When running test vectors from the command line, it would be really useful if any test errors reported also showed the set and sequence numbers (if applicable), or failing that, at least the line number in the test vector file. Or maybe both. This would make it much easier to locate which vector an error refers to.
To give an example:
I have a large spreadsheet of test vectors, and export it to a CSV file to run those tests. The spreadsheet (and resultant CSV) has many blank lines and lines with comments. Therefore there is no 1-to-1 correlation between vector numbers and line number in the CSV file.
So if I get a test error like so:
Error on test vector 75:
IBF_SETn = 0 (expected 1)
It's hard to locate which vector this actually is, because it isn't on line 75, it's actually on line 92 of my file! The test vector number (75) is completely useless information in this scenario.
If it instead showed something like this:
Error on test vector 75 (set 8, sequence 19, line 92):
IBF_SETn = 0 (expected 1)
Things would be much more convenient.
As things are, I am forced to load up my test vector file in the GUI, run them, then scroll around and look for the failures.
When running test vectors from the command line, it would be really useful if any test errors reported also showed the set and sequence numbers (if applicable), or failing that, at least the line number in the test vector file. Or maybe both. This would make it much easier to locate which vector an error refers to.
To give an example:
I have a large spreadsheet of test vectors, and export it to a CSV file to run those tests. The spreadsheet (and resultant CSV) has many blank lines and lines with comments. Therefore there is no 1-to-1 correlation between vector numbers and line number in the CSV file.
So if I get a test error like so:
It's hard to locate which vector this actually is, because it isn't on line 75, it's actually on line 92 of my file! The test vector number (75) is completely useless information in this scenario.
If it instead showed something like this:
Things would be much more convenient.
As things are, I am forced to load up my test vector file in the GUI, run them, then scroll around and look for the failures.