HinTak/Font-Validator

Complete: myfont.report.xml

Opened this issue · 5 comments

I appreciate fontVal’s message that a given report is complete (https://github.com/HinTak/Font-Validator/blob/master/FontValidator/CmdLineInterface.cs#L80).

However (in my case), it would be more useful if the output would actually report the html file (not the .xml) – I could then open the file directly from the terminal. Am I missing something? Does anyone use the .xml file directly?

Thanks!

This is perhaps a somewhat oversight/historical omission. The GUI version on Windows uses MSIE. MSIE can combine xsl and xml on-the-fly. When it was ported to Linux/Mac with Mono or Wine, it was found that the emulated web engine, be it mozilla or webit, does not allow loading of xsl from disk, for security reasons. So there are some code in the GUI, conditional on Linux/Mono or Wine, to generate the html instead. I think that is also the default on Mac with the command line too, since Safari does not let you load xsl with xml either.

So which platform(s) are you on? I am just saying that generation of html is the default on non-windows, I think.

On both Linux and mac, xsltproc is available as standard. You can do `xsltproc --output file.html xslfile xmlfile . The xsl file is deposited next to the xml files.

I am using FontValidator as a command line tool, wrapped in a script file with other tools.
My environment is macOS, and I use iTerm (which provides clickable link-access to file names).

For some reason I assumed that the html file was peeking into the XML, but it turns out I actually only the HTML file is really needed in my case. My original remark was really just about the suffix in the output message – maybe other users could benefit from changing it, too. Realizing that the XML is not needed on my platform, I wonder why it is generated at all?

Xml is the 'structured' output, from which other formatted outputs can be derived, via a style sheet (xsl), through an XSL transformation. So you write a different style sheet to transform the same data to different formats - currently we have one for html (where rows / columns becomes html tables), and one for plain text (where all the structure indicators are removed, and replaced with just spaces and indentations).

I am sorry, my request is much simpler than that. So I wonder if we’re talking about the same thing here.
Would it be possible to print

Complete: myfont.otf.report.html
instead of
Complete: myfont.otf.report.xml
?

Thanks!