Follow-up Comment #5, bug #67939 (group groff):

[comment #1 comment #1:]
> Even if this change is conceptually a good idea (I can see
> arguments both for and against),

Not to be coy...
Pros:
* eliminates gs dependency for this test
* as reported, significantly cuts run time

The con is largely theoretical: The "%%Pages:" line is not required in a valid
PostScript document.

$ echo 'Hello world!' | groff | grep -v '^%%Pages: ' > still_works.ps

Load still_works.ps into your favorite PostScript viewer to verify removing
that line has no effect on the output.  Indeed, because PostScript
interpreters ignore % lines, the %%Pages line can lie, which will fool grep
but not gs.

$ echo 'Hello world!' | groff | sed 's/^%%Pages: 1/%%Pages: 666/' | grep
'^%%Pages: ' | cut -d\  -f2
666
$ echo 'Hello world!' | groff | sed 's/^%%Pages: 1/%%Pages: 666/' | gs -o
/dev/null/ -sDEVICE=bbox - 2>&1 | grep HiResBoundingBox | wc -l
1

However, we're testing groff output, and we can expect groff to output
PostScript that follows certain conventions, such as a %%Pages line with an
accurate count.  So I don't see any real-world downside to this change; Bjarni
just needs to iron out the implementation.


    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?67939>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/

Attachment: signature.asc
Description: PGP signature

Reply via email to