On 2/9/21 4:15 PM, Patrice Dumas wrote:
in general, nothing is ignored at the parsing stage. That way the
original document can be more easily regenerated from a format without
information loss such as TexinfoXML, and also because in some situations
(@verbatim, @verb, @html, @example...) we want to keep all the spaces as
is.
Actually, I don't think it is correct to preserve invisible white-space at
end-of-line, even in verbatim modes. A texinfo file should be viewed a text
file.
Distinctions between different kinds of line-ending is not part of the
"information set" of a text file, and neither is white-space at end-of-line:
Consider a deck of punch cards (which some of us are old enough to have used for
writing programs), or OCR from a printed page.
Of course this is bikeshedding: I don't care enough to submit a patch or
argue strongly. But in general a text-file processor (such as a compiler)
which treats different line endings (including end-of-line whitespace)
differently
is IMO doing it wrong. (Assuming no backward-compatibility issues, of course.)
--
--Per Bothner
p...@bothner.com http://per.bothner.com/
