Treebrowser plugin works incorrectly, if in project settings has relative path. Problem in treebrowser/src/treebrowser.c file. The problem appears when use (pseudocode):
GeanyProject *project = geany->app->project;
dir = project->base_path ;
and not checked the path type - absolute or relative base_path.
The correct realization in the file geany-plugins/projectorganizer/src/prjorg-utils.c (function gchar *get_project_base_path(void)).

A similar issue: geany/geany#686

The link to imageshak seems to be broken. Link target is: http://img43.imageshack.us/img43/9053/screenshotsmallq.png

Most of the changes made in the ~/.config/geany/plugins/multiterm/multiterm.conf are not applied; some options affected (may be more) are:
full_width_tabs
save_tabs
track_title
track_cwd

Please add them

This is a very common paradigm used by GNOME, KDE, MacOSX, Windows, etc., and following it makes our UI more consistent with these and easier to use by their users. As Geany itself follows this now (since geany/geany@9652a34, 1.24), it also makes plugin UIs more consistent inside Geany itself.

There is already a PR for project-organizer: #243.

I tagged with plugins that seem to have issues with this, but I might not have found all of them, and I may have misinterpreted some of them (unlikely, but who knows).

The screenshot at url http://plugins.geany.org/images/plugins/addons/addonsb.png is missing.

It is referenced in the addons documentation here : http://plugins.geany.org/addons.html

Problem are arguments in wrong order for show_stats_dialog() call.

Mine code is located at /home/user/Загрузки but when I run geany debugger it show error "Can't find source code file /home/user/\320\227\320\260\320\263\321\200\321\203\320\267\320\272\320\270/natural.cxx"

The site plugins.geany.org on sourceforge.com seems to be unavailable.

When starting Geany with enabled webhelper plugin it is warning:

Breakpoint 1, handler_log (domain=0x7ffff7a6f89b "Gtk", level=G_LOG_LEVEL_CRITICAL,
msg=0x7fff98020890 "IA__gtk_entry_completion_set_text_column: assertion `column >= 0' failed", data=0x0) at ../src/log.c:121
121 {
(gdb) bt

0 handler_log (domain=0x7ffff7a6f89b "Gtk", level=G_LOG_LEVEL_CRITICAL, msg=0x7fff98020890 "IA__gtk_entry_completion_set_text_column: assertion `column >= 0' failed",

data=0x0) at ../src/log.c:121

1 0x00007ffff5a0a473 in g_logv () from /lib/x86_64-linux-gnu/libglib-2.0.so.0

2 0x00007ffff5a0a622 in g_log () from /lib/x86_64-linux-gnu/libglib-2.0.so.0

3 0x00007fffe7ac7336 in create_toolbar (self=0x9a6810) at ../webhelper/src/gwh-browser.c:984

4 gwh_browser_init (self=0x9a6810) at ../webhelper/src/gwh-browser.c:1136

5 0x00007ffff5ce2a97 in g_type_create_instance () from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0

6 0x00007ffff5cc7818 in ?? () from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0

7 0x00007ffff5cc92d1 in g_object_newv () from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0

8 0x00007ffff5cc991c in g_object_new () from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0

9 0x00007fffe7ac8109 in gwh_browser_new () at ../webhelper/src/gwh-browser.c:1232

10 0x00007fffe7aca04b in plugin_init (data=) at ../webhelper/src/gwh-plugin.c:430

11 0x00000000004ce946 in plugin_load (plugin=0xdc41c0) at ../src/plugins.c:614

12 plugin_new (fname=fname@entry=0xc96680 "/usr/local/lib/geany/webhelper.so", load_plugin=load_plugin@entry=1, add_to_list=add_to_list@entry=0) at ../src/plugins.c:769

13 0x00000000004d00af in load_active_plugins () at ../src/plugins.c:963

14 plugins_load_active () at ../src/plugins.c:1090

15 0x00000000004c0107 in main (argc=1, argv=0x7fffffffe168) at ../src/main.c:1177

This one is hard to reproduce (sorry), and I have no idea what causes this or what is happening.

Steps to reproduce sometimes:

1. Install and enable GeanyLaTeX (in case it is relevant: I have only the "auto-completion" activated)
2. Create and open: A latex file and a Makefile, in the same directory (I use the makefile to compile the latex document)
3. Edit the latex file over a long time, intermittently calling "make" via the shortcut "Shift+F9" (as found in the menu)

Actual behaviour: Sometimes the "Make" menu item becomes and stays disabled. This also disables the keyboard shortcut, so Shift+F9 does not do anything anymore. Switching between geany-tabs or re-opening the file does not help. The only way to work around this is to restart Geany.

Expected behaviour: Shift+F9 and the "Make" menu item stay enabled for the whole time.

Versions: Geany 1.25 "Veed" // "(compiled on 2015-07-12 or later)"

Sorry that I can't provide a more reliable way of reproducing the bug. I have already played with different things, and have, so far, only come to the conclusion that disabling the GeanyLaTeX pluging avoids this bug. However, this is not a guarantee, since I "only" tried that for a week. Otherwise I run into that bug roughly once every two weeks, so this experiment isn't conclusive.

I continue using Geany, so if anyone can give me a better "test", or instructions on what to look for via gdb while this is happening, I'd gladly do that.

The provided command to compile geanylatex's documentation to HTML is not working. Provided command command is

htlatex geanylatex.tex xhtml -cvalidate -interaction=batchmode

Happend on Debian Testing with Version 3.14159265-2.6-1.40.15

If I fold some peace of code in editor it will be still visible in overview plugin

Geany and Overview are latest, from git (built 1 week ago)

I was using git-new-workdir and recently switched to using the newer form git worktree however found this stops the plugin working.

The reason appears to be that instead of a .git directory in the 'worktee directory' there is a .git file that contains a reference to the original directory. e.g. gitdir: /home/ubuntu/origdir.git/.git/worktrees/newdir.git.

When compiling webhelper, it thorows warnings:

[164/243] c: webhelper/src/gwh-settings.c -> _build_/webhelper/src/gwh-settings.c.1.o
../webhelper/src/gwh-settings.c: In function ‘gwh_settings_install_property’:
../webhelper/src/gwh-settings.c:199:5: warning: ‘g_value_set_char’ is deprecated (declared at /usr/include/glib-2.0/gobject/gvaluetypes.h:179): Use 'g_value_set_schar' instead [-Wdeprecated-declarations]
     HANDLE_BASIC_TYPE (CHAR,    Char,    char)
     ^
[165/243] c: webhelper/src/gwh-plugin.c -> _build_/webhelper/src/gwh-plugin.c.1.o
../webhelper/src/gwh-plugin.c: In function ‘plugin_init’:
../webhelper/src/gwh-plugin.c:424:5: warning: ‘g_thread_init’ is deprecated (declared at /usr/include/glib-2.0/glib/deprecated/gthread.h:265) [-Wdeprecated-declarations]
     g_thread_init (NULL);

Using GTK 2.24.25, GLib 2.42.1

Nuff said. As well as broken tab names and broken GOTOline function.

Hello.

Markdown plugin does not highlight html comments that are used in markdown as well.

Building debugger from git I find that in the debug tab of the message window (F4) there is an empty frame with the title 'command line arguments' . It appears that this should be and editable input.

There is no way to enter command line arguments for the debugger.

Hi!

Would be too difficult to add the same features of autoclose to multi-line comments??
Tentatively, It has two additional complications:

1. Symbol(s) depend on the language you're coding. I guess that could be overtaken getting that info from the filetype files.
2. Usually multi-line comments use more than one symbol. This is more tricky, but maybe autoclose could be checking char by char if the user is trying to insert a block comment...Anyway, some features of Geany/autoclose could still be added: like deleting pairing brace with shift + backspace or moving around with Ctrl + b.

What do you think?

Btw, why is not this plug-in merged into the main Geany code? I don't find any reasons why it should not be.

Overview plugin is not compiling with waf

//.....
Summary: 
Install Geany Plugins 1.25 in            : /usr/local 
Using GTK version                        : 2.24.25 
Using Geany version                      : 1.25 
Compiling Git revision                   : 8f70319 
Plugins to compile                       : overview 
Plugins to skip due to missing dependencies :  
'configure' finished successfully (0.075s)
Waf: Entering directory `/home/frlan/quellen/geany-plugins/_build_'
Traceback (most recent call last):
  File "/home/frlan/quellen/geany-plugins/.waf-1.6.11-30618c54883417962c38f5d395f83584/waflib/Scripting.py", line 97, in waf_entry_point
    run_commands()
  File "/home/frlan/quellen/geany-plugins/.waf-1.6.11-30618c54883417962c38f5d395f83584/waflib/Scripting.py", line 151, in run_commands
    run_command(cmd_name)
  File "/home/frlan/quellen/geany-plugins/.waf-1.6.11-30618c54883417962c38f5d395f83584/waflib/Scripting.py", line 143, in run_command
    ctx.execute()
  File "/home/frlan/quellen/geany-plugins/.waf-1.6.11-30618c54883417962c38f5d395f83584/waflib/Scripting.py", line 345, in execute
    return execute_method(self)
  File "/home/frlan/quellen/geany-plugins/.waf-1.6.11-30618c54883417962c38f5d395f83584/waflib/Build.py", line 98, in execute
    self.execute_build()
  File "/home/frlan/quellen/geany-plugins/.waf-1.6.11-30618c54883417962c38f5d395f83584/waflib/Build.py", line 101, in execute_build
    self.recurse([self.run_dir])
  File "/home/frlan/quellen/geany-plugins/.waf-1.6.11-30618c54883417962c38f5d395f83584/waflib/Context.py", line 128, in recurse
    user_function(self)
  File "/home/frlan/quellen/geany-plugins/wscript", line 254, in build
    bld.recurse(enabled_plugins)
  File "/home/frlan/quellen/geany-plugins/.waf-1.6.11-30618c54883417962c38f5d395f83584/waflib/Context.py", line 112, in recurse
    exec(compile(function_code,node.abspath(),'exec'),self.exec_dict)
  File "/home/frlan/quellen/geany-plugins/overview/wscript_build", line 36, in <module>
    defines = [ subst_vars('OVERVIEW_PREFS_UI_FILE="${PKGDATADIR}/overview/prefs.ui"', bld.env) ]
NameError: name 'subst_vars' is not defined

geanypg - translatable author

 #: ../geanypg/src/geanypg.c:39
 msgid "Hans Alves <[email protected]>"

Is it really needed for author to be translatable?

Just numbers

#: ../scope/data/scope.glade.h:51
msgid "_1"
msgstr "_1"

#: ../scope/data/scope.glade.h:52
msgid "_2"
msgstr "_2"

#: ../scope/data/scope.glade.h:53
msgid "_4"
msgstr "_4"

#: ../scope/data/scope.glade.h:54
msgid "_8"
msgstr "_8"

Tough in some languages, e.g. I found that:

./po/it.po:6062:#: ../scope/data/scope.glade.h:54
./po/it.po-6063-msgid "_8"
./po/it.po-6064-msgstr "_"

Empty formatters

#: ../scope/src/break.c:1250
#, c-format
msgid ", %s"

And others:

#: ../scope/src/debug.c:115 ../scope/src/debug.c:120 ../scope/src/utils.c:37
#, c-format
msgid "%s: %s."

#: ../scope/src/debug.c:598 ../scope/src/thread.c:93
#, c-format
msgid "%s."

#: ../scope/src/utils.c:45
#, c-format
msgid "%s: %s"

#: ../webhelper/src/gwh-settings.c:804
#, c-format
msgid "%s:"

#: ../geanygendoc/src/ggd-file-type-loader.c:871
#, c-format
msgid "%s:%u:%u: %s"

Symbols

#: ../scope/data/scope.glade.h:117
msgid "@"

#: ../scope/data/scope.glade.h:115
msgid "-"

#: ../scope/src/inspect.c:66
msgid "..."

#: ../scope/src/views.c:526
msgid "…"

P.S. Searched using multiline regexp msgid (.+)\nmsgstr \1

There seems to be some kind of race when quickly changing files using e.g. Ctrl+Tab. When pressing Ctrl+Tab several times in a row it appears git-changebar diffs the previous file with the current file thinking it is the same file and shows the diff in the changebar.

This is a bit hard to trigger because normally the switch is too fast for this to happen. I have seen it with the linux kernel project with all the tags loaded to make Geany a little slower but maybe it's enough to have just two really big files open and cycle between them (they would have to be version-controlled of course).

While GeanyPy itself translatable, plugins included in it have no translation strings. It's easy to localize python:

import gettext
_ = lambda x: gettext.ldgettext("geany-plugins", x)

or

from gettext import gettext as _

Upd: parts of GeanyPy written in Python aren't translatable too:

        lbl = gtk.Label("Choose plugins to load or unload:")

I have just built a clean git clone of geany and geany-plugins.
./configure --disable-all-plugins --enable-scope

Using Debug | Setup Program dialogue I attempt to set "Working dir" but it will not accept a dir. In fact this seems to be a FILE open dialogue. I have to select file and click "open", then manually edit off the file name part

Fedora 20, using gtk2 2.24.27

I have not worked with GTK+ recently but it seems the relevant call is :

gtk.FileChooser.set_action()
gtk.FILE_CHOOSER_ACTION_SELECT_FOLDER

https://developer.gnome.org/gtkmm/stable/classGtk_1_1FileChooser.html#ac455a88d080c044c03abb343c46b38db

For some reason the ./waf updatepo script doesn't work. It complains output missing a file. The reason seems to be that (recently?) the files in multiterm are changed from .c to .vala. Simply faking the system by setting a symbolic link for each .c file in this ./multiterm/src directory let the waf script run correctly, like:

ln -s ./multiterm/src/config.vala ./multiterm/src/config.c

But that's a workaround, not a permanent solution, I think.

I want to build geany-plugins from source to apply a patch for cross-debugging but it seems that master has a requirement for geany 1.25 which is not available on my distro ( 1.24 installed ).

I don't see a branch for geany-plugins 1.24

neither do I find a changelog for geany-plugins to see where the 1.25 condition comes from. I suspect I should be able to build the scope plugin with what I have .

Is there no plugins version that is compatible with geany 1.24 ?

Thanks.

When compiling with recent Gtk2, warning appears:

../geanysendmail/src/geanysendmail.c: In function ‘add_stock_item’:
../geanysendmail/src/geanysendmail.c:202:3: warning: ‘gdk_pixbuf_new_from_inline’ is deprecated (declared at /usr/include/gdk-pixbuf-2.0/gdk-pixbuf/gdk-pixbuf-core.h:314) [-Wdeprecated-declarations]
   GdkPixbuf *pb = gdk_pixbuf_new_from_inline(-1, mail_pixbuf, FALSE, NULL);

^

I'm using RHEL 6.6 based OS with Equinox theme. Settings window of overview plugin looks ugly:

P.S. After changing options my current opened document view disappears... This is only on RHEL, others are OK.

A thing I really appreciate from GEdit is its GIT plugin: it allows you to see which lines were added/modified/deleted from the previous commit. In GeanyVCS there is the diff tool, but it isn't as convenient as GEdit's plugin. I think it would be really nice if this feature were added to GeanyVCS.

Geany is an excellent tool but gdb support unfortunately does not support cross-debugging, which is a rapidly growing need. I certainly don't want to be forced to use some ugly mastodont like eclipse just for that.

This patch is very short, so it seems that there is little functionality missing to achieve this. Would it be possible to integrate something like this into geany-plugins?
http://faumarz.blogspot.fr/2014/03/gdb-and-geany-ide-remote-cross-debugging.html

This relies on symlinking the cross debugger to gdb which is a hack. It really needs an additional config for the full name : arm-none-eabi-gdb , or whatever.

Starting the remote gdbserver debugging session could probably be done in the existing project properties dialogue. by defining an 'execute command' . The supplied patch prevents spawning gdb directly. This could be integrated more cleanly, with minimal changes.

Thanks.

At the Addons plugi nREADME there is a "http://" image references even when accessed via https://.
This is the full http:// image URL is hard-coded.

When the plugin Numbered Bookmarks is switched on, Geany crashes instantly on start-up. Running GDB shows the following warning:

Program received signal SIGSEGV, Segmentation fault.
0x00007fffd153f2f4 in SetMarker (sci=0x9fa4a0, bookmarkNumber=1, markerNumber=2, line=247) at geanynumberedbookmarks.c:885
885 fd=GetFileData(document_get_current()->file_name);

I assume that at start-up the call document_get_current() will return a NULL pointer? And asking a file_name member will cause a segfault?

Tested on geany-plugins commit 4c11373 of Nov. 8th, 2015.

When compiling debugger plugin with waf using git c1a1a1d warnings about deprecated functions are printed:

[ 43/243] c: debugger/src/debug_module.c -> _build_/debugger/src/debug_module.c.1.o
../debugger/src/dconfig.c: In function ‘saving_thread_func’:
../debugger/src/dconfig.c:314:2: warning: ‘g_cond_timed_wait’ is deprecated (declared at /usr/include/glib-2.0/glib/deprecated/gthread.h:285) [-Wdeprecated-declarations]
  while (!g_cond_timed_wait(cond, change_config_mutex, &interval));
  ^
../debugger/src/dconfig.c: In function ‘config_init’:
../debugger/src/dconfig.c:461:2: warning: ‘g_mutex_new’ is deprecated (declared at /usr/include/glib-2.0/glib/deprecated/gthread.h:277) [-Wdeprecated-declarations]
  change_config_mutex = g_mutex_new();
  ^
../debugger/src/dconfig.c:462:2: warning: ‘g_cond_new’ is deprecated (declared at /usr/include/glib-2.0/glib/deprecated/gthread.h:281) [-Wdeprecated-declarations]
  cond = g_cond_new();
  ^
../debugger/src/dconfig.c:463:2: warning: ‘g_thread_create’ is deprecated (declared at /usr/include/glib-2.0/glib/deprecated/gthread.h:104): Use 'g_thread_new' instead [-Wdeprecated-declarations]
  saving_thread = g_thread_create(saving_thread_func, NULL, TRUE, NULL);
  ^
../debugger/src/dconfig.c: In function ‘config_destroy’:
../debugger/src/dconfig.c:474:2: warning: ‘g_mutex_free’ is deprecated (declared at /usr/include/glib-2.0/glib/deprecated/gthread.h:279) [-Wdeprecated-declarations]
  g_mutex_free(change_config_mutex);
  ^
../debugger/src/dconfig.c:475:2: warning: ‘g_cond_free’ is deprecated (declared at /usr/include/glib-2.0/glib/deprecated/gthread.h:283) [-Wdeprecated-declarations]
  g_cond_free(cond);
  ^

Using GTK 2.24.25, GLib 2.42.1

With the new proxy plugin support in Geany and eventual update of GeanyPy to use it, we should consider adding support to the build system(s) to allow first-class Python plugins.

A few things that should be considered initially:

• GeanyPy plugins should force --enable-geanypy. Or however the dependency is best cared for.
• They should probably be compiled to .pyc files on installation, similar to GeanyPy's .py files.
• Whether we should have each their directory in $topsrcdir, or just install all the plugins found in GeanyPy's plugins directory.
• Other stuff I'm sure...

I needed to delete geanyprj, overview, and projectorganizer to get everything to compile. They all had local problems in the code.

I recently built libgit2 on Windows. I will publish a ZIP archive soon (AFAIK there are no official Windows binary builds available).

There are only two issues I found so far:

• repository paths contain backslashes:

According to libgit2/libgit2#3012 libgit2 does not support backslashes as path seperator in repository paths. However, the current GCB code constructs the path of the file in the repository from the absolute file path and so on Windows it contains backslashes.
The following diff works for me though there is maybe a more elegant solution. It's a naive approach by simply replacing all backslashes by forward slashes:

diff --git a/git-changebar/src/gcb-plugin.c b/git-changebar/src/gcb-plugin.c
index 57afbcb..4db07b7 100644
--- a/git-changebar/src/gcb-plugin.c
+++ b/git-changebar/src/gcb-plugin.c
@@ -332,8 +332,17 @@ worker_thread (gpointer data)

     if (repo) {
       const gchar *relpath = path + strlen (git_repository_workdir (repo));
+#ifdef G_OS_WIN32
+      GString *repo_path = g_string_new(relpath);
+      /* relpath contains backslashes on Windows but libgit requires repository paths
+       * to be built of slashes, so replace them. */
+      utils_string_replace_all(repo_path, "\\", "/");
+#endif

-      job->blob = repo_get_file_blob (repo, relpath);
+      job->blob = repo_get_file_blob (repo, repo_path->str);
+#ifdef G_OS_WIN32
+      g_string_free(repo_path, TRUE);
+#endif
     } else {
       job->blob = NULL;
     }

• native EOL characters confuse the diff:

On Windows git repositories often use a setting which converts line endings in files to the native line ending characters. This confuses the generated diff and so all lines are marked as changed.
One could change that GIT setting but this could cause further troubles.
The best workaround I found so far is ignoring white space in the generated diff. This however is more than we want here because, AFAIU, this ignores any whitespace changes at the end of the line.
So I don't suggest to do this unconditionally but maybe a plugin setting could be added to let the user choose and a hint that this might be especially useful on Windows.
The option itself:

diff --git a/git-changebar/src/gcb-plugin.c b/git-changebar/src/gcb-plugin.c
index 57afbcb..4db07b7 100644
--- a/git-changebar/src/gcb-plugin.c
+++ b/git-changebar/src/gcb-plugin.c
@@ -547,7 +556,7 @@ diff_blob_to_doc (const git_blob   *old_blob,

   /* no context lines, and no need to bother about binary checks */
   opts.context_lines = 0;
-  opts.flags = GIT_DIFF_FORCE_TEXT;
+  opts.flags = GIT_DIFF_FORCE_TEXT | GIT_DIFF_IGNORE_WHITESPACE_EOL;

   ret = git_diff_blob_to_buffer (old_blob, NULL, buf, len, NULL, &opts,
                                  NULL, hunk_cb, NULL, payload);

I would take care to include the plugin and the libgit2 library in the Windows installer and related stuff.

Strings like This plugin provides or alike should be avoided in plugins description, and most occurrences of the "plugin" word should probably be double checked.

The point is that at the very least in the Plugin Manager any occurrences of the word "plugin" used to refer to the plugin itself is plain redundant and just makes strings longer, less clear and less consistent.

In other places it might be OK to use the word "plugin" to refer to the plugin itself, if it could be confused with something else.

Neither of the Markdown plugin's parser libraries supports tables, mostly because it's not "standard" Markdown, but it would be extremely useful to support this.

@b4n There's a (non-critical) crash in git-changebar:

1. deactivate git-changebar in plugin manager
2. make some change in a version-controlled document and undo it (so redo can be performed)
3. activate git-changebar in plugin manager
4. perform redo in the editor

This is the backtrace I get:

Program received signal SIGSEGV, Segmentation fault.
__strcmp_ssse3 () at ../sysdeps/x86_64/multiarch/../strcmp.S:209
209 ../sysdeps/x86_64/multiarch/../strcmp.S: No such file or directory.
(gdb) bt
#0  __strcmp_ssse3 () at ../sysdeps/x86_64/multiarch/../strcmp.S:209
#1  0x00007ffff5a7ffc9 in g_str_equal () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#2  0x00007ffff5a7f5f0 in g_hash_table_lookup () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#3  0x00007ffff5a9f3c0 in g_quark_from_static_string () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#4  0x00007fffc8ad01ec in update_diff (
    path=0x7fffe539cdc0 <error: Cannot access memory at address 0x7fffe539cdc0>, 
    contents=0x7fffc8cd2400 <G_blob_contents>, data=0xc) at ../git-changebar/src/gcb-plugin.c:814
#5  0x00007fffc8ad046e in get_cached_blob_contents_async (user_data=<optimized out>, 
    callback=0x7fffc8ad01a0 <update_diff>, force=0, path=<optimized out>)
    at ../git-changebar/src/gcb-plugin.c:446
#6  do_update_diff_idle (doc_id=12, force=0) at ../git-changebar/src/gcb-plugin.c:847
#7  0x00007ffff5a90703 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#8  0x00007ffff5a8fce5 in g_main_context_dispatch () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#9  0x00007ffff5a90048 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#10 0x00007ffff5a9030a in g_main_loop_run () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#11 0x00007ffff7076447 in gtk_main () from /usr/lib/x86_64-linux-gnu/libgtk-x11-2.0.so.0
#12 0x00007ffff79c97f1 in main_lib (argc=1, argv=0x7fffffffe1b8) at ../src/libmain.c:1215
#13 0x00007ffff75a8ec5 in __libc_start_main (main=0x4006a0 <main>, argc=1, argv=0x7fffffffe1b8, 
    init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, 
    stack_end=0x7fffffffe1a8) at libc-start.c:287
#14 0x00000000004006ce in _start ()

It saves a heck of a lot of time when you can simply select the top context menu item as the correct spelling, like SublimeText does, or like Firefox does right now as I edit this post. It is annoying to have to run the mouse over the "Spelling suggestion" submenu, especially when the mouse move slips off the menu area. This'd be awesome, thanks. (Also, if anyone can send me a note on how to tweak plugins, I might be able to hack it myself.)

The Overview plugin uses CAIRO_ANTIALIAS_GOOD which is only available since Cairo 1.12.
But on Windows, the latest available binary build of Cairo (at least on gtk.org) is 1.10 and so currently the plugin cannot be compiled (see http://nightly.geany.org/win32/build_win32_plugins_stderr.log).

We could either:
a) require Cairo 1.12 for this particular plugin
b) adjust the code to only use CAIRO_ANTIALIAS_GOOD if available
c) require Cairo 1.12 generally in G-P

c) would actually be possible, I found some binaries on http://sourceforge.net/projects/dia-installer/files/cairo/1.12.16/cairo-devel-1.12.16.zip/download.

I've built geany with gtk3 with

        ./autogen.sh  --enable-gtk3 --disable-html-docs --prefix=/usr
        make 
        sudo make install 

And then installed the plugins via

sudo apt-get install geany-plugins

and yet I don't see those in the plugin manager.
Is there something else I should do ?

geany-plugins-markdown-1.24-3 on fedora 21 is unable to read images from local files with relative directory specification. Some other markdown editors with markdown preview (atom, retext) do that.

When downloading a complete web page with Firefox, the images are downloades and are linked in the markdown file as follows:

the geany plugin behaves like if the image is unreachable. It works however if the link is modified as follows:


The plugin does not highlight all the words that match current word under cursor but only the words which are visible on the screen. During scrolling a document no more highlighted word appears.

With Geany 10a04f71c GeanyLua plugin is not compiling anymore.

 [167/248] cshlib: _build_/automark/src/automark.c.1.o -> _build_/automark/automark.so
 [169/248] cshlib: _build_/commander/src/commander-plugin.c.1.o -> _build_/commander/commander.so
 In file included from ../geanylua/glspi_sci.c:11:0:
 ../geanylua/glspi_sci.h:532:31: error: ‘SCI_SETKEYSUNICODE’ undeclared here (not in a function)
   {"SETKEYSUNICODE", SLT_VOID, SCI_SETKEYSUNICODE, SLT_BOOL, SLT_VOID},
                                ^
 ../geanylua/glspi_sci.h:533:31: error: ‘SCI_GETKEYSUNICODE’ undeclared here (not in a function)
   {"GETKEYSUNICODE", SLT_BOOL, SCI_GETKEYSUNICODE, SLT_VOID, SLT_VOID},
                                ^
 Waf: Leaving directory `/home/frlan/quellen/geany-plugins/_build_'
 Build failed
  -> task in 'libgeanylua' failed (exit status 1): 
    {task 140593153502544: c glspi_sci.c -> glspi_sci.c.2.o}
 ['/usr/bin/gcc', '-DHAVE_CONFIG_H', '-O2', '-g', '-DDEBUG', '-fPIC', '-pthread', '-pthread', '-pthread', '-pthread', '-pthread', '-pthread', '-pthread', '-pthread', '-I/home/frlan/quellen/geany-plugins/_build_', '-I/usr/include/lua5.1', '-I/usr/include/glib-2.0', '-I/usr/lib/x86_64-linux-gnu/glib-2.0/include', '-I/usr/include/gtk-2.0', '-I/usr/lib/x86_64-linux-gnu/gtk-2.0/include', '-I/usr/include/gio-unix-2.0', '-I/usr/include/cairo', '-I/usr/include/pango-1.0', '-I/usr/include/atk-1.0', '-I/usr/include/pixman-1', '-I/usr/include/libpng12', '-I/usr/include/gdk-pixbuf-2.0', '-I/usr/include/harfbuzz', '-I/usr/include/freetype2', '-I/usr/local/include/geany', '-I/usr/local/include/geany/tagmanager', '-I/usr/local/include/geany/scintilla', '-DG_LOG_DOMAIN="GeanyLUA"', '-DGTK', '-DGTK', '../geanylua/glspi_sci.c', '-c', '-o', 'geanylua/glspi_sci.c.2.o']

This seems to be related to a changed API on bundled Scintilla shipped with Geany core. Happens with waf as well as autotools.

Linux: Ubuntu trusty (3.13.0-68-generic / i686)
Geany 1.25 (built on Jul 13 2015 with GTK 2.24.23, GLib 2.40.2)

Installing from Ubuntu repositories (geany v1.23) caused the same problem, so I tried installing from ppa:geany-dev/ppa. Problem remains. All packages starting with "geany" installed, except for geany-plugin-markdown, geany-plugin-scope.

I have tried debugging a python script, (setting debug target to .py file and .pyc) - both gave me "Error laoding file" when I press the Start button

I then tried debugging a bash script (.sh) - same problem.

Then I downloaded a sample .c source file, compiled it and tried to debug it - same problem.

Starting geany from shell in order to try to get some output or stack trace did not help much. All I got was the following: (output on console immediately after pressing "OK" on "Error loading file" dialog)

(geany:30320): GLib-CRITICAL **: g_string_free: assertion 'string != NULL' failed

(geany:30320): GLib-CRITICAL **: g_string_free: assertion 'string != NULL' failed

(geany:30320): GLib-CRITICAL **: g_string_free: assertion 'string != NULL' failed

(geany:30320): GLib-CRITICAL **: g_string_free: assertion 'string != NULL' failed

(geany:30320): GLib-CRITICAL **: g_string_free: assertion 'string != NULL' failed

(geany:30320): GLib-CRITICAL **: g_string_free: assertion 'string != NULL' failed

(geany:30320): GLib-CRITICAL **: g_string_free: assertion 'string != NULL' failed

(geany:30320): GLib-CRITICAL **: g_string_free: assertion 'string != NULL' failed

(geany:30320): GLib-CRITICAL **: g_string_free: assertion 'string != NULL' failed

(geany:30320): GLib-CRITICAL **: g_string_free: assertion 'string != NULL' failed

(geany:30320): GLib-CRITICAL **: g_hash_table_destroy: assertion 'hash_table != NULL' failed

Is there any way I can get a decent stack trace of what caused the "Error loading file"?

GeanyPy is not available on Windows.

Opening this issue to remind myself to look into that.

Here is the description:
https://bugs.launchpad.net/ubuntu/+source/geany-plugins/+bug/1354747?comments=all

I think the issue is in file "debugger/src/dbm_gdb.c" at line 372 when geany tries to execute gdb command "-exec-run &".
I fix it by removing the character "&"

exec_async_command("-exec-run");

gdb doesn't specify something about this
https://sourceware.org/gdb/current/onlinedocs/gdb/GDB_002fMI-Program-Execution.html#index-g_t_002dexec_002drun-2853

Best regards

I have HTML file of size ~400 KB. If SpellCheck plugin enabled and I reload that HTML, Geany hangs after reloading for 4-5 seconds. I unchecked checkbox "check spelling when opening document". It doesn't check after document opening but does that after reloading.

Profiler info:

 \_ sc_speller_process_line (89.6% | 3.6% self)
 | \_ enchant_dict_check (47.0% | 0.7% self)
 |   \_ enchant_pwl_check (44.7% | 0.7% self)
 |     \_ g_hash_table_foreach (31.5% | 9.4% self)
 |       \_ g_strndup (10.5% | 0.5% self)
 |         \_ g_malloc (9.9% | 0.3% self)
 |           \_ malloc (9.6% | 9.6% self)
 |       \_ free (7.0% | 7.0% self)
 | \_ MySpellChecker::checkWord (24.0% | 0.2% self)
 |   \_ Hunspell::spell (22.5% | 1.1% self)
 |     \_ Hunspell::checkword (21.3% | 0.2% self)
 |       \_ AffixMgr::affix_check (20.3% | 0.0% self)
 |         \_ AffixMgr::suffix_check (20.3% | 0.5% self)
 |           \_ SfxEntry::checkword (19.8% | 1.7% self)
 |             \_ AffixMgr::lookup (18.0% | 0.2% self)
 |               \_ HashMgr::lookup (17.9% | 17.5% self)

If Geany Debugger is enabled (from the Plugin Manager) while the window width is less than 700px, Geany crashes.

Incidentally, if Geany Debugger is already enabled, I cannot resize the window to less than 700px wide, which prevents me from being able to use Geany on half my screen. It would be nice if this minimum width requirement could be removed as well.