GNOME Bugzilla – Bug 344848
epiphany fails to configure even if --disable-spell-checker is invoked
Last modified: 2006-06-19 13:51:25 UTC
Please describe the problem: I don't know why there is a spellchecker support in a browser but nevertheless the cvs version fails to build even if i ./configure with either --disable-spell-checker or --enable-spell-checker=no My xulrunner is compiled with spellchecker and epiphany still wants 'enchant' ? So we have a perfectly useable spellchecker in mozilla but epiphany want's to install yet another spellchecker ? The dependencies are going crazy ! Steps to reproduce: 1. Attempt to do a ./configure --disable-spell-checker Actual results: configure fails Expected results: configure succeeds and build is succesful Does this happen every time? yes Other information: 1. If enchant is mandatory, please remove the misleading option in the configure file. 2. Why is there a spellchecker in a browser ? 3. Why not use the standard mozilla/seamonkey spellchecker ?
Example output % ./configure --with-gecko=xulrunner --disable-spell-checker --enable-certificate-manager --enable-spell-checker=no checking for a BSD-compatible install... /bin/install -c checking whether build environment is sane... yes checking for gawk... gawk checking whether make sets $(MAKE)... yes checking whether to enable maintainer-specific portions of Makefiles... no checking build system type... i686-pc-linux-gnu checking host system type... i686-pc-linux-gnu checking for style of include used by make... GNU checking for gcc... gcc ...[stuff deleted]... checking whether to enable C++ RTTI... no checking whether we have a gtk 2 gecko build... yes checking whether we have a gecko debug build... no checking for gecko version... 1.9 checking for GECKO... checking whether we can compile and run XPCOM programs... yes checking whether typeaheadfind is broken... no checking nsIX509Cert.h usability... yes checking nsIX509Cert.h presence... yes checking for nsIX509Cert.h... yes checking whether to include PSM support... yes checking whether to enable native gtk+ filepicker... no checking nsINonBlockingAlertService.h usability... yes checking nsINonBlockingAlertService.h presence... yes checking for nsINonBlockingAlertService.h... yes checking nsIGlobalHistory3.h usability... yes checking nsIGlobalHistory3.h presence... yes checking for nsIGlobalHistory3.h... yes checking nsIIOService2.h usability... yes checking nsIIOService2.h presence... yes checking for nsIIOService2.h... yes checking nsIMutableArray.h usability... yes checking nsIMutableArray.h presence... yes checking for nsIMutableArray.h... yes checking mozISpellCheckingEngine.h usability... yes checking mozISpellCheckingEngine.h presence... yes checking for mozISpellCheckingEngine.h... yes checking for the @mozilla.org/permissionmanager;1 XPCOM component... yes checking for the @mozilla.org/PopupWindowManager;1 XPCOM component... yes checking for the @mozilla.org/cookie/permission;1 XPCOM component... yes checking for the @mozilla.org/permissions/contentblocker;1 XPCOM component... yes checking X11/XF86keysym.h usability... yes checking X11/XF86keysym.h presence... yes checking for X11/XF86keysym.h... yes checking whether Python support is requested... autodetect checking for a Python interpreter with version >= 2.3... python checking for python... /usr/bin/python checking for python version... 2.4 checking for python platform... linux2 checking for python script directory... ${prefix}/lib/python2.4/site-packages checking for python extension module directory... ${exec_prefix}/lib/python2.4/site-packages checking for headers required to compile python extensions... found checking whether we can build a shared library depending on libpython... yes checking for PYGTK... checking for pygtk defs... /usr/share/pygtk/2.0/defs checking for pygtk codegen... /usr/bin/python /usr/share/pygtk/2.0/codegen/codegen.py checking for pygtk h2def... /usr/bin/python /usr/share/pygtk/2.0/codegen/h2def.py checking whether gcc understands -fno-strict-aliasing... yes checking whether to build the deskop file plugin... yes checking for DESKTOP_FILE_PLUGIN_DEP... checking whether to enable zeroconf bookmarks support... yes checking whether to build the certificates manager... yes checking whether to enable NetworkManager support... no checking whether spell checking support is requested... no checking for SPELLCHECKER... configure: error: Package requirements (enchant >= 1.0) were not met. Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively you may set the SPELLCHECKER_CFLAGS and SPELLCHECKER_LIBS environment variables to avoid the need to call pkg-config. See the pkg-config man page for more details.
(In reply to comment #0) > I don't know why there is a spellchecker support in a browser For spell checking form input fields. > but nevertheless > the cvs version fails to build even if i ./configure with either > --disable-spell-checker > or > --enable-spell-checker=no This should be fixed in cvs now. > Other information: > 3. Why not use the standard mozilla/seamonkey spellchecker ? Because it sucks. It's just a backend, wich an insufficient interface to provide our planned UI, and it uses an internal copy of myspell, hardly what I would call the best spell checker available.
> For spell checking form input fields. How cool is that. I can now spellcheck my name and address in input forms !! > This should be fixed in cvs now. Thanks > Because it sucks. It's just a backend, wich an insufficient interface to > provide our planned UI, and it uses an internal copy of myspell, hardly what I > would call the best spell checker available. Someone should tell the mozilla/seamonkey team then and log and rfc to change it. Why not submit the patches into their trunk instead of adding this into your local tree. Just a suggestion.
Reopening since this is not fixed. I did a cvs checkout and configure today and it doesn't appear to be resolved: % rm -f configure % ./autogen.sh % ./configure --with-gecko=xulrunner --disable-spell-checker --enable-spell-checker=no But still end up in checking whether to build the deskop file plugin... yes checking for DESKTOP_FILE_PLUGIN_DEP... checking whether to enable zeroconf bookmarks support... yes checking whether to build the certificates manager... no checking whether to enable NetworkManager support... no checking whether spell checking support is requested... no checking for SPELLCHECKER... configure: error: Package requirements (enchant >= 1.0) were not met. Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively you may set the SPELLCHECKER_CFLAGS and SPELLCHECKER_LIBS environment variables to avoid the need to call pkg-config. See the pkg-config man page for more details.
Should be really fixed now, was a wrong test statement.
Confirmed