summaryrefslogtreecommitdiff
path: root/ChangeLog
Commit message (Collapse)AuthorAgeFilesLines
* [docs]: Further clarify end-of-sentence detection.G. Branden Robinson2023-04-241-0/+10
| | | | | | | | | | | | | (One hopes.) * doc/groff.texi (Sentences): * man/roff.7 (Concepts): Do it. Fixes <https://savannah.gnu.org/bugs/?64091>. Thanks to Ingo Schwarze and Dave Kemper for the report. Also fix whitespace nit. Thanks to Dave Kemper, he of the unflinching osprey gaze.
* [docs]: Minimally document `tag`, `taga` requests.G. Branden Robinson2023-04-141-0/+9
| | | | | | | | | | | | | | | | | | | * doc/groff.texi (Postprocessor Access): * man/groff.7 (Request short reference): Do it. Fixes <https://savannah.gnu.org/bugs/?62695>. I don't know if we need to keep these. They do much the same thing as `device`, and are implemented nearly identically to each other. They have only one user, the also undocumented "devtag.tmac". One difference is that the `device` request constructs a "macro" (internally to GNU troff), and `tag{,a}` construct strings (from groff's string class, not the C++ standard library's). Possibly the type used for storage of the node contents is intended to address some of the same problems we've seen crop up in issues like Savannah #63074.
* [docs]: Fix typo.G. Branden Robinson2023-04-141-0/+10
|
* Move ChangeLog entry to correct file.G. Branden Robinson2023-04-141-10/+0
|
* ChangeLog: Annotate regression fix.G. Branden Robinson2023-04-011-0/+4
|
* [docs]: Fix Savannah #63812.G. Branden Robinson2023-03-221-0/+9
| | | | | | | | | | | * doc/groff.texi (Operators in Conditionals): * man/groff.7.in (Conditional expressions): Clarify how the output comparison operator is recognized. Fixes <https://savannah.gnu.org/bugs/?63812>. Thanks to John Gardner for the report. ANNOUNCE: Update bug counts.
* Revert "Make pdfmom more versatile."G. Branden Robinson2023-03-201-8/+0
| | | | | | | | | | This reverts commit 0290924f05f823039c546f5b14422af7eef70644. This has been separately committed on the post-1.23.0 branch so that it is easier for Bertrand to decide where to place a tag for 1.23.0.rc4 in this branch, or which commits to cherry-pick from that branch before doing so. (It is not yet decided what commits since 1.23.0.rc3 from this branch of post-1.23.0 will comprise 1.23.0.rc4--possibly all.)
* Make pdfmom more versatile.Deri James2023-03-091-0/+8
| | | | | | | * src/devices/gropdf/pdfmom.pl: Add flag -roff which severs hardcoded link to the mom macros. * src/devices/gropdf/pdfmom.pl: Document changes.
* Use a better type for symbol hashes.G. Branden Robinson2023-03-061-0/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | * bootstrap.conf: Add "stdint" module to ensure that the `uintptr_t` type is available. * src/include/symbol.h: Include <stdint.h> for `uintptr_t`. (class symbol): (symbol::hash): Change return type from `unsigned long`, which causes build failures on 64-bit MinGW, to `uintptr_t`. (symbol::hash): Use a C++ type cast, not a C-style one. Thanks to Bruno Haible for reporting the build failure in the 64-bit MinGW environment, and for suggesting a remedy. Also update editor aid comments; drop old style Emacs file-local variable setting. Any day you get to use the foot-Howitzer reinterpret_cast is a good one. Commits since 0221b657fe tested on: Debian bullseye Debian sid 2023-03-06 chroot macOS 12 OpenBSD 7.2 Solaris 10 Solaris 11
* [groff]: Revise a test to be more revealing.G. Branden Robinson2023-03-061-0/+21
| | | | | | | | | | | | | | | | | | | * src/roff/groff/tests/initialization_is_quiet.sh: Stop using "set -e". Instead use `fail` variable and `wail` function (and lowercase names for our internal variables) like many of our other tests. If the "unset" shell built-in fails, skip the test (prompted by /usr/xpg4/bin/sh on Solaris). Attempt every groff locale, with and without compatibility mode initially enabled, instead of stopping at the first failure. Report standard error and standard output content separately. Use groff's `-a` flag to prepare the standard output, for readability. * PROBLEMS: Document that this test might be skipped rather than failing on Solaris. (What actually happens depends on which shell you run it with, and we advise a variety of approaches.) Thanks to Bruno Haible for feedback regarding mysterious failures of this test on GNU/Hurd and NetBSD systems.
* bootstrap.conf: Add "stdbool-c99" gnulib module.G. Branden Robinson2023-03-061-0/+7
| | | | | | | * bootstrap.conf (gnulib_modules): Add "stdbool-c99" per recommendation from Bruno Haible. Fixes build problem on Solaris using Sun compiler.
* Makefile.am (AR): Remove hardcoded value.Bruno Haible2023-03-061-0/+8
| | | | | | | | * Makefile.am (AR): Remove hardcoded value. Let Automake use the value from config.status. [Fixes build problem on 64-bit AIX. Problem appears to date back to commit 5fec19d453, 2014-08-15. --GBR]
* [arch]: Rename target to fix non-GNU make failure.Bruno Haible2023-03-061-0/+6
| | | | | | * arch/misc/misc.am (shdeps.sed): Rename target from this... ($(SH_DEPS_SED_SCRIPT)): ...to this, to work better with make(1) on FreeBSD, NetBSD, and AIX.
* [build]: Discard now-unneeded Autoconf macro.G. Branden Robinson2023-03-021-0/+9
| | | | | | | | | | ...and variables. * configure.ac: Stop calling `GROFF_POPPLER`. Stop populating the Automake conditional `HAVE_PDFTOOLS`. Eliminate chatter about their availability in the configuration report. * m4/groff.m4 (GROFF_POPPLER): Delete.
* [mom]: Always generate test script.G. Branden Robinson2023-03-021-0/+10
| | | | | | | | | | | | | | | | | | | [mom]: Generate test script even if we'll skip it. * contrib/mom/examples/test-mom.sh.in: Check availability of required pdfinfo(1) and pdfimages(1) commands at test time; skip if they aren't present. * contrib/mom/mom.am [!HAVE_PDFTOOLS]: Generate the test script even if these tools aren't available. This approach also has the slight advantage that if you build groff, then install these tools after building (or even testing), you can simply run the tests again with "make check" without having to "make distclean" and start all over. As a developer, I find it confusing when the (total) quantity of test scripts bounces around from one host environment to another.
* m4/groff.m4 (GROFF_URW_FONTS_NOTICE): Clarify.G. Branden Robinson2023-03-011-0/+10
| | | | | | | | | | * m4/groff.m4 (GROFF_URW_FONTS_NOTICE): Adapt wording of notice to presence and identity of Ghostscript interpreter command. Clarify that it is gropdf(1) specifically that traverses the Ghostscript search path reported by its "-h" option. Continues <https://savannah.gnu.org/bugs/?63808>. Thanks to Deri James for the ongoing discussion.
* [devpdf]: Revise tests.G. Branden Robinson2023-02-251-0/+17
| | | | | | | | | | | | | | * font/devpdf/tests/check-default-foundry.sh.in: Test only the base 14 fonts of the PDF standard unconditionally. Test the remainder from the set of 35 commonly distributed only if a Ghostscript interpreter was detected at configuration time, because the latter fonts _must_ be embedded in PDF documents. If they're not present, skip the test rather than failing it. * font/devpdf/tests/check-urw-foundry.sh.in: Skip test if no URW fonts detected at configuration time, rather than failing it. * m4/groff.m4 (GROFF_GROPDF_PROGRAM_NOTICE, GROFF_URW_FONTS_NOTICE): Drop warnings of expected test failures. The tests no longer fail in the anticipated circumstances.
* [devpdf]: Generate tests from template files.G. Branden Robinson2023-02-251-0/+19
| | | | | | | | | | | | | | | | | | | [devpdf]: Generate tests from template files, so we can populate the test scripts with information determined at configuration. The default foundry test depends on $GHOSTSCRIPT, and the URW foundry test on $urwfontsdir. * font/devpdf/tests/check-default-foundry.sh: * font/devpdf/tests/check-urw-foundry.sh: Rename these... * font/devpdf/tests/check-default-foundry.sh.in: * font/devpdf/tests/check-urw-foundry.sh.in: ...to these. * font/devpdf/devpdf.am (font_devpdf_default_test) (font_devpdf_urw_test): New variables store names of generated test scripts. (font/devpdf/tests/check-default-foundry.sh): (font/devpdf/tests/check-urw-foundry.sh): New targets produce test scripts from corresponding .in files.
* [devpdf]: Trivially refactor.G. Branden Robinson2023-02-251-0/+11
| | | | | | | | | | Rename sed-substitutum [Lat.] from "@GROFF_GHOSTSCRIPT_INTERPRETERS@" to "@GHOSTSCRIPT@" for clarity and brevity; this is a scalar value containing the Autoconf-determined name of the Ghostscript interpreter. It is not the same as the replacement that occurs in contrib/pdfmark. * font/devpdf/devpdf.am: * font/devpdf/util/BuildFoundries.pl: Do it.
* [build]: Don't treat Ghostscript fonts as URW.G. Branden Robinson2023-02-251-0/+17
| | | | | | | | | | | | | | | | [build]: Stop scraping output of Ghostscript executable with "-h" option to attempt to find URW fonts. Fonts that ship with Ghostscript are regarded as the "default" foundry, not the URW foundry (though they often ultimately originate with URW fonts). They are often missing Adobe Font Metric (AFM) files, so it is impossible for groff to generate font description files for them at build time. * m4/groff.m4 (GROFF_URW_FONTS_CHECK): Drop `AC_REQUIRE` on `GROFF_AWK_PATH`. Drop awk-based scraping of Ghostscript "-h" output. Annotate need for sync between this list of characteristic font file names and the one in BuildFoundries. Thanks to Deri James for the ongoing discussions.
* configure.ac: Drop now-redundant explicit check.G. Branden Robinson2023-02-251-0/+6
| | | | | | * configure.ac: Drop now-redundant explicit `GROFF_URW_FONTS_CHECK`. `GROFF_GROPDF_DEPENDENCIES_CHECK` `AC_REQUIRE`s it as of commit ec001d2a23, 18 February.
* m4/groff.m4: Improve URW font check.G. Branden Robinson2023-02-251-0/+9
| | | | | | | | * m4/groff.m4 (GROFF_URW_FONTS_CHECK): When looking for a characteristic URW font by its file name, also check for "URWGothic-Book" with no file extension. Avoids false negatives in URW Type 1 font detection when using Ghostscript 9.53.3. Thanks to Deri James for suggesting the test procedure that uncovered this flaw.
* [doc]: Uninstall groff Texinfo in HTML better.G. Branden Robinson2023-02-221-0/+8
| | | | | | | * Makefile.am (uninstall_groffdirs): Remove "html.mono" and "html.node" directories corresponding to HTML version of our Texinfo manual. * doc/doc.am (uninstall-html): Uninstall HTML version of our Texinfo manual more reliably.
* [doc]: Handle Texinfo output formats consistently.G. Branden Robinson2023-02-221-0/+18
| | | | | | | | | | | | | | | | | | | | [doc]: Handle output formats of our Texinfo manual more consistently. * doc/doc.am (install-doc): Add dependency on (Automake standard) target 'install-dvi'. (maintainer-clean-local): Remove our Texinfo manual in plain text format. (install-data-local): Add dependency on new target 'install-txt'. (install-txt): Install our Texinfo manual in plain text format. (uninstall-local): Add dependency on new target 'uninstall-txt'. (uninstall-txt): Uninstall our Texinfo manual in plain text format. Thanks to T. Kurt Bond for noticing the discrepancy.
* doc/doc.am: Fix install-{pdf,html}-local targets.G. Branden Robinson2023-02-221-0/+5
| | | | | | | | * doc/doc.am (install-pdf-local, install-html-local): Enable rules to work in out-of-tree builds. These rules aren't made by default (and so weren't causing failures); that's a separate issue.
* doc/doc.am: Drop redundant dependency.G. Branden Robinson2023-02-221-0/+7
| | | | | | * doc/doc.am (uninstall-hook): Drop dependency on "uninstall_mom"; this is not the appropriate place to declare it, and "uninstall_groffdirs" already depends on it in mom's Automake file.
* [tests]: Have more tests report output.G. Branden Robinson2023-02-221-0/+12
| | | | | | | | | | * src/roff/groff/tests/ab_works.sh: * src/roff/groff/tests/handle_special_input_code_points.sh: * src/roff/groff/tests/initialization_is_quiet.sh: * src/roff/groff/tests/msoquiet_works.sh: * src/roff/groff/tests/soquiet_works.sh: * tmac/tests/an-ext_MR-works.sh: * tmac/tests/an_MR-works.sh: Do it.
* [tests]: Avoid failures when $GROFF_ENCODING set.G. Branden Robinson2023-02-221-0/+17
| | | | | | | | | | | | | | | * src/roff/groff/tests/ab_works.sh: * src/roff/groff/tests/handle_special_input_code_points.sh: * src/roff/groff/tests/initialization_is_quiet.sh: * src/roff/groff/tests/msoquiet_works.sh: * src/roff/groff/tests/soquiet_works.sh: * tmac/tests/an-ext_MR-works.sh: * tmac/tests/an_MR-works.sh: * tmac/tests/an_font-remapping-does-not-affect-titles.sh: Unset $GROFF_ENCODING before running test because preconv(1) confounds these tests. Thanks to Alexis for reporting this problem.
* [ms]: Rename test.G. Branden Robinson2023-02-221-0/+8
| | | | | | | | | | | | | | * tmac/tests/s_no-excess-space-around-displays.tmac: Rename... * tmac/tests/s_no-excess-space-around-displays.sh: ...to this. * tmac/tmac.am (tmac_TESTS): Update. The original name of this test was a thoughtless error; renaming it more intelligibly facilitates a necessary procedure for getting the tests to run on Solaris 10 (where /bin/sh is a pre-POSIX clay tablet version). gsed -i -e '1s@#!/bin/sh@#!/usr/xpg4/bin/sh@' \ `find . -name '*.sh' | grep /tests/` PATH=/usr/xpg4/bin:$PATH gmake check
* [gropdf]: Fix Savannah #63824 (2/2).G. Branden Robinson2023-02-221-0/+15
| | | | | | | | | | | | | | | | | [gropdf]: Revise tests to run unconditionally, rather than configuring them away at build time, which can cause a distribution archive to be incorrectly structured. Update configuration notices when optional dependencies are absent. * font/devpdf/devpdf.am (font_devpdf_TESTS): Remove `USE_GROPDF` and `HAVE_URW_FONTS` conditionals. * m4/groff.m4 (GROFF_GROPDF_PROGRAM_NOTICE, GROFF_URW_FONTS_NOTICE): Warn reader that a gropdf test failure is to be expected. Fixes <https://savannah.gnu.org/bugs/?63824> (2/2). ANNOUNCE: Update bug counts.
* [grohtml]: Fix Savannah #63824 (1/2).G. Branden Robinson2023-02-221-0/+18
| | | | | | | | | | | | | | | | | [grohtml]: Revise tests to check for requisite programs at test time and skip if not found, rather than configuring them away at build time, which can cause a distribution archive to be incorrectly structured. * m4/groff.m4: Annotate requisite program list since we're violating the DRY principle. * src/roff/groff/groff.am (groff_TESTS): Populate unconditionally. * src/roff/groff/tests/html_works_with_grn_and_eqn.sh: * src/roff/groff/tests/smoke-test_html_device.sh: Check for requisite programs and skip test if any are not found. Fixes <https://savannah.gnu.org/bugs/?63824> (1/2).
* [man pages]: Define page-local `MR` fallback.G. Branden Robinson2023-02-221-0/+79
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | [man pages]: Provide page-local fallback definition of new `MR` macro. I didn't think I was going to have to do this, but the premier site on the web for viewing Linux man pages, Michael Kerrisk's man7.org, has been pulling snapshots of the pages themselves without upgrading the underlying macros, and so man page cross references set with `MR` are getting lost in its presentations. (I acknowledge: Ingo Schwarze warned me something like this could happen.) This definition is intended as a stopgap measure only. I want to revert this after groff 1.23 is released and has spread to some reasonable degree. * contrib/chem/chem.1.man: * contrib/eqn2graph/eqn2graph.1.man: * contrib/gdiffmk/gdiffmk.1.man: * contrib/glilypond/glilypond.1.man: * contrib/gperl/gperl.1.man: * contrib/gpinyin/gpinyin.1.man: * contrib/grap2graph/grap2graph.1.man: * contrib/hdtbl/groff_hdtbl.7.man: * contrib/mm/groff_mm.7.man: * contrib/mm/groff_mmse.7.man: * contrib/mm/mmroff.1.man: * contrib/mom/groff_mom.7.man: * contrib/pdfmark/pdfroff.1.man: * contrib/pic2graph/pic2graph.1.man: * contrib/rfc1345/groff_rfc1345.7.man: * man/groff.7.man: * man/groff_char.7.man: * man/groff_diff.7.man: * man/groff_font.5.man: * man/groff_out.5.man: * man/groff_tmac.5.man: * man/roff.7.man: * src/devices/grodvi/grodvi.1.man: * src/devices/grohtml/grohtml.1.man: * src/devices/grolbp/grolbp.1.man: * src/devices/grolj4/grolj4.1.man: * src/devices/gropdf/gropdf.1.man: * src/devices/gropdf/pdfmom.1.man: * src/devices/grops/grops.1.man: * src/devices/grotty/grotty.1.man: * src/devices/xditview/gxditview.1.man: * src/preproc/eqn/eqn.1.man: * src/preproc/eqn/neqn.1.man: * src/preproc/grn/grn.1.man: * src/preproc/pic/pic.1.man: * src/preproc/preconv/preconv.1.man: * src/preproc/refer/refer.1.man: * src/preproc/soelim/soelim.1.man: * src/preproc/tbl/tbl.1.man: * src/roff/groff/groff.1.man: * src/roff/nroff/nroff.1.man: * src/roff/troff/troff.1.man: * src/utils/addftinfo/addftinfo.1.man: * src/utils/afmtodit/afmtodit.1.man: * src/utils/grog/grog.1.man: * src/utils/hpftodit/hpftodit.1.man: * src/utils/indxbib/indxbib.1.man: * src/utils/lkbib/lkbib.1.man: * src/utils/lookbib/lookbib.1.man: * src/utils/pfbtops/pfbtops.1.man: * src/utils/tfmtodit/tfmtodit.1.man: * src/utils/xtotroff/xtotroff.1.man: * tmac/groff_man.7.man.in: * tmac/groff_me.7.man: * tmac/groff_ms.7.man: * tmac/groff_trace.7.man: * tmac/groff_www.7.man: Do it. Fixes <https://savannah.gnu.org/bugs/?63825>. Thanks greatly to Alexis for identifying an issue with mandoc(1)'s handling of an earlier iteration of this fallback, and to John Gardner for identifying a method of detecting mandoc as the renderer at formatting time. Tested with groff 1.22.4, mandoc 1.14.5, and groff Git HEAD (of course). The sed script I used follows. /^\.cp 0/a\ .\ .\\" Define fallback for groff 1.23's MR macro if the system lacks it.\ .nr do-fallback 0\ .if !\\n(.f .nr do-fallback 1 \\" mandoc\ .if \\n(.g .if !d MR .nr do-fallback 1 \\" older groff\ .if !\\n(.g .nr do-fallback 1 \\" non-groff *roff\ .if \\n[do-fallback] \\{\\\ . de MR\ . ie \\\\n(.$=1 \\\ . I \\%\\\\$1\ . el \\\ . IR \\%\\\\$1 (\\\\$2)\\\\$3\ . .\ .\\}\ .rr do-fallback ANNOUNCE: Update bug counts. Credit Alexis with assistance.
* Revert "[man pages]: Define page-local `MR` fallback."G. Branden Robinson2023-02-211-71/+0
| | | | | | | | | | | This reverts commit 18d708e489758636ff9e168eee2592591755eb61. Unfortunately my clever hack [sic] does not work with mandoc 1.14.5, causing that tool to truncate the entire text of the man page after the summary description. The good news is that there is a simpler replacement that works better, forthcoming.
* ChangeLog: Fix wording goof.G. Branden Robinson2023-02-201-1/+1
|
* [man]: Improve portability of a test.1.23.0.rc3G. Branden Robinson2023-02-201-0/+8
| | | | | | | | | * tmac/tests/an_TS-do-not-keep-tables-when-cR-set.sh: Improve portability of script by using shell constructs instead of the nonstandard GNU coreutils 'seq' utility. Resolves test failure on Solaris 10. * HACKING: Document this problem.
* m4/groff.m4 (GROFF_MAKE_DEFINES_RM): Honor $MAKE.G. Branden Robinson2023-02-201-0/+7
| | | | | | | * m4/groff.m4 (GROFF_MAKE_DEFINES_RM): Test the make(1) in the environment variable $MAKE if defined, instead a literal "make". Required on (some) Solaris 10 configurations where traditional make(1) is not installed but GNU make is installed as "gmake".
* tmac/tests/latin2_works.sh: Fix missing backslash.G. Branden Robinson2023-02-191-0/+6
| | | | | | * tmac/tests/latin2_works.sh: Fix missing backslash in printf. Solaris printf(1) was sensitive to this error; macOS and GNU printf were not. Fixes a test failure on Solaris 11.
* [groff]: Improve portability of a test.G. Branden Robinson2023-02-191-0/+9
| | | | | | | | | * src/roff/groff/tests/regression_savannah_58153.sh: Improve portability. Avoid the unpredictability of implementations when putting backslashes inside a groff-piped printf shell command inside a here document inside a command substitution by changing the groff escape character to something meaningless to the shell and to printf ('@'). Fixes a test failure on Solaris 11.
* Revert "[gropdf]: Bifurcate test of font availability."G. Branden Robinson2023-02-191-15/+0
| | | | | | | | | | | This reverts commit 6e001a7f397438d7f09e97155b3c1ec89c1fc855. Per Deri James, having a Ghostscript executable installed but not a full set of PostScript level 2 font files installed (the "base 35" plus groff's "EURO") in discoverable places is _not_ a supported groff configuration scenario as far as gropdf is concerned. A test failure is therefore expected in that case.
* [gropdf]: Bifurcate test of font availability.G. Branden Robinson2023-02-181-0/+15
| | | | | | | | | | | | | | * font/devpdf/tests/check-default-foundry.sh: Bifurcate test of font availability. The Ghostscript executable can be present in the host environment without any Type 1 fonts. The PDF standard requires that its base 14 be present in the renderer, so groff font descriptions for those 14 fonts (plus groff's "EURO" font) should always be available. But the remaining fonts from the "Base 35" PostScript level 2 set won't necessarily be, and so their font descriptions won't be, either. Skip the test rather than failing it in that case, because the Ghostscript executable can have any number of fonts built-in (in "%rom%" as "gs -h" puts it). We'll need a more sophisticated Autoconf test to discern this scenario from a true failure case.
* [gropdf]: Recognize more URW file names.G. Branden Robinson2023-02-181-0/+7
| | | | | | * font/devpdf/Foundry.in: Recognize URW foundry replacements for Helvetica Bold-Oblique and Helvetica Oblique under the file names "NimbusSans-BoldItalic.t1" and "NimbusSans-Italic.t1", respectively.
* [build]: Uninstall more fastidiously.G. Branden Robinson2023-02-181-0/+9
| | | | | | | | | | | | Try to remove the configured PDF documentation directory in the event it is empty, but do not fail if it isn't. (It can be a directory shared with other groff components; we don't know in what order the uninstall targets will serialize, but the last one run should succeed.) * contrib/mom/mom.am (uninstall_mom): * contrib/pdfmark/pdfmark.am (uninstall-pdfmark-hook): * contrib/sboxes/sboxes.am (uninstall_sboxes): * doc/doc.am (uninstall-pdf): Do it.
* [build]: Detangle pdfroff, gropdf config (2/2).G. Branden Robinson2023-02-181-0/+20
| | | | | | | | | | | | | | | | | | | | | | * m4/groff.m4: Give pdfroff its own Autoconf macros to handle dependency checking, build objectives, and user notice. (GROFF_PDFROFF_DEPENDENCIES_CHECK): New macro requires `GROFF_AWK_PATH` and `GROFF_GHOSTSCRIPT_PATH`, determines whether pdfroff can be used at build time, and (if not) constructs part of message to be shown to user explaining why. (GROFF_PDFROFF_PROGRAM_NOTICE): New macro requires `GROFF_PDFROFF_DEPENDENCIES_CHECK` and emits message if needed. (GROFF_GHOSTSCRIPT_AVAILABILITY_NOTICE): Drop mention of impact on pdfroff since its dedicated notice covers this now. * configure.ac: Call the new macros at appropriate times. Produce a new Automake macro, `USE_PDFROFF`, to replace inapposite use of `USE_GROPDF` in pdfmark.am. * contrib/pdfmark/pdfmark.am: Use `USE_PDFROFF` instead of `USE_GROPDF`.
* [build]: Detangle pdfroff, gropdf config (1/2).G. Branden Robinson2023-02-181-0/+20
| | | | | | | | | | | | | | | | | | * m4/groff.m4 (GROFF_GROPDF_DEPENDENCIES_CHECK): Add `AC_REQUIRE`ment on `GROFF_URW_FONTS_CHECK`. Per discussion with Deri James, if _either_ Ghostscript or the URW fonts are avilable, gropdf will be fully functional, consequent to commit d5515, 22 June. Drop dependency on awk; it is used only at configuration time (in an Autoconf macro) and since Savannah #62775 was resolved (19 September), it has not been strictly necessary. (It is useful for searching more locations for URW fonts, but several others are searched even if it is absent.) Stop populating notice text here, instead moving it... (GROFF_GROPDF_PROGRAM_NOTICE): ...here, since it no longer needs to be dynamically constructed. Continues <https://savannah.gnu.org/bugs/?63808>. Thanks to Deri James for the continued discussion.
* [build]: Rename GROFF_CHECK_GROPDF_PROGRAMS macro.G. Branden Robinson2023-02-181-0/+9
| | | | | | | | | [build]: Rename `GROFF_CHECK_GROPDF_PROGRAMS` macro to `GROFF_GROPDF_DEPENDENCIES_CHECK` to generalize for greater accuracy in forthcoming change. * configure.ac: * m4/groff.m4: Do it.
* [gropdf]: Don't test if hobbled.G. Branden Robinson2023-02-181-0/+16
| | | | | | | | | | | | | | | | | [gropdf]: Don't run automated tests if 'gropdf' will be operating with reduced function. If neither Ghostscript nor the URW fonts are available at configuration time, there is no point testing for successful build-time population of the font descriptions for the default and URW foundries. * font/devpdf/devpdf.am (font_devpdf_TESTS) [USE_GROPDF]: Run "check-default-foundry" only if gropdf is fully functional. (font_devpdf_TESTS) [USE_GROPDF && HAVE_URW_FONTS]: Run "check-urw-foundry" only if the URW fonts were found. Continues <https://savannah.gnu.org/bugs/?63808>. Thanks to Deri James for the continued discussion.
* [gropdf]: Rename tests.G. Branden Robinson2023-02-181-0/+11
| | | | | | | | | | | | ...to more accurately characterize their purpose. * font/devpdf/tests/basic-fonts-present.sh: * font/devpdf/tests/urw-fonts-present.sh: Rename these... * font/devpdf/tests/check-default-foundry.sh: * font/devpdf/tests/check-urw-foundry.sh: ...to these. * font/devpdf/devpdf.am (font_devpdf_TESTS): Reflect rename.
* [gropdf]: Revise tests to be foundry-focussed.G. Branden Robinson2023-02-181-0/+16
| | | | | | | | | | | | | | * font/devpdf/tests/basic-fonts-present.sh: Stop trying to match font descriptions in the "devps" directory with ones in "devpdf"; instead, test whether "BuildFoundries" did its job. Test for font descriptions corresponding to the full 35 PostScript Level 2 font repertoire, plus groff's "EURO". * font/devpdf/tests/urw-fonts-present.sh: Drop stale comment and rename variable for better parallelism with the other test above. Continues <https://savannah.gnu.org/bugs/?63808>. Thanks to Deri James for the continued discussion.
* [man]: Tweak fix to Savannah #63768.G. Branden Robinson2023-02-181-0/+14
| | | | | | | | | | | | * tmac/an.tmac (MR): Ensure `an*url` always has a value, defaulting to "format 1" (man:page(section)). Resequence the macOS URL formats to sort the contemporary one before the others. * tmac/man.local: Reflect resequencing of integer assignments to formats. Clarify historicity of annotations. Continues <https://savannah.gnu.org/bugs/?63768>. Thanks to John Gardner for further discussion.
* [ms]: Port a test to Solaris 11 sed.G. Branden Robinson2023-02-171-0/+9
| | | | | | | | * tmac/tests/s_TC-works-with-percent-in-custom-titles.sh: Put newlines after opening braces in sed scripts. macOS and GNU sed tolerate their absence, but this sed does not. * HACKING: Document this problem.
View Lorry Controller Status