Activity for gnuplot
-
Ethan Merritt modified ticket #2830
build problems under Fedora 42
-
The command "set terminal qt size ..." changes the size of the plot window
-
SweetTyler posted a comment on ticket #2832
I also suspect that it is a bug of Fedora. Please discard this report.
-
I cannot reproduce this. For me the "load" command resets the size and it stays that way unless explicitly changed by dragging a corner of the window. On the other hand, I'm not even sure what I expect would happen. Should the size given in a previously created file override the size that has been adjusted to the current user's preference? I can see an argument either way. I do think it's a bit odd that you and I see different behavior. Could that be due to the window manager? I'm using Qt/Plasma...
-
My original title might be misleading; a better one could be “Plot window size cannot be reset correctly ...” or something similar.
-
The command "set terminal qt size ..." changes the size of the plot window
-
Mugundan MCW created ticket #831
Add support for installing X64 gnuplot installer on Windows ARM64
-
gnuplot released /gnuplot/6.0.3/gnuplot-6.0.3-c23.tar.gz
-
-
-
-
Ethan A Merritt committed [7efb45] on Git Repository
Add autoconf version check to the release checklist
-
Ethan A Merritt committed [7d1fb9] on Git Repository
x11: rearrange include statements to make -std=c17 and -std=c23 happy
-
-
sounds great!
-
Shigeharu TAKENO created ticket #2831
japanese tooltip text is broken of wgnuplot
-
Followup: autoconf 2.72 has modified the AC_HEADER_STDBOOL macro to better accommodate c23. The gcc15 + c23 problem is ameliorated when this version of autoconf is used to prepare gnuplot's configure script. I am considering to release a gnuplot 6.0.4 distribution package earlier than usual so that people trying to build on very recent linux distro versions (Fedora 42/43 in particular) get the benefit of an updated configure script and do not have to fiddle with compiler switches or locally regenerating...
-
Followup: autoconf 2.72 has modified the AC_HEADER_STDBOOL macro to better accommodate c23. The gcc15 + c23 problem is ameliorated when this version of autoconf is used to prepare gnuplot's configure script. I am considering to release a gnuplot 6.0.4 distribution package earlier than usual so that people trying to build on very recent linux distro versions (Fedora 23 in particular) get the benefit of an updated configure script and do not have to fiddle with compiler switches or locally regenerating...
-
And finally, thank you for your help!
-
export CFLAGS="-std=gnu17" make the job done (no warning, no error), and no need to modify syscfg.h.
-
Thank you for trying. I don't think I will be able to take this any further unless/until I have a newer gcc (and maybe libc?) that exhibits this problem. And that may not happen if it turns out this breakage is specific to Fedora. As I said, I do have gcc15 here, but it is one release behind yours (15.2.0) and does not have whatever tweaks Fedora made as discussed in that developers blog a gave a link to. The following active bugs in the Fedora Bugzilla tracker seem relevant: https://bugzilla.redhat.com/show_bug.cgi?id=2342124...
-
I replaced src/syscfg.h with yours and then ran "configure", followed by "make". The error messages were as same as before. If you need any detail, please let me know.
-
I see what is going on. Starting with Fedora 42 the default compiler settings have been changed to use -std=c23. This is a work in progress both on the Fedora packaging side and the stabilization of of the gcc15 compiler. Breakage is expected. See for example https://developers.redhat.com/articles/2025/04/10/6-usability-improvements-gcc-15# You can probably bypass the _Bool error by using the gcc compiler switch -std=c17 (although I don't have Fedora 42 to test on). I would like to find a fix that...
-
build problems under Fedora 42
-
I see what is going on. Starting with Fedora 42 the default compiler settings have been changed to use -std=c23. This is a work in progress both on the Fedora packaging side and the stabilization of of the gcc15 compiler. Breakage is expected. See for example https://developers.redhat.com/articles/2025/04/10/6-usability-improvements-gcc-15# You can probably bypass the _Bool error by using the gcc compiler switch -std=c17 (although I don't have Fedora 23 to test on). I would like to find a fix that...
-
Please find attached config.log and config.h. the command line I ran was simply "./configure" after I untar the downloaded source package. The third file, "screen_printout" was the direct screen output after command "./configure". When I tried to compile gnuplot, the first trouble was a complaint of "QtCore" couldn't be found, after I installed Qt (both 5 and 6) then I got the errors that I mentioned above.
-
Huh. Going back to the error message you initially reported: /syscfg.h:388:15: error: ‘_Bool’ does not name a type 388 | # define bool _Bool | ^ It should only be possible to reach this line of code if the system header file <stdbool.h> is missing. This header should be provided by all compilers that are c99 compliant. There is some odd wording in the c23 documentation that implies this file is "obsolescent" but what that means in practice to existing source code I don't know, since the compiler...
-
My gcc version from "gcc --version" is: gcc (GCC) 15.2.1 20250924 (Red Hat 15.2.1-2) and Qt version fom "qmake --version" is: QMake version 3.1 Using Qt version 6.9.2 in /usr/lib64 and my OS from "cat /etc/os-release" is: NAME="Fedora Linux" VERSION="43 (Workstation Edition Prerelease)" RELEASE_TYPE=development ID=fedora VERSION_ID=43 VERSION_CODENAME="" PRETTY_NAME="Fedora Linux 43 (Workstation Edition Prerelease)" ANSI_COLOR="0;38;2;60;110;180" LOGO=fedora-logo-icon
-
That is strange. Could you please report what version of gcc (or clang?) you are using, and what compiler flags were selected by the configure script? For example, as noted in the INSTALL guide it may help to use $ export CXXFLAGS="-Wall -g -Og -fPIC -std=c++11" Also check the qt version. If it's a super recent qt6 but Fedora also supplies optional qt5 libraries, try configuring with ./configure --with-qt=qt5. FWIW I can attest it all works with gcc 15.2.0 or clang 20.1.7 and qt 6.8. If there are...
-
Just found building from the source is not easy! I could easily do it when I was on CentOS 7.9, however, with the latest Fedora, I got MANY errors related with head files and/or libraries. ... /syscfg.h:388:15: error: ‘_Bool’ does not name a type 388 | # define bool _Bool | ^~~~~ ./syscfg.h:404:18: note: in expansion of macro ‘bool’ 404 | #define TBOOLEAN bool | ^~~~ ./gadgets.h:87:5: note: in expansion of macro ‘TBOOLEAN’ 87 | TBOOLEAN hypertext; | ^~~~~~~~ ./syscfg.h:388:15: error: ‘_Bool’ does...
-
Thanks for the reply. The following is what outputs from "show version long": gnuplot> show version long G N U P L O T Version 6.0.3 patchlevel 3 last modified 2025-06-01 Copyright (C) 1986-1993, 1998, 2004, 2007-2025 Thomas Williams, Colin Kelley and many others gnuplot home: http://www.gnuplot.info faq, bugs, etc: type "help FAQ" immediate help: type "help" (plot window: hit 'h') Compile options: +READLINE -LIBREADLINE +HISTORY +OBJECTS +STATS +WATCHPOINTS +EXTERNAL_FUNCTIONS +POLARGRID +COMPLEX_FUNCS...
-
Thanks for the reply. The following what outputs from "show version long": gnuplot> show version long G N U P L O T Version 6.0.3 patchlevel 3 last modified 2025-06-01 Copyright (C) 1986-1993, 1998, 2004, 2007-2025 Thomas Williams, Colin Kelley and many others gnuplot home: http://www.gnuplot.info faq, bugs, etc: type "help FAQ" immediate help: type "help" (plot window: hit 'h') Compile options: +READLINE -LIBREADLINE +HISTORY +OBJECTS +STATS +WATCHPOINTS +EXTERNAL_FUNCTIONS +POLARGRID +COMPLEX_FUNCS...
-
This is a known limitation of the minimal built-in "readline" functionality that gnuplot provides if no external library was provided. It is less functional than either of the two library options: libreadline or libedit The choice of which input path to use is made when the program is built. The default is to use libreadline if available, falling back to libedit if available, and falling back to the minimal built-in code if neither library is available. Or you can request any of the three options...
-
long command in terminal-qt messed up by cursor movement
-
Ethan A Merritt committed [0baecb] on Git Repository
update NEWS and RELEASE_NOTES to match current state of 6.0.4alpha (23 Sep 2025)
-
Ethan A Merritt committed [0c4e05] on Git Repository
hpgl: font name parsing overruns the string by one char
-
Ethan A Merritt committed [9fa287] on Git Repository
change http://gnuplot.info to http://www.gnuplot.info in all documentation
-
Ethan A Merritt committed [32d903] on Git Repository
docs: clean up the heading for New Features in the html output
-
Ethan A Merritt committed [47ab2b] on Git Repository
set/show/save imaginary_i
-
Ethan A Merritt committed [0e4789] on Git Repository
show/save "set micro" was not saving any user-specified string
-
Ethan A Merritt committed [93d1f1] on Git Repository
New gprintf format specifiers %C and %Ci
-
Ethan A Merritt committed [81ecf6] on Git Repository
Transparent sixelgd option doesn't work
-
Ethan A Merritt committed [ea3e1a] on Git Repository
Allow "smooth bezier with filledcurves between"
-
Ethan A Merritt committed [482185] on Git Repository
document linetypes 0 and -1; include lt 0 in the output from "show linetypes"
-
Ethan A Merritt committed [6cfcd4] on Git Repository
Fix above/below treatment of filledcurves in polar mode
-
Ethan A Merritt committed [566d4f] on Git Repository
Partially deprecate the "sample" keyword in plot commands (back-port from 6.1)
-
Ethan A Merritt committed [bc7dbe] on Git Repository
infinity has no decimal point
-
Ethan A Merritt committed [0b5314] on Git Repository
docs: cleanup
-
Ethan A Merritt committed [827d42] on Git Repository
update pm3d/README
-
Ethan A Merritt committed [5296d1] on Git Repository
move TODO and README.emacs to project_notes/ directory
-
Riccardo Lombardi modified a comment on ticket #593
Hello Ethan, thank you for your feedback. I agree that this is not an actual problem, it's more like being perfectionist/obsessive about the real pdf size. I also agree that 0,15mm difference is unnoticeable in the real world. However, some more exotic software or clients could suffer from this small difference. Some examples I found in my investigation: https://forum.affinity.serif.com/index.php?/topic/232218-affinity-designer-v2-size-adding-005mm-when-exporting-to-pdf/ https://community.adobe.com/t5/indesign-discussions/cannot-set-page-size-with-211-1-mm-exports-pdf-as-211-0-mm/m-p/9014231...
-
Hello Ethan, thank you for your feedback. I agree that this is not an actual problem, it's more like being perfectionist/obsessive about the real pdf size. I also agree that 0,15mm difference is unnoticeable in the real world. However, some more exotic software could suffer from this small difference. Some examples I found in my investigation: https://forum.affinity.serif.com/index.php?/topic/232218-affinity-designer-v2-size-adding-005mm-when-exporting-to-pdf/ https://forum.affinity.serif.com/index.php?/topic/150163-different-document-dimensions-after-pdf-export/...
-
Thank you for your suggestion, but Is there an actual problem here? I understand that A3 paper dimensions cannot be expressed as an integral number of points. I am wondering what are the circumstances where the residual fractional point (approx 0.15 mm) matters in practice. For what it's worth, the pdf file produced by gnuplot is identified as "Landscape A3" when I open it in a pdf viewer (okular in this case). I agree that your proposed change is harmless with respect to pdf output. I am uncertain...
-
Fontconfig warning: using without calling FcInit()
-
MBoehme posted a comment on ticket #2817
I want to give an update. Looks like this error was caused by "cairo graphics" lib on my system. After recompiling the lib, the error does not happens any more.
-
Hi, a quick update about the feature request. Disclaimer: I have some programming skills, but I have zero experience in team-based coding or tools like GitHub/SourceForge. I locally installed VB Studio and managed to compile a local version of gnuplot with mingw. After messing around with the code (with the help of Copilot and Chatgpt), I came to the conclusion that the solution is quite simple: The variables device_xmax/device_ymax of the plot_struct in "src/wxterminal/gp_cairo.h" (line 115) are...
-
Hi, a quick update about the feature request. Disclaimer: I have some programming skills, but I have zero experience in team-based coding or tools like GitHub/SourceForge. I locally installed VB Studio and managed to compile a local version of gnuplot with mingw. After messing around with the code (with the help of Copilot and Chatgpt), I came to the conclusion that the solution is quite simple: The variables device_xmax/device_ymax of the plot_struct in "gp_cairo.h" (line 115) are currently declared...
-
Hi, a quick update about the feature request. Disclaimer: I have some programming skills, but I have zero experience in team-based coding or tools like GitHub/SourceForge. I locally installed VB Studio and managed to compile a local version of gnuplot with mingw. After messing around with the code (with the help of Copilot and Chatgpt), I came to the conclusion that the solution is quite simple: The variables device_xmax/device_ymax of the plot_struct in "gp_cairo.h" (line 115) are currently declared...
-
Yes. You can apply a filter operation to the data. Since filtering is done prior to plotting, any points that are filtered out will not appear in the output. See help filter if. In your case: set table $k plot $data using 1:2 if (2 <= column(1) && column(1) <= 4) unset table
-
Daniel Dan K. posted a comment on ticket #2829
Hello Ethan, Thank you for your response. I missed the point in the documentation where the limitations of the plot style with table are clearly described. Now if I plot into a datablock using table mode and not using with table, all points are plotted including an additional column indicating whether they are in range or not. Example: set table $k plot [2:4] $data unset table outputs: # Curve 0 of 1, 5 points # Curve title: "$data" # x y type 1 1 o 2 2 i 3 4 i 4 8 i 5 16 o Can I somehow limit the...
-
increase precision in pdf paper size
-
plotting in table mode ignores x-range
-
From the documentation: gnuplot> help with table This discussion applies only to the special plot style with table. To avoid any style-dependent processing of the input data being tabulated (filters, smoothing, errorbar expansion, secondary range checking, etc), or to increase the number of columns that can be tabulated, use the keyword "table" instead of a normal plot style. In this case the output does not contain an extra column containing a flag i, o, u indicating inrange/outrange/undefined....
-
I filed this ticket. Obviously, one can file bug reports without being logged in. I did not know this.
-
Anonymous created ticket #2829
plotting in table mode ignores x-range
-
Add cdot as multiplication sign
-
OK. I'll look to see what the glitch is with cherry-picking the 6.1 fix, and queue it for 6.0.4
-
Orion Poplawski posted a comment on ticket #2781
Not particularly. I'm just going through various bug reports on gnuplot in Fedora and came across the CVE report related to this crash.
-
The fix used for 6.1 does not apply cleanly to the stable branch. Given that the bug had been in the code for decades with no reported problems, it did not seem worth the effort to back-port or work out a different fix for 6.0. Is this issue causing you a problem in current stable (6.0.3)?
-
Any ETA on a new release with this fix? Thanks.
-
Ah - looks like this: a5897feadc4be73b0ffd8458556c47117bd24d03 hpgl: font name parsing overruns the string by one char if no comma is present in the font name. E.g. set term pcl set title "Title" font "sans" # no comma in font name plot x Bug 2781
-
What commit fixed this? It seems like this was assigned CVE-2025-3359 so it would be nice to have a reference
-
Ethan A Merritt committed [3cbe37] on Git Repository
change http://gnuplot.info to http://www.gnuplot.info in all documentation
-
Ethan A Merritt committed [9cac78] on Git Repository
minor cleanup to project notes
-
Ethan A Merritt committed [73b4a0] on Git Repository
New repository subdirectory .../project_notes
-
Ethan A Merritt committed [1e4c45] on Git Repository
move obsolete LaTeX tutorial files to docs/old/
-
Ethan A Merritt committed [9f75a9] on Git Repository
docs: clarify where to find a summary of new features under development
-
Ethan A Merritt committed [8a1b6a] on Git Repository
do not allow user-supplied string as a format specifier
-
Transparent sixelgd option doesn't work
-
James Holderness posted a comment on ticket #2828
neither the original nor the patched version work correctly in the KDE konsole terminal Sorry I should have mentioned that. There are a number of Sixel terminals on which this is not expected to work (at least not any better than it did before), because they simply don't support transparency. In the case of Konsole, they've got an open issue tracking that functionality here: https://bugs.kde.org/show_bug.cgi?id=479125
-
Ethan A Merritt committed [1bbdf4] on Git Repository
Transparent sixelgd option doesn't work
-
Ticket moved from /p/gnuplot/support-requests/308/ Can't be converted: _milestone: _priority: 5
-
Thank you. For what it's worth neither the original nor the patched version work correctly in the KDE konsole terminal. But I can confirm that the patched version works better in wezterm. I am reclassifying this as a bug report so that the commit message can refer back to it.
-
Transparent sixelgd option doesn't work
-
Ethan A Merritt committed [78ba08] on Git Repository
Allow "smooth bezier with filledcurves between"
-
Enable 'smooth bezier' for 'with filledcurves between' style
-
Hiroki Motoyoshi created ticket #592
Enable 'smooth bezier' for 'with filledcurves between' style
-
"show linetype 0" fails with error message
-
Ethan A Merritt committed [fdddfb] on Git Repository
document linetypes 0 and -1; include lt 0 in the output from "show linetypes"
-
"show linetype 0" fails with error message
-
I can confirm that kimageformats 6.16.0 gives the spurious error and 6.17.0 does not.
-
Ethan A Merritt committed [989e5c] on Git Repository
pm: reset flag term_force_init for each plot
-
Tatsuro MATSUOKA modified a comment on ticket #2826
It may be related to the below https://sourceforge.net/p/gnuplot/bugs/2397/ If I set the environment variable PANGOCAIRO_BACKEND=fc (fontconfig) it seems to be better.
-
Part of letters missing from ylabel/y2label
-
Thanks, setting this environment variable fixes this issue for me. 👍
-
It may be related to the below https://sourceforge.net/p/gnuplot/bugs/2397/ If I set the environment variablePANGOCAIRO_BACKEND=fc (fontconfig) it seems to be better.
-
Part of letters missing from ylabel/y2label
-
Part of letters missing from ylabel/y2label
-
The change should be in kimageformats-6.18.0, if I am reading the repository notes correctly.
-
gnuplot-6.0.x - qt terminal error "IFFChunk::innerFromDevice: unkwnown chunk"
