Hi, the error message the test matches on occurs only once in the GNU datamsh sources. That same line has resulted in a build failure on Fedora for the armv7hl platform[1], which is 32bit. I have suggested two possible fixes[2], but cannot test them, since I do not have an ARM system for build tests available. Those compile fixes just use the correct format specifier for the variable holding the line number.
This might fix the tests as well, but I don't know this. Perhaps you would like to try this. Of course, datamash 1.7 is quite old and thus the patches agains current git might not apply cleanly. HTH, Erik [1] https://lists.gnu.org/archive/html/bug-datamash/2022-01/msg00003.html [2] https://lists.gnu.org/archive/html/bug-datamash/2022-01/msg00004.html -- The most effective debugging tool is still careful thought, coupled with judiciously placed print statements. -- Brian W. Kernighan