it can be replaced by std::vector or do you suggest to implement custom byte vector ?
There was a problem hiding this comment.
libstdc++ has changed in incompatible ways 6 times in the last ~20 years, making it very hard to provide binaries of HTMLDOC that depend on it.
this patch removes these limitations
test command is:
cat ./test.html | htmldoc --webpage -t pdf - > test.pdf
test.html is attached
test.html.zip