2008-06-04 Behdad Esfahbod Released vte-0.16.14. * NEWS: Updated. * configure.in: Bumped version to 0.16.14. Bumped libtool version to 11:18:2. 2008-06-04 Behdad Esfahbod Bug 536632 – vte build failure in ring.c:210: error: expected expression before 'do' * src/debug.h: * src/ring.c: * src/ring.h: With latest glib (after bug #519026), g_error is expanding to "do {...} while(0)" instead of "{...}". This breaks build in debugging mode. Use g_critical instead. Also add some G_STMT_START/END to some macros. 2008-05-29 Chris Wilson * src/pty.c (merge_environ): Replace previous commit with the better patch by Christian Persch. 2008-05-29 Chris Wilson * src/pty.c (merge_environ): Fix bug in patch to use g_listenv() as it only returns the variable name and we need to use g_getenv() to retrieve its value. 2008-05-27 Behdad Esfahbod * src/iso2022.c (_vte_iso2022_ambiguous_width): Oops. Fix thinko in last commit. 2008-05-27 Behdad Esfahbod Bug 535022 – ambiguous width in utf8 locale * src/iso2022.c (_vte_iso2022_ambiguous_width): Recognize env var settings VTE_CJK_WIDTH=narrow and VTE_CJK_WIDTH=wide. Any other value means "auto" based on locale, as it was doing previously. 2008-05-21 Chris Wilson * src/pty.c (merge_environ): Silence compiler warning about using environ as a local variable name. 2008-05-21 Richard Hult Bug 534148 – Use g_listenv() instead of environ * src/pty.c: (merge_environ): Use g_listenv() instead of non-portable environ extern. 2008-05-19 Kjartan Maraas * configure.in: automake doesn't like to substitute more than one variable at a time. * src/vtedraw.c: (_vte_draw_get_using_fontconfig): * src/vtetc.c: (_vte_termcap_find_boolean): Fix two compiler warnings. 2008-05-09 Chris Wilson * src/vtedraw.c (_vte_draw_text): Add the text to draw to the debug output. 2008-05-09 Chris Wilson Fix "GLib-CRITICAL **: g_io_add_watch_full: assertion `channel != NULL' failed" * src/debug.c (_vte_debug_parse_string): * src/debug.h: * src/vte-private.h: * src/vte.c (vte_terminal_emit_adjustment_changed), (_vte_terminal_adjust_adjustments_full), (vte_terminal_scroll_lines), (vte_terminal_maybe_scroll_to_bottom), (_vte_terminal_insert_char), (vte_terminal_catch_child_exited), (mark_input_source_invalid), (_vte_terminal_connect_pty_read), (mark_output_source_invalid), (_vte_terminal_connect_pty_write), (_vte_terminal_disconnect_pty_read), (_vte_terminal_disconnect_pty_write), (_vte_terminal_fork_basic), (vte_terminal_eof), (_vte_terminal_enable_input_source), (vte_terminal_io_read), (vte_terminal_handle_scroll), (vte_terminal_finalize), (vte_terminal_class_init), (vte_terminal_set_pty), (process_timeout), (update_repeat_timeout), (update_timeout): It was possible for _vte_terminal_enable_input_source() to be called after the input closed with G_IO_HUP. The minimal change would just have been to add a guard to check the pty_master was still valid before reattaching the source. Instead I removed the redundant duplication of the input and output channels and added lots of debugging. 2008-05-08 Chris Wilson Bug 516869 – vte displays nothing on GTK+/DirectFB Original patch by Jérémy Bobbio. * src/vte.c (vte_terminal_class_init), (update_regions), (update_repeat_timeout), (update_timeout): gdk_window_process_all_updates() is ineffective on the gtk+-directfb and gtk+-quartz backends, so in order to process updates immediately one must call gdk_window_process_updates() on the desired GdkWindows instead.