# Ann: SWI-Prolog 10.1.6 **URL:** \ **Category:** Releases\ **Created:** [April 22, 2026, 7:32pm UTC](https://swi-prolog.discourse.group/t/ann-swi-prolog-10-1-6/9577 "2026-04-22T19:32:59Z")\ **Posts on this page:** 8\ **Page:** 1
**Author:** ![jan](https://yyz2.discourse-cdn.com/free1/user_avatar/swi-prolog.discourse.group/jan/32/4_2.png) [@jan](https://swi-prolog.discourse.group/u/jan)\ **Post date:** [April 22, 2026, 7:32pm UTC](https://swi-prolog.discourse.group/t/ann-swi-prolog-10-1-6/9577/1 "2026-04-22T19:32:59Z")
Dear SWI-Prolog user, SWI-Prolog 10.1.6 is ready for download. The main theme of this release is improved Unicode support, from the stream layer all the way up to the terminal. Highlights: **Building!** - There is a **new dependency**. Our unmaintained copy of `utf8proc.c`, underlying `library(unicode)` has been removed. All major platforms now support this package and thus we use the platform version. Install using **Core** - Unicode-aware stream position. `Supdatepos()` now counts East-Asian wide characters (CJK, emoji) as two columns and combining marks as zero, using `wcwidth()` (with a bundled `mk_wcwidth.c` fallback on platforms that lack it). `format/2`’s tab-stop arithmetic and the new `write_size/4` predicate inherit the correct column widths. - New `write_size(+Term, -Width, -Height, +Options)` replaces the now-deprecated `write_length/3`. It reports both the longest output line and the number of lines a term would occupy when printed, respects Unicode column widths, and accepts `max_width(+N)` / `max_height(+N)` options that bound the runtime on huge terms by aborting the write as soon as a threshold would be exceeded. - `code_type(Code, width(W))` exposes `wcwidth()` at the Prolog level. **Library(unicode) / package utf8proc** - The package no longer vendors its own copy of `utf8proc.c`; it now depends on the external `libutf8proc` shipped with every major distro (and built as a shared library in the flatpak and MinGW images). Distributions that ship an older `libutf8proc` still compile and run: feature probes in the package’s CMakeLists.txt gate optional fields and functions behind `HAVE_UTF8PROC_*` macros. - `unicode_property/2` grows `width`, `ambiguous_width`, `bidi_mirrored`, `uppercase`/`lowercase`/`titlecase`, and `indic_conjunct_break` properties; drops the rarely useful `comp_exclusion` and `control_boundary`. - New predicates `atom_graphemes/2`, `string_graphemes/2`, `unicode_version/1`, `unicode_codepoint_valid/1`, `unicode_casefold/2`, and `unicode_nfkc_casefold/2`. - Full PlDoc rewrite and a regression suite (80 plunit tests). **Terminal (XPCE / libedit)** - A unified Unicode-aware cell model in the xpce terminal, so grapheme clusters (NFD letters, emoji ZWJ sequences, regional-indicator flags) render and navigate as single user-perceived characters. Many related fixes: resize reflow, caret placement across combining marks and wide characters, DCH/ICH column arithmetic, and rendering of long NFD+emoji runs. - libedit uses bracketed-paste mode — pasting long text no longer triggers TAB completion, and is considerably faster on Windows. - On Windows, non-BMP code points (emoji, supplementary-plane CJK) now survive the UTF-16 round trip through `ReadConsoleInput`, `el_wgetc`, and `el_gets` unharmed. - `SIGWINCH` now correctly rewraps the current input line at the new column count, in both xterm-style and xpce terminals. **XPCE, modernisation** - The SDL3 backend continues to replace X11 bindings. This release removes a large set of X-only methods (`frame->hide`, `->grab_pointer`, `->open`, …); adds SDL3 implementations for `image<-rotate`, `image<-scale`, `window->flash`, reading/saving images from/to streams, SVG loading, and more; and ships SVG icons for most development tools. **Assorted fixes** - Read buffer reallocation bug in `'..` numeric literals (#1483). - GC crash when a nondeterministic foreign predicate triggered a constraint wakeup. - `spy/1` again enables debug mode. - `listing/2`: missing declaration of the `source(+Boolean)` option. - WASM class/property name conversion under Unicode. Enjoy — Jan # SWI-Prolog Changelog since version 10.1.5 - CI: Added utf8proc as dependency - ENHANCED: mkchangelog stops changelog entries at the first blank line A blank line in a commit body now terminates that commit’s changelog entry. Subject + first body paragraph become the entry; deeper detail stays visible in `git log` but is hidden from the generated changelog. - TEST: write\_size/4 — max\_width, max\_height, Unicode combining, wide chars New write\_size unit in tests/core/test\_write.pl covers: - ADDED: write\_size(+Term, -Width, -Height, +Options) write\_size/4 replaces the now-deprecated write\_length/3. It returns both the maximum line Width and the Height (line count) the term would occupy when printed with write\_term. Options max\_width(+N) and max\_height(+N) cap the computation — the underlying Swrite\_lss aborts as soon as either threshold would be exceeded, bounding the runtime for huge terms. - CLEANUP: pl\_write\_term refactored to PRED\_IMPL; foreign\_t → bool Convert pl\_write\_term2 / pl\_write\_term3 (and pl\_print2) from the legacy foreign\_t/pl\_\* calling convention to PRED\_IMPL, and change pl\_write\_term3’s signature to return bool. Also drop its entries from the foreigns table in pl-ext.c now that they are registered via PRED\_DEF in pl-write.c. - CLEANUP: bool return types for pl-text and pl-string APIs Convert int → bool for PL\_unify\_text\_range, PL\_promote\_text, PL\_mb\_text, PL\_canonicalise\_text\_ex, PL\_concat\_text, PL\_save\_text, PL\_text\_recode, and the return path of several PRED\_IMPL functions in pl-string.c. Introduce get\_text\_t / cmp\_t enums for the tri-valued get/compare cases so the type tells the reader what to expect instead of relying on int conventions. - ENHANCED: Make format/2 position logic aware of Unicode column logic. - ADDED: Unicode-aware Supdatepos() using wcwidth() Supdatepos() now counts double-width characters (CJK, emoji) as 2 columns and combining marks as 0, via wcwidth(). On systems lacking a native wcwidth() we fall back to the embedded mk\_wcwidth.c from Markus Kuhn’s public-domain implementation; src/CMakeLists.txt adds it to SRC\_CORE when HAVE\_WCWIDTH is false. - ADDED: code\_type/2: code\_type(Code, width(W)) Provides an interface to wcwidth(). - UNICODE: Fixed reading in `DerivedCoreProperties.txt` for Unicode-17.0. Required to update our tables to Unicode version 17. - FIXED: Missing declaation of listing/2 for option source(+Boolean) - FIXED: #1483 `'` numbers can be misread. A possible read buffer reallocation can cause misreading the base and result in a syntax error or wrong value. - FIXED: WASM: Convert class and property names with Unicode. - CLEANUP: Remove sigaction configuration - FIXED: GC error If a non-deterministic foreign predicate triggers a constraint wakeup. - FIXED: spy/1 to enable debug mode again - PORT: Fix cmake SDL3 import handling for non-standard configurations (PGO) ## Package debian - ADDED: libutf8proc-dev as new dependency. ## Package libedit - FIXED: swipl.exe console uses ReadConsoleInputW (Unicode) for input libedit4pl.c’s Windows console read\_char called ReadConsoleInput, which without UNICODE/\_UNICODE defined expands to the ANSI variant ReadConsoleInputA. The A variant fills uChar.AsciiChar as a CP\_437 / CP\_ACP byte. We then read uChar.UnicodeChar from the same union, which gets at best a low-byte-truncated value (the A-call leaves the high byte of UnicodeChar undefined; in practice it’s 0, so the byte value is read as a BMP code point of the same numeric value). - FIXED: pl\_getc reassembles UTF-16 surrogate pair on Windows editline.pl’s bracketed\_paste/3 handler collects pasted bytes with el\_getc/2 (foreign pl\_getc) and then calls string\_codes/2 on the result. On Windows wchar\_t is a 16-bit UTF-16 code unit, so a supplementary code point (emoji like 🤩, most non-BMP CJK) lives in the line buffer as a surrogate pair and el\_wgetc returned each half as its own int. The resulting code list contained lone surrogates, and string\_codes/2 aborted with - ENHANCED: libedit4pl.c — reassemble supplementary code points on read (Windows) On Windows wchar\_t is a 16-bit UTF-16 code unit, but read\_char read one UTF-8 code point at a time and stored it directly in a wchar\_t: - ADDED: Use “bracketed paste mode” in the terminal. This avoids interpreting TAB as completion and speeds up long paste in Windows. ## Package pldoc - ENHANCED: doc\_latex translates common Unicode punctuation to LaTeX pdflatex under T1 encoding rejects non-ASCII bytes, so any PlDoc comment containing an em-dash, arrow, fancy quote or ellipsis produced hundreds of “non-ASCII character” warnings and the glyph often did not render. ## Package protobufs - CLEANUP: reformat with pprint:print\_term (changed whitespace) - CLEANUP: fix some typos ## Package ssl - PORT: Update for openssl 4.0 - CLEANUP: Remove dead conditional code. Now only support OpenSSL 1.1.1 and LibreSSL. ## Package utf8proc - BUILD: Probe optional libutf8proc features and gate on HAVE\_UTF8PROC\_\* Older distro libutf8proc (Ubuntu LTS) ships a version that predates charwidth, ambiguous\_width, indic\_conjunct\_break, the EXTENDED\_PICTOGRAPHIC boundclass, and single-codepoint case mappings. The binding now probes those at configure time via check\_c\_source\_compiles / check\_symbol\_exists and emits HAVE\_UTF8PROC\_\* macros in config.h; unicode\_property/2 branches and the associated foreign registrations are guarded by the matching #ifdef, so the binding compiles against any utf8proc version from 2.4 onwards. - TEST: Full regression suite for library(unicode) Replace the stub test\_utf8proc.pl with a plunit suite covering the features introduced in the previous commit: - ENHANCED: Review unicode\_property/2, add grapheme and casefold APIs Rewrite unicode4pl.c around the current utf8proc property struct and use SWI-Prolog’s public PL\_type\_error / PL\_domain\_error / PL\_resource\_error helpers and `bool` return types. - BUILD: Use external libutf8proc via pkg-config The package used to bundle utf8proc.c (plus its generated utf8proc\_data.c), a local Makefile, a Ruby gem wrapper, and PostgreSQL binding code — all inherited from a pre-Julia upstream that no longer matches the maintained library. utf8proc is now maintained by the Julia project at [GitHub - JuliaStrings/utf8proc: a clean C library for processing UTF-8 Unicode data · GitHub](https://github.com/JuliaStrings/utf8proc) and ships in every major package manager, so there is no reason to vendor it here. ## Package xpce - FIXED: rlc\_resize, rlc\_shift\_lines\_up, rlc\_clear\_from\_cursor + tests Three related fixes to the xpce ring-buffer terminal and a new regression-test unit that exercises them: - FIXED: terminal paints each wide-char cluster at its column-grid x Pango’s natural advance for an emoji (or other wide glyph) runs a pixel or two wider than the 2\*cw the terminal grid allocates. Feeding a long same-flags run like “sskll​😀😀😀😀😀😀😀” as a single s\_print\_utf8 call made each emoji nudge the subsequent glyphs rightward, so the rendered text drifted past the column-grid cursor that rlc\_paint\_text walks in integer cell units. With the caret walked to the end the text and the caret no longer meet, and a selected “see​😀” left the trailing “aap” shifted left. - FIXED: s\_print\_utf8 reads baseline from the new text, not the stale layout s\_print\_utf8 was calling pango\_layout\_get\_baseline BEFORE pango\_layout\_set\_text, so `baseline` reflected whatever text was in the cached PangoLayout from the previous call. Consecutive calls with different scripts (e.g. an emoji glyph followed by plain ASCII) baseline-aligned each draw against the previous call’s text, so the second draw could land a few pixels above or below the intended baseline. - FIXED: raise META\_OFFSET above Unicode range META\_OFFSET (used to tag Meta-prefixed key events in the same integer space as character code points) was 1\<\<16 = 65536. Any non-BMP code point (emoji like 🤩 U+1F612, most supplementary-plane characters) is \>= 65536 and collided with the Meta flag. - TEST: terminal\_non\_bmp — exercise supplementary-plane code points Add five tests covering the non-BMP path that matters most on Windows, where wchar\_t is 16-bit UTF-16 and a single code point (U+10000 and above) occupies two buffer slots as a surrogate pair. Linux stores the same code point in one wchar\_t; these tests pin both platforms to the same observable behaviour: - TEST: extend random-test model to multi-row layout Teach test\_terminal\_random’s pure-Prolog model to lay clusters out across rows (model\_layout/8) so it can verify content that legitimately wraps. Rows are grouped by the row they occupy and each row is compared to xpce’s \<-row. The per-session content cap is lifted from one row to 900 visual columns so random sessions routinely exercise the libedit+xpce wrap path. - TEST: regression for re\_refresh wide-at-cursor using visual column Locks in the libedit fix: long NFD content + emoji at cursor must not bump the computed cursor row based on code-point count. - FIXED: rlc\_insert must not clamp tl-\>size to b-\>width rlc\_insert extended the line’s cell size only up to b-\>width when opening a gap for the inserted character. For a line whose content is NFD-heavy (combining marks occupy cell slots but zero visual columns) the cell count can legitimately exceed b-\>width while the visible content still fits, and clamping trimmed the trailing cells — most visibly the final cluster’s combining mark, which disappeared from the display after a mid-line insert. - FIXED: ANSI column positioners must not clamp caret\_x to width-1 rlc\_set\_caret (CUP / HVP) and rlc\_set\_caret\_x (CHA / HPA) clamped the resulting caret\_x to b-\>width-1 after converting the requested visual column via rlc\_vcol\_to\_cell. For plain narrow content that’s a no-op, but for rows containing combining marks or wide characters the cell index at visual column K legitimately exceeds K (a base + combiner occupies two cells for one column). The clamp silently snapped the caret back to the middle of the row, so libedit’s re\_refresh\_cursor —which uses CHA to position after cross-wrap moves — landed at the physical midpoint of an NFD-filled line instead of its right edge. - FIXED: count wide-char clusters as 2 visual columns in DCH rlc\_delete\_chars decremented its visual-column budget by the base cell’s per-cell width (1), which is wrong for a wide char: the cluster spans base + placeholder = 2 visual columns. So a DCH with count=2 only removed one wide cluster _plus_ a character from the next cluster, instead of a single wide cluster. - FIXED: preserve cell content when rlc\_put lands a combiner mid-line rlc\_put’s combining-mark branch used to write the combiner into the cell at caret\_x and advance. For a re-render that inserts an NFD cluster with libedit’s T\_IC + overwrite, the overwrite’s combining mark clobbered the base that T\_IC had just shifted into place, dropping the following character from the visible row. - TEST: add test\_terminal\_random/2,3 and first regression it surfaced Introduces a random driver that types and edits through libedit and after each command verifies the cursor and row text against a pure- Prolog model of where they should be. Single-row v1 (buffer capped to fit on one row). On the first divergence the driver prints the seed, the full session command history, and the expected/actual state, then throws. - TEST: Added terminal test framework. - ADDED: epilog/1 option object(-Ref) Get the xpce object reference of the new terminal. - ENHANCED: Unicode-aware terminal cell model Teach the xpce terminal’s line buffer to store Unicode content by code point, with per-cell display width, and to move the caret by grapheme clusters or by visual columns as the context requires. Summary of the model: - ADDED: `terminal_image<-cursor_position` and `terminal_image<-row` This allows for writing terminal handling tests. - ADDED: `text_item->placeholder` Modernizing the UI - ADDED: `library(pce_backcomp)`: Provide better backward compatibility - MODIDIED: Renamed `->fill_pattern` to `->fill` Only filling with colours is currently supported. Even if we re-introduce filling with images, the name `fill_pattern` is not accurate. - MODIFIED: Use .SVG images for most of the icons of the development tools This patch replaces most 16x16 .PNG image files with .SVG files, most of which are scaled to the context, e.g., the line height. Furthermore - ADDED: Support loading `image` objects from SVG data - FIXED: Windows `display<-selection` to return canonical newlines. This patch also fixes a memory leak: the selection string as obtained from SDL3 was not freed. The leak applies to all platforms. - FIXED: `figure->shadow` rendering. - FIXED: `image<-clip` and `image<-scale` to same size Preserve transparency of the source. - ADDED: Support loading images from SVG files. This uses the SDL-Image library support for _nano-svg_. This is a subset of SVG. - MODIFIED: Deleted `image<-monochrome`. This is a bit outdated. Use `<-grayscale` instead. - SDL: Implemented `image<-rotate`. - ENHANCED: Allow resizing tool button images. - SDL: Implemented `image<-scale`. - ENHANCED: Several PceDraw issues. - ENHANCED: Default arrow size. - FIXED: `ellipse->shadow` drawing. - ENHANCED: Slider look handling. Now implements `gtk` and `motif` Deleted old `open_look` and `win` (black-and-white) looks. - FIXED: Pen±fill handling for box and arc/ellipse - FIXED: Properly draw 3D elevated triangles - ENHANCED: sizing 3D triangles to indicate popup menus based on the font metrics. - FIXED: draw 3D triangle correctly. - FIXED: Dynamic switching `menu_bar->look`. - FIXED: Use 64-bit millisecond timer Now uses SDL\_GetTick() for portability and consistently type as `int64_t`. This avoids 49 days overflow on Windows. - SDL: Implemented `@pce<-user`. - MODIFIED: Deleted console methods on `@pce` This deletes `@pce->`console_label`, `@pce-\>expose\_console`, `@pce-\>iconify\_console`and`@pce-\>show\_console``. - FIXED: _Edit_ menu from profiler. - ADDED: PceEmacs find\_references (M-?) to reuse the window unless pinned. - SDL: Implemented `frame<-image`. - SDL: Implemented `window->flash`. - MODIFIED: Deleted `window->grab_keyboard` Modern GUI frameworks no longer allow global grabbing of the keyboard. - SDL: Added enable/disable screen saver. - SDL: Implemented `display->bell` Was a stub. Now generates a sentesized beep. - SDL: Re-implement reading images from arbitrary streams. - SDL: Re-added support for images in saves objects. - SDL: Support `image->save` Currently only for PNG format and only to a file. - SDL: Implemented `image<-pixel`. This patch also introduces @rgba, a hash table that maps RGBA quadruples into existing colour objects. - SDL: Implemented `window->pointer` (was a dummy) Does not work on all platforms. Still succeeds silently. - FIXED: `frame->background` for open frames. - FIXED: Help menu of Profiler to open correct page. - CLEANUP: Remove RCS version control on classes. - MODIFIED: Deleted `display->open` SDL has no such notion. It could be in _headless_ mode, but then it still has a display that can be considered “open”. - MODIFIED: Deleted unimplemented methods on class `frame`: These methods have been removed as there is no cross-platform support for them: - MODIFIED: Deleted `frame<-thread` All windows are managed by the SDL main thread. - DELETED: `frame->grab_pointer`. This is not supported by security-aware window systems. - MODIFIED: Deleted `frame<->border`. - MODIFIED: Deleted `frame->hide`. This commit removes several low-level bindings that are not supported by SDL and typically considered outdated. - CLEANUP: Remove left overs of built-in XBM images - CLEANUP: Remove X11 resource reference database. - CLEANUP: Remove `SDL_GRAPHICS` macro (now unconditionally) - CLEANUP: Remove `WIN32_GRAPHICS` configuration - CLEANUP: Remove configuration for sigaction and union wait. - FIXED: Use a safe way to send messages from a singal handler. This is used to inform xpce when the status of a child process has changed. - CLEANUP: Remove CMake define `USE_SIGINFO` - CLEANUP: Remove various X11 traces - PORT: Support SDL \< 3.2.12, which does not provide accumulative wheel events. - FIXED: Windows: `frame<-{open,save}_file` work around for SDL3.4.0 bug. - PORT: Windows: get rid of remaining XBM images. Fixes XopenImage(@ms\_left\_arrow\_image/image) message and shows correct left arrow for pullright menus. ---
**Author:** ![swi](https://avatars.discourse-cdn.com/v4/letter/s/51bf81/32.png) [@swi](https://swi-prolog.discourse.group/u/swi)\ **Post date:** [April 23, 2026, 3:16am UTC](https://swi-prolog.discourse.group/t/ann-swi-prolog-10-1-6/9577/2 "2026-04-23T03:16:21Z")
Pasting text broke in vi mode with the following in .swiplrc: ```prolog % vi mode in libedit :- multifile editline:el_setup/1. editline:el_setup(Stream) :- el_bind(Stream, \['-v'\]). el_bind(Stream, \["^I", complete\]), % el_bind(Stream, \["^[?", show_completions]), el_bind(Stream, \["^R", isearch_history\]). ``` thanks for the new release! ---
**Author:** ![jan](https://yyz2.discourse-cdn.com/free1/user_avatar/swi-prolog.discourse.group/jan/32/4_2.png) [@jan](https://swi-prolog.discourse.group/u/jan)\ **Post date:** [April 23, 2026, 11:08am UTC](https://swi-prolog.discourse.group/t/ann-swi-prolog-10-1-6/9577/3 "2026-04-23T11:08:34Z")
> [@swi](#): > > Pasting text broke in vi mode with the following in .swiplrc: Thanks. Pushed a fix. The way bracketed paste is handled is basically incompatible with `vi`s moded operation (at least how libedit implements this), so I added some calls that allows us to make bracketed paste a mode that can be controlled and added `el_get(Stream, editor(-Editor))`. We now only enable this in `emacs` mode. Supporting it in `vi` mode requires implementation in the low-level `libedit`. ---
**Author:** ![Boris](https://yyz2.discourse-cdn.com/free1/user_avatar/swi-prolog.discourse.group/boris/32/7486_2.png) [@Boris](https://swi-prolog.discourse.group/u/Boris)\ **Post date:** [April 25, 2026, 9:31am UTC](https://swi-prolog.discourse.group/t/ann-swi-prolog-10-1-6/9577/4 "2026-04-25T09:31:32Z")
TL;DR: Not sure how to find utf8proc when installed with Homebrew. Workaround: use `-DINSTALL_DOCUMENTATION=OFF`. * * * > [@jan](#): > > **Building!** > > - There is a **new dependency**. Our unmaintained copy of `utf8proc.c`, underlying `library(unicode)` has been removed. All major platforms now support this package and thus we use the platform version. Install using > - Debian based systems: `apt-get install libutf8proc-dev` > - Redhat/Fedora based systems: `dnf install utf8proc-devel` > - Macports: `port install utf8proc` Has anyone managed to use the Homebrew installation of utf8proc on MacOS? I _think_ I have installed it: ```prolog $ brew reinstall utf8proc ==> Fetching downloads for: utf8proc ✔︎ Bottle Manifest utf8proc (2.11.3) Downloaded 7.5KB/ 7.5KB ✔︎ Bottle utf8proc (2.11.3) Downloaded 107.4KB/107.4KB ==> Reinstalling utf8proc ==> Pouring utf8proc--2.11.3.arm64_sequoia.bottle.1.tar.gz 🍺 /opt/homebrew/Cellar/utf8proc/2.11.3: 15 files, 476KB ==> Running `brew cleanup utf8proc`... Disable this behaviour by setting `HOMEBREW_NO_INSTALL_CLEANUP=1`. Hide these hints with `HOMEBREW_NO_ENV_HINTS=1` (see `man brew`). ``` The necessary header is there: ```prolog $ ls /opt/homebrew/opt/utf8proc/include utf8proc.h ``` but then CMake finds _some_ thing, but not the right thing: ```prolog -- Checking for module 'libutf8proc' -- Found libutf8proc, version 2.11.3 -- Performing Test HAVE_UTF8PROC_CHARWIDTH -- Performing Test HAVE_UTF8PROC_CHARWIDTH - Failed -- Performing Test HAVE_UTF8PROC_AMBIGUOUS_WIDTH -- Performing Test HAVE_UTF8PROC_AMBIGUOUS_WIDTH - Failed -- Performing Test HAVE_UTF8PROC_BIDI_MIRRORED -- Performing Test HAVE_UTF8PROC_BIDI_MIRRORED - Failed -- Performing Test HAVE_UTF8PROC_INDIC_CONJUNCT_BREAK -- Performing Test HAVE_UTF8PROC_INDIC_CONJUNCT_BREAK - Failed -- Performing Test HAVE_UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC -- Performing Test HAVE_UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC - Failed -- Looking for utf8proc_unicode_version -- Looking for utf8proc_unicode_version - not found -- Looking for utf8proc_codepoint_valid -- Looking for utf8proc_codepoint_valid - not found -- Looking for utf8proc_grapheme_break_stateful -- Looking for utf8proc_grapheme_break_stateful - not found -- Looking for utf8proc_tolower -- Looking for utf8proc_tolower - not found -- Looking for utf8proc_toupper -- Looking for utf8proc_toupper - not found -- Looking for utf8proc_totitle -- Looking for utf8proc_totitle - not found ``` Any ideas? Not a show stopper of course, the documentation is available online. EDIT: for clarity, here is what I see in my build/CMakeCache.txt on MacOS (Apple Silicon). I cannot tell what is the difference between utf8proc and libutf8proc. On my Linux machine the library is indeed called libutf8proc and everything works out splendidly. ```prolog UTF8PROC_CFLAGS:INTERNAL=-I/opt/homebrew/Cellar/utf8proc/2.11.3/include;-DUTF8PROC_EXPORTS UTF8PROC_CFLAGS_I:INTERNAL= UTF8PROC_CFLAGS_OTHER:INTERNAL=-DUTF8PROC_EXPORTS UTF8PROC_FOUND:INTERNAL=1 UTF8PROC_INCLUDEDIR:INTERNAL=/opt/homebrew/Cellar/utf8proc/2.11.3/include UTF8PROC_INCLUDE_DIRS:INTERNAL=/opt/homebrew/Cellar/utf8proc/2.11.3/include UTF8PROC_LDFLAGS:INTERNAL=-L/opt/homebrew/Cellar/utf8proc/2.11.3/lib;-lutf8proc UTF8PROC_LDFLAGS_OTHER:INTERNAL= UTF8PROC_LIBDIR:INTERNAL=/opt/homebrew/Cellar/utf8proc/2.11.3/lib UTF8PROC_LIBRARIES:INTERNAL=utf8proc UTF8PROC_LIBRARY_DIRS:INTERNAL=/opt/homebrew/Cellar/utf8proc/2.11.3/lib UTF8PROC_LIBS:INTERNAL= UTF8PROC_LIBS_L:INTERNAL= UTF8PROC_LIBS_OTHER:INTERNAL= UTF8PROC_LIBS_PATHS:INTERNAL= UTF8PROC_MODULE_NAME:INTERNAL=libutf8proc UTF8PROC_PREFIX:INTERNAL=/opt/homebrew/Cellar/utf8proc/2.11.3 UTF8PROC_STATIC_CFLAGS:INTERNAL=-I/opt/homebrew/Cellar/utf8proc/2.11.3/include;-DUTF8PROC_EXPORTS UTF8PROC_STATIC_CFLAGS_I:INTERNAL= UTF8PROC_STATIC_CFLAGS_OTHER:INTERNAL=-DUTF8PROC_EXPORTS UTF8PROC_STATIC_INCLUDE_DIRS:INTERNAL=/opt/homebrew/Cellar/utf8proc/2.11.3/include UTF8PROC_STATIC_LDFLAGS:INTERNAL=-L/opt/homebrew/Cellar/utf8proc/2.11.3/lib;-lutf8proc UTF8PROC_STATIC_LDFLAGS_OTHER:INTERNAL= UTF8PROC_STATIC_LIBDIR:INTERNAL= UTF8PROC_STATIC_LIBRARIES:INTERNAL=utf8proc UTF8PROC_STATIC_LIBRARY_DIRS:INTERNAL=/opt/homebrew/Cellar/utf8proc/2.11.3/lib UTF8PROC_STATIC_LIBS:INTERNAL= UTF8PROC_STATIC_LIBS_L:INTERNAL= UTF8PROC_STATIC_LIBS_OTHER:INTERNAL= UTF8PROC_STATIC_LIBS_PATHS:INTERNAL= UTF8PROC_VERSION:INTERNAL=2.11.3 UTF8PROC_libutf8proc_INCLUDEDIR:INTERNAL= UTF8PROC_libutf8proc_LIBDIR:INTERNAL= UTF8PROC_libutf8proc_PREFIX:INTERNAL= UTF8PROC_libutf8proc_VERSION:INTERNAL= ``` ---
**Author:** ![jan](https://yyz2.discourse-cdn.com/free1/user_avatar/swi-prolog.discourse.group/jan/32/4_2.png) [@jan](https://swi-prolog.discourse.group/u/jan)\ **Post date:** [April 26, 2026, 8:29am UTC](https://swi-prolog.discourse.group/t/ann-swi-prolog-10-1-6/9577/5 "2026-04-26T08:29:46Z")
Odd. Just updated by Homebrew on an M1 running latest MacOS. Builds like a charm. The way to debug is to run ``` cmake -U '*UTF*' . ``` and then look in the build log that is in the CMakeFiles. ---
**Author:** ![Boris](https://yyz2.discourse-cdn.com/free1/user_avatar/swi-prolog.discourse.group/boris/32/7486_2.png) [@Boris](https://swi-prolog.discourse.group/u/Boris)\ **Post date:** [April 26, 2026, 9:16am UTC](https://swi-prolog.discourse.group/t/ann-swi-prolog-10-1-6/9577/6 "2026-04-26T09:16:31Z")
I did run the following, from the swipl-devel directory: ```prolog $ git checkout master && git pull && git submodule update --init $ mkdir build $ cd build $ cmake -U '*UTF*' \ -DCMAKE_INSTALL_PREFIX="$HOME" \ -DSWIPL_PACKAGES_JAVA=OFF \ -DEPILOG=ON \ -G Ninja .. ``` The result of this was that there is no longer any mention of UTF in any of the output and logs, and at the end it fails exactly as before: ```prolog $ ninja [2113/2371] Generating pldoc2tex FAILED: [code=1] man/pldoc2tex /Users/booris/install/swipl-devel/build/man/pldoc2tex cd /Users/booris/install/swipl-devel/build/man && /Users/booris/install/swipl-devel/build/src/swipl -f none --no-packs -o /Users/booris/install/swipl-devel/build/man/pldoc2tex --class=development --no-autoload --no-stand_alone -c /Users/booris/install/swipl-devel/man/pldoc2tex.pl -- --nopce ERROR: /Users/booris/install/swipl-devel/build/home/library/ext/pldoc/doc_latex.pl:1724: ERROR: Exported procedure unicode:unicode_version/1 is not defined ERROR: /Users/booris/install/swipl-devel/build/home/library/ext/pldoc/doc_latex.pl:1724: ERROR: Exported procedure unicode:unicode_codepoint_valid/1 is not defined ERROR: /Users/booris/install/swipl-devel/build/home/library/ext/pldoc/doc_latex.pl:1724: ERROR: Exported procedure unicode:atom_graphemes/2 is not defined ERROR: /Users/booris/install/swipl-devel/build/home/library/ext/pldoc/doc_latex.pl:1724: ERROR: Exported procedure unicode:string_graphemes/2 is not defined Warning: [Thread main] Halting with status 1 due to 4 errors and 0 warnings ``` This is exactly the same error which I get when I do not add the `-U '*UTF*'` flag when I run cmake. Odd indeed. Don’t waste time on this though. ---
**Author:** ![Boris](https://yyz2.discourse-cdn.com/free1/user_avatar/swi-prolog.discourse.group/boris/32/7486_2.png) [@Boris](https://swi-prolog.discourse.group/u/Boris)\ **Post date:** [April 27, 2026, 12:38pm UTC](https://swi-prolog.discourse.group/t/ann-swi-prolog-10-1-6/9577/7 "2026-04-27T12:38:22Z")
> [@jan](#): > > Odd. Just updated by Homebrew on an M1 running latest MacOS. Builds like a charm. The way to debug is to run I am stymied. I now read the [CMakeLists.txt file in packages/utf8proc](https://github.com/SWI-Prolog/packages-utf8proc/blob/80e05394f33c01d78823f1d8e5ff0099ced80357/CMakeLists.txt). I do not understand why the tests fail. [You have defined a function](https://github.com/SWI-Prolog/packages-utf8proc/blob/80e05394f33c01d78823f1d8e5ff0099ced80357/CMakeLists.txt#L24-L34): ```prolog function(utf8proc_has_field FIELD VAR) check_c_source_compiles(" #include int main(void) { utf8proc_property_t p = {0}; unsigned x = p.${FIELD}; (void)x; return 0; } " ${VAR}) endfunction() ``` The [check\_c\_course\_compiles function](https://cmake.org/cmake/help/latest/module/CheckCSourceCompiles.html) supposedly uses `CMAKE_REQUIRED_INCLUDES`; [this variable is set in CMakeLists.txt](https://github.com/SWI-Prolog/packages-utf8proc/blob/80e05394f33c01d78823f1d8e5ff0099ced80357/CMakeLists.txt#L16) with the correct value: ```prolog Variable "CMAKE_REQUIRED_INCLUDES" was accessed using MODIFIED_ACCESS with value "/opt/homebrew/Cellar/utf8proc/2.11.3/include". ``` This seems correct: ```prolog $ ls /opt/homebrew/Cellar/utf8proc/2.11.3/include utf8proc.h ``` If I make a file and compile it, it all works (I replaced `${FIELD}` with the first checked field, `charwidth`). ```prolog $ cat foo.c #include int main(void) { utf8proc_property_t p = {0}; unsigned x = p.charwidth; (void)x; return 0; } $ gcc -I/opt/homebrew/Cellar/utf8proc/2.11.3/include foo.c $ ./a.out $ echo $? 0 ``` Why does the test fail? Again, this is what it says: ```prolog -- Performing Test HAVE_UTF8PROC_CHARWIDTH -- Performing Test HAVE_UTF8PROC_CHARWIDTH - Failed ``` I checked the run-time values of all used variables by adding ```prolog variable_watch(VARIABLE_NAME) ``` in the CMakeLists.txt file inside packages/utf8proc, they all seem to have the correct values. I tested all of this inside the package/utf8proc directory, git says: ```prolog $ git status HEAD detached at 80e0539 ``` This seems to be correct too, this is the relevant commit in swipl-devel: ```prolog $ git show 8bdd070950f0036899888827c22b19e9667ef80e commit 8bdd070950f0036899888827c22b19e9667ef80e Author: Jan Wielemaker Date: Thu Apr 23 13:57:43 2026 +0200 Updated utf8proc diff --git a/packages/utf8proc b/packages/utf8proc index a6ee0fde9..80e05394f 160000 --- a/packages/utf8proc +++ b/packages/utf8proc @@ -1 +1 @@ -Subproject commit a6ee0fde94f669e3cb64c0959c5bf298feb89dcd +Subproject commit 80e05394f33c01d78823f1d8e5ff0099ced80357 ``` ---
**Author:** ![jan](https://yyz2.discourse-cdn.com/free1/user_avatar/swi-prolog.discourse.group/jan/32/4_2.png) [@jan](https://swi-prolog.discourse.group/u/jan)\ **Post date:** [April 27, 2026, 1:32pm UTC](https://swi-prolog.discourse.group/t/ann-swi-prolog-10-1-6/9577/8 "2026-04-27T13:32:51Z")
Probably the clue is in CMakeFiles/CMakeConfigureLog.yaml. That gives you the executed commands and their output. You may have to re-run the failing parts using ``` cmake -U '*UTF*' . ```