ich [Thu, 5 Aug 2010 11:25:50 +0000 (11:25 +0000)]
Beginning of the HTML validation of webcit pages. Changing Doctype to Transitional makes it easier to remove errors, since most tables and the like are not yet ready for Strict
* add defines for IGNET parts
* shared_room_add needs to iterate over the available configured remote nodes, as display_ignetconf/aide_ignetconf_section do
* shared_room_add: add ID to the table column, so we can later find it
* shared_room_removal: here I need some help, some little js code to hide the matching column in the add-table.
* call new directory creation function
* use enums for the different setup questions / tasks
* move getenv strings to array
* move most of the getenv things into the sub/subfunction
* change some strings so they're similar to the debian.org questions; we want to share the .po files.
* LogTemplateError(): allow referencing random tokens for the [] at the start of the errormessage.
* GetNextParameter(): return next param via double pointer, so we can mention it in errors.
* NewTemplateSubstitute() follow syntax change of GetNextParameter ()
Art Cancro [Tue, 27 Jul 2010 14:12:07 +0000 (14:12 +0000)]
* Changed some instances of 'struct CitContext *CCC = CC;' to 'struct CitContext *CCC = MyContext();' in order to demonstrate one way in which we might eliminate the CC macro
Art Cancro [Mon, 26 Jul 2010 20:23:11 +0000 (20:23 +0000)]
* When searching instant message logs for transcripts to flush, if it is discovered that at least one party has logged out, flush immediately instead of waiting 5 minutes.
Art Cancro [Fri, 23 Jul 2010 20:23:18 +0000 (20:23 +0000)]
* Rewrote vCtdlLogPrintf() to NOT use any buffers at all, only v*printf() type functions. We do this by writing the log data and metadata in separate calls. In practice, this will cause them to appear on the same line even when other threads are concurrently writing because we are using buffered I/O. Unfortunately, this does NOT fix the server crash that occurs when saving instant message transcripts to disk. It points us in the right direction, though.
Art Cancro [Thu, 22 Jul 2010 14:49:53 +0000 (14:49 +0000)]
* Fixed bug introduced in r8304. CtdlSockGetLine() needs to call FlushStrBuf() at the beginning, just like CtdlClientGetLine() does. This should fix POP3 aggregation.
Art Cancro [Sun, 11 Jul 2010 03:11:26 +0000 (03:11 +0000)]
* Eliminated the C code to shorten long subject lines and add an ellipsis. It doesn't cut off in the right place anyway. Matt put something in the style sheet to let the browser render the ellipsis. It works in Opera and IE, and it'll work in Mozilla soon, degrading to a simple clipping otherwise.
Art Cancro [Sun, 11 Jul 2010 02:34:53 +0000 (02:34 +0000)]
* Replaced normalizeHeaderTable() which I had taken out. Even though the column widths are honoring fixed percentage widths, the header table and the message table display slightly offset due to the presence of the scroll bar. normalizeHeaderTable() fixes that.
Art Cancro [Sun, 11 Jul 2010 02:23:02 +0000 (02:23 +0000)]
* Eliminated the <tbody> from the message summary view. JS now replaces the entire table including the <table></table> tags. FINALLY got the browser to honor fixed column widths with no overflow, keeping the column proportions at 50/30/20 regardless of the width of the browser window.
* I also eliminated the fix_scrollbar_bug div in this view because it is apparently not needed anymore (it was intended to fix a rendering bug in IE5 and IE6). There is no need to wage an aggressive war on this div, but I will remove it from various places when the opportunity arises.
* switch to enum, enums are nice.
* compare_preference() best guess compare two preference sets
* set_preference_backend() ignore config write force, if the new preference value equals the previous.
Art Cancro [Mon, 28 Jun 2010 20:47:38 +0000 (20:47 +0000)]
* Removed z-index manipulation from datepicker. The old behavior was causing the datepicker to pop-under in IE, which is of course not the desired effect.