Closed craigsapp closed 5 months ago
I will resolve this in April/May, when I am planning on finishing the corpus, unless you need a change to the server copy before then.
No hurry.
The problem was that the awk script that I wrote some time back to find spacing errors didn't take record type into account (e.g., reference record, interpretation, etc.). So it couldn't handle the use case where you have spaces on an interpretation record because I ignored spaces in the tab-spaced columns due to the complexity of nested spaces in stops. I have written a more robust Perl program that handles this. I then wrapped that in another Perl program that filters out files that don't actually have spacing errors.
For the main branch, this resulted in the following files being found to have errors:
For the editing branch, this resulted in the following files being found to have errors:
On main, the sole error in scriabin-op08_no03.krn
is spaces on an interpretation record:
For the editing branch, the error in scriabin-op08_no03.krn
is the same.
For scriabin-op07_no01.krn
these two records have too many tabs:
Error on main fixed in 5eac98e7b0b482a0d03afc5b3fab486dc65fe0fc
Errors on editing branch fixed in ce277d74185afe788b2a176850ef4d12755008d2
In scriabin-op08-n03.krn, there it a space instead of a tab:
(The gray box highlights where the space should be converted into a tab).