Add input validation to each symbology

bwipp / postscriptbarcode

Barcode Writer in Pure PostScript

https://bwipp.terryburton.co.uk

MIT License

462 stars 64 forks source link

Add input validation to each symbology #17

Closed terryburton closed 1 year ago

terryburton commented 10 years ago

Add proper input validation with detailed error messages to all symbologies:

[x] auspost
[x] azteccode
[x] codablockf
[x] code128
[x] code16k
[x] code49
[x] code93
[x] code93ext
[x] codeone
[x] daft
[x] databarexpanded
[x] databarexpandedcomposite
[x] databarexpandedstacked
[x] databarexpandedstackedcomposite
[x] databarlimitedcomposite
[x] databaromni
[x] databaromnicomposite
[x] databarstackedcomposite
[x] databarstackedomnicomposite
[x] databartruncatedcomposite
[x] dotcode
[x] ean13composite
[x] ean8composite
[x] gs1-128
[x] gs1-128composite
[x] gs1-cc
[x] gs1datamatrix
[x] hanxin
[x] japanpost
[x] maxicode
[x] micropdf417
[x] msi
[x] onecode
[x] pdf417
[x] posicode
[x] qrcode
[x] telepen
[x] upcacomposite
[x] upcecomposite

terryburton commented 1 year ago

@gitlost Hi. From now on, anything that causes the PS interpreter to fail without calling the BWIPP raiseerror routine should be considered to be a bug.

gitlost commented 1 year ago

Hi, thanks, just a few things I noticed on running the Zint tests with BWIPP enabled:

[x] in msi, the check when badmod11 is set should allow any checktype that contains a mod11 I think, i.e. mod11 or ncrmod11 or mod1110 or ncrmod1110
[x] in code49, the option concat clashes with the built-in concat operator - maybe rename to append?
[x] in pdf417, the check of eclevel should be from 0 to 8, not 1 to 5
[x] in datamatrix, the doly/dmre logic needs to be done in the second metrics loop also

If you like, I can do PRs for these....

terryburton commented 1 year ago

in msi, the check when badmod11 is set should allow any checktype that contains a mod11 I think, i.e. mod11 or ncrmod11 or mod1110 or ncrmod1110

I think you're right.

in code49, the option concat clashes with the built-in concat operator - maybe rename to append?

Agreed. Note the workaround I put in BWIPP for this, but I agree we should rename it to append if it causes trouble downstream.

in pdf417, the check of eclevel should be from 0 to 8, not 1 to 5

Doh.

in datamatrix, the doly/dmre logic needs to be done in the second metrics loop also

If you like, I can do PRs for these....

That would be excellent, thank you.

gitlost commented 1 year ago

Cool, will do.