Hi Deri, At 2026-02-03T16:37:34+0000, Deri wrote: > On Monday, 2 February 2026 17:48:11 GMT G. Branden Robinson wrote: > > > As you failed to generate a proper basic run, consideration of what > > > happens next is moot. > > > > I'm going to disregard the rest of your message, as you're having > > obvious trouble comprehending mine. > > > > Let me know when you make sense of it. > > Ah, the mists are clearing. My instructions were confusing, sorry.
Acknowledged. Properly supporting multiple configurations can be a real PITA. > "after make these fonts available..." > > Would have been much better expressed as > > "after running 'make' you should see these fonts" Font _descriptions_... Yeah, but see them _where_? In the build tree, or the installation directory? > (sometimes my v. slow typing makes me too succinct so my true meaning > is lost). Also there always lurks the devil of terminological precision. :) > Due to this confusion you showed me the contents of the devpdf install > directory after 'make install'. > > I'm very glad you did. My focus is on building the appropriate files > correctly, I assumed that the install phase would promote those built > objects to the proper install directory, so I don't test that as often > as I should. I would not have realised this is not what is happening > now. Where "now" = 1.24.0.rc2, I think that's a correct assessment. The install phase will not do things properly if we don't take explicit steps to ensure that it does. Automake handles a lot of stuff for us, but a device-independent troff's font management infrastructure is not among that stuff. > Using current master (same for rc2) in the basic mode (no gs command > and no separate urw fonts found) running 'make' generates these > files:- > > [derij@pip build (master)]$ ls font/devpdf > CB CI DESC enc/ Foundry HBI HR S symbolsl.afm TB TI util/ > CBI CR download EURO HB HI map/ SS symbolsl.pfb TBI TR ZD > > These are the correct files which should be promoted to the install directory > for basic mode. ...minus the "util" directory, which hasn't been installed since groff 1.22.3. A more salient point is that we see only font descriptions for the PDF Base 14 fonts. That seems okay to me. (I'd be okay with installing font descriptions for all PostScript level 2 Base 35 fonts at this "basic" support level, too, supposing that with Free Software versions of these being widely available, PDF readers may have them "built-in" even if the standard doesn't require them to do so.) Further, since that's what my working copy is doing, so I'd rather not perturb it unless someone feels strongly that we need to change it for "correctness". > [derij@pip build (master)]$ sudo make install > [...] > [derij@pip build (master)]$ ls /usr/local/share/groff/1.24.0/font/devpdf > AB BMB CB DESC Foundry HNB HR NI PI symbolsl.afm TI > ABI BMBI CBI download HB HNBI map/ NR PR symbolsl.pfb TR > AI BMI CI enc/ HBI HNI NB PB S TB ZCMI > AR BMR CR EURO HI HNR NBI PBI SS TBI ZD > > Exactly like you showed me when I was showing you font/devpdf. If I > now look again at the devpdf build directory:- > > [derij@pip build (master)]$ ls font/devpdf > AB BMB CB DESC Foundry HNB HR NI PI symbolsl.afm TI ZD > ABI BMBI CBI download HB HNBI map/ NR PR symbolsl.pfb TR > AI BMI CI enc/ HBI HNI NB PB S TB util/ > AR BMR CR EURO HI HNR NBI PBI SS TBI ZCMI > > These extra (non base-14) fonts have appeared in the build directory > and been installed. Is this normal make install behaviour. Not if you "make distclean" and re-"configure" after uninstalling whatever packages supply the URW Base 35 fonts on your system. :) > It has changed the files after all the groff pdf documents have been > produced by 'make' and then 'make check'ed. So 'make check' has been > run before the contents of font/devpdf has been altered. We rely on a stamp file, font/devpdf/stamp. Unfortunately, the stamp file is not made out of date by NOT removing files that it _no longer_ depends on. So a possibility exists for running the tests against a configuration that no longer reflects the state of the system. https://cgit.git.savannah.gnu.org/cgit/groff.git/tree/font/devpdf/devpdf.am?h=1.24.0.rc2#n244 Note that: 1. The stamp file depends on the contents of the dynamically populated `devpdffontdata` macro. 2. The stamp file will be out of date (because it's been deleted) if one cleans the build tree. That's good! > These extra fonts which have appeared must not be present if gropdf is > in basic mode, since any font beyond the base-14 has to be embedded > for it to be displayed properly by a viewer, You know available PDF readers better than I do. Do _none_ of them supply their own copies of the PostScript level 2 base 35? > and the fonts are not available without either ghostscript or a > separate urw font directory being located. > > In groff 1.23.0 both the build and install directory contained just > the base-14 fonts:- > > [derij@pip build ((1.23.0))]$ ls /usr/local/share/groff/1.23.0/font/devpdf > > CB CI DESC enc/ Foundry HBI HR S TBI TR > CBI CR download EURO HB HI map/ TB TI ZD Not true for me. My groff 1.23.0 was "configured" differently, with the base 35 fonts available. $ (cd ~/groff-1.23.0/share/groff/1.23.0/font/devpdf/ && echo * | fold -sw 72) AB ABI AI AR BMB BMBI BMI BMR CB CBI CI CR DESC EURO Foundry HB HBI HI HNB HNBI HNI HNR HR NB NBI NI NR PB PBI PI PR S TB TBI TI TR U-AB U-ABI U-AI U-AR U-BMB U-BMBI U-BMI U-BMR U-CB U-CBI U-CI U-CR U-HB U-HBI U-HI U-HNB U-HNBI U-HNI U-HNR U-HR U-NB U-NBI U-NI U-NR U-PB U-PBI U-PI U-PR U-S U-TB U-TBI U-TI U-TR U-ZCMI U-ZD ZCMI ZD download enc map Neither your directory contents nor mine are erroneus. The discrepancy is due to the activity of the "configure" script, which affects the installation payload. > [derij@pip build ((1.23.0))]$ l font/devpdf > CB CI DESC enc/ Foundry HBI HR S TB TI util/ > CBI CR download EURO HB HI map/ stamp TBI TR ZD > > I think the only changes to the devpdf directory since 1.23.0 was an > extra line in 'download' and the addition of the new font SS and its > corresponding .pfb file. Well, not really. When I diff my groff 1.23.0 installation with one from HEAD corresponding to my working copy, I see far more differences than that. $ diff -urN ~/groff-1.23.0/share/groff/1.23.0/font/devpdf ~/groff-HEAD/share/groff/1.24.0/font/devpdf | diffstat -w72 AB | 457 +-- ABI | 457 +-- AI | 457 +-- AR | 457 +-- BMB | 457 +-- BMBI | 457 +-- BMI | 457 +-- BMR | 457 +-- CB | 457 +-- CBI | 457 +-- CI | 457 +-- CR | 457 +-- DESC | 2 Foundry | 12 HB | 457 +-- HBI | 457 +-- HI | 457 +-- HNB | 457 +-- HNBI | 457 +-- HNI | 457 +-- HNR | 457 +-- HR | 457 +-- NB | 457 +-- NBI | 457 +-- NI | 457 +-- NR | 457 +-- PB | 457 +-- PBI | 457 +-- PI | 457 +-- PR | 457 +-- S | 316 +- SS | 54 TB | 457 +-- TBI | 457 +-- TI | 457 +-- TR | 457 +-- U-AB | 6407 ++++++++++++++++++++++++++++++++++----------- U-ABI | 6646 +++++++++++++++++++++++++++++++++++----------- U-AI | 6558 +++++++++++++++++++++++++++++++++++----------- U-AR | 6378 ++++++++++++++++++++++++++++++++++---------- U-BMB | 6507 +++++++++++++++++++++++++++++++++++---------- U-BMBI | 6556 +++++++++++++++++++++++++++++++++++----------- U-BMI | 6613 +++++++++++++++++++++++++++++++++++----------- U-BMR | 6591 +++++++++++++++++++++++++++++++++++----------- U-CB | 2470 ++++++----------- U-CBI | 2451 ++++++----------- U-CI | 2467 ++++++----------- U-CR | 2506 ++++++----------- U-HB | 6548 +++++++++++++++++++++++++++++++++++----------- U-HBI | 6614 +++++++++++++++++++++++++++++++++++----------- U-HI | 6665 +++++++++++++++++++++++++++++++++++------------ U-HNB | 6212 +++++++++++++++++++++++++++++++++---------- U-HNBI | 6509 ++++++++++++++++++++++++++++++++++----------- U-HNI | 6539 +++++++++++++++++++++++++++++++++++----------- U-HNR | 6210 +++++++++++++++++++++++++++++++++---------- U-HR | 6289 ++++++++++++++++++++++++++++++++++---------- U-NB | 6549 +++++++++++++++++++++++++++++++++++----------- U-NBI | 6388 ++++++++++++++++++++++++++++++++++----------- U-NI | 6546 +++++++++++++++++++++++++++++++++++----------- U-NR | 6593 +++++++++++++++++++++++++++++++++++----------- U-PB | 6524 +++++++++++++++++++++++++++++++++++----------- U-PBI | 6525 +++++++++++++++++++++++++++++++++++----------- U-PI | 6584 +++++++++++++++++++++++++++++++++++----------- U-PR | 6604 +++++++++++++++++++++++++++++++++++----------- U-S | 400 +- U-TB | 6613 +++++++++++++++++++++++++++++++++++----------- U-TBI | 6617 +++++++++++++++++++++++++++++++++++----------- U-TI | 6591 +++++++++++++++++++++++++++++++++++----------- U-TR | 6607 +++++++++++++++++++++++++++++++++++----------- U-ZCMI | 6501 +++++++++++++++++++++++++++++++++++---------- U-ZD | 422 +- ZCMI | 457 +-- ZD | 407 +- download | 141 map/dingbats.map | 207 + map/symbol.map | 9 map/symbolchars | 13 map/text.map | 56 symbolsl.afm | 51 symbolsl.pfb |binary 80 files changed, 156424 insertions(+), 59725 deletions(-) Howver, that looks much more dramatic than it is. 1. All of the font descriptions add a comment field that you wanted so that gropdf could infer a Unicode code point for each glyph. (Maybe for groff 1.25, I'd like to change that by extending the syntax of the fourth field of a glyph description, so that "36" becomes "36,u0024" or something like that. Parsing comments for semantic information is a fraught business.) This change accounts for...some massive percentage of the differences shown above. 2. The "DESC" file added the new "SS" font description to the default loading of mounting positions (as you noted). 3. Comments in files like "Foundry" saw (mostly cosmetic) updates. 4. Comments in the "U-" font descriptions for the URW foundry now disclose exactly how afmtodit(1) was run to produce them. 5. Comments in the "U-" font descriptions now reflect changes consequent to the newer(?)[1] version of the URW font files we're using. 6. The "SS" font description file is new, as you noted. 7. For me the "download" file is populated differently. Here's one example of many. - AvantGarde-Book /usr/share/fonts/type1/gsfonts/a010013l.pfb + AvantGarde-Book /usr/share/fonts/type1/urw-base35/URWGothic-Book.t1 This doesn't alarm me. This is host- and configuration-dependent stuff. 8. The dingbats.map file is much more populated. There's a Savannah ticket asking for this, and it was done. 9. The "symbolchars" file got deleted. Oddly, I can't find the commit that did this. Its contents seem redundant with "symbol.map", however. 10. Uppercase Greek letters were added to "text.map". 11. "symbolsl.afm" is new. This goes with the "SS" font description file and the "symbolsl.pfb" font file you noted. Regards, Branden [1] I'm utterly confused by this: -# FullName URW Gothic L Demi Oblique -# Version 1.06 -# FamilyName URW Gothic L +# FullName URW Gothic Demi Oblique +# Version 1.00 +# FamilyName URW Gothic ... -# Notice (Copyright (URW)++,Copyright 1999 by (URW)++ Design & Development; Cyrillic glyphs added by Valek Filippov (C) 2001-2005) -# Comment Generated by FontForge 20070723 -# Comment Creation Date: Thu Aug 2 15:52:13 2007 +# Notice (URW)++,Copyright 2014 by (URW)++ Design & Development +# Comment Copyright (URW)++,Copyright 2014 by (URW)++ Design & Development +# Comment Creation Date: 5/17/2017 So, uh... Version 1.06 is "(Copyright (URW)++,Copyright 1999 by (URW)++ Design & Development; Cyrillic glyphs added by Valek Filippov (C) 2001-2005)", and was produced by FontForge in 2007. ...but... Version 1.00 is Copyright 2014 by (URW)++ Design & Development and generated in 2017. Will URW's next release be numbered 0.7, maybe? ¯\_(ツ)_/¯ This change doesn't cause any problems that I've noticed, and as long as it does not, I'm content to leave font designers to munch their magic mushrooms in peace.
signature.asc
Description: PGP signature
