summaryrefslogtreecommitdiffstats
path: root/perl-install/interactive.pm
Commit message (Collapse)AuthorAgeFilesLines
...
* cleaning the utf8 support stuff:Thierry Vignaud2003-01-281-3/+1
| | | | | | | | | | - consolidate check_for_xserver() to check for x11 access - introduce prepare_gtk2() to do what gtk+2 needs, so that tools that're not part of drakxtools (aka: rpmdrake, mcc, ...) can just do : unshift @::textdomains, 'drakconf'; prepare_gtk2();
* strip authentification out of interactive->vnew into common.pmThierry Vignaud2003-01-271-14/+1
| | | | | | | | | | | | | | | | | | | | | | This enable apps that still use gtk+-1.2 via my_gtk to still work when they only need interactive->vnew('su') just to get root capabilities. Indeed, interactive load ugtk if it has access to the X server, which make my_gtk and ugtk fight for the cpu forever... which is bad imho... drakfloppy and net_monitor were converted to use it and are now usuable again ("i leave ... again" commit). it would also enable to complete spec 64 by enabling rpmdrake to use kdesu under kde and consolehelper under gnome. rpmdrake is indeed the last program to not behave like the running desktop to get root capability. unless someone is against this, i'll convert the mcc and all other gtk+ pure tools to do not use anymore interactive just to get root capability. btw, i fixed the infamous SECTOR_SIZE warning...
* (ask_from_list, ask_from_list_, ...): option nocancel addedPascal Rigaux2003-01-241-11/+15
| | | | | (ask_yesorno): no "Cancel" or "Previous" button
* fix non latin1 post-install perl-gtk2 apps, seems like perl upgrading ↵Guillaume Cottenceau2003-01-241-0/+2
| | | | strings to utf8 is somewhat broken (dunno why), so now we bind the codeset or our textdomains to utf8 and tag the translated strings to utf8, when using gtk
* add some documentation about the various possible fieldsPascal Rigaux2003-01-221-1/+13
|
* when the return value of ask_from_ and ask_from is not used, do not have aPascal Rigaux2003-01-211-0/+2
| | | | | "Previous ->" or "Cancel" button (beware, black magic here :)
* fix ask_okcancel in wizard modePascal Rigaux2003-01-071-1/+1
|
* use "if any" instead of "if grep", and various other occurences of "any", ↵Pascal Rigaux2003-01-071-3/+3
| | | | "every", "partition"
* nicer debug messagePascal Rigaux2003-01-071-0/+1
|
* - callbacks "Add", "Modify" and "Remove" return undef when they failPascal Rigaux2003-01-051-3/+7
| | | | | - callback "Add" return the new added element (not handled by interactive emulation, but correctly handled by interactive::gtk native ask_from__add_modify_removeW)
* fix call to method ask_from__add_modify_removeWPascal Rigaux2003-01-031-1/+1
|
* (ask_from__add_modify_remove): check all callbacks are givenPascal Rigaux2003-01-021-0/+2
|
* (ask_from__add_modify_remove): fixPascal Rigaux2003-01-021-7/+4
|
* add ask_from__add_modify_remove, and use it for bootloader entriesPascal Rigaux2003-01-021-0/+26
|
* remove unused variables or rename them with an underscore (eg: $o becomes $_o)Pascal Rigaux2002-12-041-6/+6
|
* "require log" causes some pb, perl thinking that "log" is the log() ↵Pascal Rigaux2002-12-041-2/+2
| | | | function. So replace it with require 'log.pm' (perl_checker will handle this)
* perl_checker adaptationsPascal Rigaux2002-11-271-3/+3
|
* no need to call common::backtrace(), use directly backtrace() (was it ↵Pascal Rigaux2002-11-261-2/+2
| | | | written that way to workaround a bug in old perl_checker?)
* s/wantarray/wantarray()/Pascal Rigaux2002-11-151-1/+1
|
* add/remove spaces to make perl_checker happyPascal Rigaux2002-11-131-1/+1
|
* replace "_" with "N" and "__" with "N_"Pascal Rigaux2002-11-061-11/+11
| | | | | | | | | | | | | | | | | | | rationale: - currently, we use _("xxx") as a shorthand for gettext("xxx"). It also used to call xgettext with --keyword=_ - alas, function &_ is global and not by package (notice esp. that _ is not exported in common.pm) - this lead to big ugly pb with packages defining their own &_, overriding common.pm's &_ - a fix is to set @::textdomains to add a new domain (the default being "libDrakX") but relying on the global "_" is still dangerous!
* - do not handle "format"ing in interactive for combosPascal Rigaux2002-11-051-15/+6
| | | | | | - do it in interactive::gtk instead (that way, "val" is back"format"ed in callbacks as wanted) - die when editable combos are used with non-strings (like arrays or hashes)
* require log will reliably fail, use require 'log.pm' insteadGuillaume Cottenceau2002-09-191-1/+1
|
* add {callbacks}{advanced}, called when "Advance" button is pressedPascal Rigaux2002-08-281-1/+1
|
* when standalone, use the name of the program for the window titlePascal Rigaux2002-08-231-0/+3
|
* remove completly setting 'ok' & 'cancel' since it's better done perPascal Rigaux2002-08-131-2/+0
| | | | | interactive::* to handle correctly the Wizard mode
* button "Previous" and "Next" are not only there in interactive::gtkPascal Rigaux2002-08-131-1/+2
|
* don't switch to 2 buttons dialog box if the labels are too long in ask_from_listPascal Rigaux2002-08-121-10/+12
| | | | | | (otherwise it's ugly (reported for XFdrake multi-head choice with 67 & 58 chars strings) (fix bug #62)
* allow cancel button even for wizard.Francois Pons2002-08-091-1/+1
|
* s/__END__/1;Thierry Vignaud2002-08-051-1/+1
|
* kill "wonderful perl"Thierry Vignaud2002-08-051-4/+1
|
* make new perl_checker happy (and that's not easy!)Pascal Rigaux2002-07-311-3/+3
|
* fix "nicer error message when {val} is not set for ask_from & list" in case ↵Pascal Rigaux2002-07-281-1/+1
| | | | of a reference to a reference (ref is REF instead of SCALAR)
* use fuzzy_pidofs (to detect if kwin is running)Pascal Rigaux2002-07-231-1/+1
|
* - Big Move 1: interactive::* hierarchyThierry Vignaud2002-07-231-6/+6
|
* nicer error message when {val} is not set for ask_from & listPascal Rigaux2002-07-191-0/+1
|
* ease testing of standalone's when non-superuser via setting TESTING=1Pascal Rigaux2002-07-151-1/+1
|
* when not in X and needs su, use consolehelper nowGuillaume Cottenceau2002-02-211-0/+4
|
* use consolehelper when no "kdeinit: kwin" processGuillaume Cottenceau2002-02-211-1/+5
| | | | | is running
* - fix non X + testing + suPascal Rigaux2002-02-161-5/+7
| | | | | | | - since "interactive" takes care of "format" for type "combo", remove "format" (so that interactive_newt which switch to type "list" don't apply "format" once again)
* favour "combo" over "list" in any casePascal Rigaux2002-02-151-1/+1
|
* if testing, don't use kdesuPascal Rigaux2002-02-141-1/+1
|
* fix (causing XFdrake to use a combo instead of a list)Pascal Rigaux2002-02-071-5/+5
|
* since "format" is now handled for combo boxes, combo boxes are used in much ↵Pascal Rigaux2002-02-031-1/+2
| | | | too many places. Fix this by forcing type "list" for some cases, and using type "list" when a ask_from has only one entry
* handle "format" for combo'sPascal Rigaux2002-01-291-6/+22
|
* fix pixel's minimal example using interactive.Thierry Vignaud2002-01-291-0/+2
|
* add a minimal examplePascal Rigaux2002-01-271-0/+6
|
* add "quit_if_double_click" and "tree_expanded" flags (request from till)Pascal Rigaux2002-01-221-0/+2
|
* add feature "allow_empty_list" for "list" entries (disables the special ↵Pascal Rigaux2002-01-221-2/+3
| | | | cases for 0 and 1 element lists)
* toggle Advanced/Basic buttonPascal Rigaux2002-01-031-0/+1
|