Bug #128280 for Perl-Tidy: --add-semicolons vs. one-line blocks

So the problem is, how can we get perltidy to add semicolons into a one-line block when they are missing, without breaking the one-line block. The suggested modification would change how perltidy behaves on all existing scripts, so it cannot be made. The default behavior approximately follows the ‘perlstyle’ doc, which says “Semicolon omitted in "short" one-line BLOCK” where “short” is left undefined. A new flag which defines “short” would solve the problem. This seems simple but it isn’t completely straightforward, because "short" is probably best defined as number of tokens, or something similar, rather than as a line length. A simpler alternative would be to make it a binary all-or-none switch. I’m not sure if this is worth the programming effort, but I will consider it if there is enough interest. Meanwhile, a crude workaround for adding semicolons to one-line blocks, without breaking them, is cat infile > perltidy -l=10 -npro | perltidy –noadd-newlines –npro | perltidy >outfile where the 10 defines “short” as a total line length. This breaks the one-line blocks, adds the semicolons, and then reforms them and finally reformats. The -npro prevents reading any .perltidyrc file, and the –noadd-newlines reconstructs one-line blocks. This might have undesirable side-effects on a complete file, such as making more one-line blocks than desired, but it could be done as a macro within an editor and applied to selected sections of code.

Maybe just add an (yes, ugly) new flag: --add-semicolons-for-one-line-blocks .

version 20190601 added flag --one-line-block-semicolons=n (-olbs=n) to control semicolons in one-line blocks. The values of n are: n=0 means no semicolons termininating simple one-line blocks n=1 means stable; do not change from input file [DEFAULT and current] n=2 means always add semicolons in one-line blocks The current behavior corresponds to the default n=1.