summaryrefslogtreecommitdiff
path: root/ntp.c
Commit message (Collapse)AuthorAgeFilesLines
* Have a common routine for converting dates and times to strings.Guy Harris2023-02-021-15/+6
| | | | | | | | | | | | | | | | | | | | | | | Have a routine that takes a buffer, a strftime format, and a struct tm * as arguments, and: * checks whether the struct tm * is null and, if so, returns a string indicating that the date and time couldn't be converted; * otherwise, passes it to strftime(), along with the buffer and the format argument and, if strftime() returns 0, meaning the string didn't fit into the buffer and thus that the buffer's contents are undefined, returns a string indicating that the date and time didn't fit into the buffer; * otherwise, returns a pointer to the buffer. Call that routine instead of directly calling strftime() in printers; that prevents printing a buffer with undefined data if the buffer isn't big enough for the string. Also, when generating file names using an strftime format, check the return value of strftime() to make sure the buffer didn't overflow.
* configure, CMakeLists.txt: don't check for snprintf().Guy Harris2023-01-301-2/+0
| | | | | | | | It's specified by the C90 standard (and, as I remember, by the C89 standard, although I no longer have my paper copy); no need to worry about ancient environments that lack it, and we have some cases where we call it in code not protected by #ifdef HAVE_STRFTIME/#endif and haven't seen any reports of problems.
* Extract NTP printingHerwin Weststrate2020-04-211-0/+85
This method is required to dissect some RADIUS attributes