[gs-devel] build report for gs-9.55 candidate

Nelson H. F. Beebe beebe at math.utah.edu
Tue May 25 13:27:34 UTC 2021


I read with interest the long posting at

	https://ghostscript.com/pdfi.html

about the work on reimplementing PDF processing in C instead of
PostScript, and I thank the ghostscript developers for that clear
explanation of why this significant work is necessary.

I downloaded and successfully built and installed the candidate
release from

	https://github.com/ArtifexSoftware/ghostpdl-downloads/releases/tag/gpdf_alpha1

on CentOS 7 and 8, and Ubuntu 20.04, x86_64 systems.  All tests
passed.

I noticed two anomalies, however:

* After a "make" and "make check", a subsequent "make install"
  continues to compile many files.  In my view, this is seriously
  wrong. The "install" target should only be copying files into their
  target locations; otherwise, a compilation failure part way through
  could result in a seriously broken partly installed ghostscript,
  necessitating recovery of installation directories from backups.

* The "make install" steps fail to install the executables bin/gpdf
  and bin/gpdl.  I installed them manually on the build systems.

-------------------------------------------------------------------------------
- Nelson H. F. Beebe                    Tel: +1 801 581 5254                  -
- University of Utah                    FAX: +1 801 581 4148                  -
- Department of Mathematics, 110 LCB    Internet e-mail: beebe at math.utah.edu  -
- 155 S 1400 E RM 233                       beebe at acm.org  beebe at computer.org -
- Salt Lake City, UT 84112-0090, USA    URL: http://www.math.utah.edu/~beebe/ -
-------------------------------------------------------------------------------


More information about the gs-devel mailing list