• last updated 15 mins ago
Constraints
Constraints: committers
 
Constraints: files
Constraints: dates
remove "untested" comment

clean up templates, improve context bar handling. still more to do.

*** empty log message ***

- removed obsolete upgrade directory

fixes rdf:Seq since the rdf parser complains about unqualified resource tags

change colors for openacs.org

fix essay answers

fix responses-export for postgresql and remove unused subselect for oracle.

    • -5
    • +5
    /openacs-4/packages/survey/survey.info
clean up del/remove functions

add a no-responses page for CSV export

file no-responses.adp was initially added on branch oacs-4-6.

    • -0
    • +0
    /openacs-4/packages/survey/www/admin/no-responses.adp
fix survey-copy query

remove unused xql file

error v_section_row used rather than v_question_row bug 1790/patch 495

removing old files

moving from dotlrn.openforce.net:/dotlrn-cvsroot/new-portal to

opeancs.org:/cvsroot/openacs-4/packages/portal

  1. … 130 more files in changeset.
Initial revision

    • -0
    • +154
    /openacs-4/contrib/packages/portal/portal.info
  1. … 130 more files in changeset.
I removed the composite index that checks for uniqueness of the community_type

and supertype pair. This index broke installation if the Oracle blocksize is

less than 8KB (which is true by default on at least some platforms).

The unique constraint was unnecessary since community_type was already declared as the

table's primary key. Because of this, only one row with a given community_type name

is possible, which means the composite key (community_type, supertype) is guaranteed

to be unique.

Removing the index will also speed community creation slightly, save a bit of

disk space, etc etc.

Note that the composite index was useless for lookups on the supertype alone

anyway (since community_type was the leftmost member of the composite key), and that

any queries on the type hierarchy ought to use the denormalized table defined

by acs-objects-metadata.sql anyway.

So the index seems entirely useless to me.

Made -export ignore variables that aren't defined ala the export_vars

proc

updating owner and vendor of acs-lang package

improving context bar handling, master template, and other minor improvements

  1. … 4 more files in changeset.
the catalog check script wasn't checking adp lookups properly, typo, improved indentation

I18N: converting the forums application to using the message catalog

    • -1
    • +4
    /openacs-4/packages/forums/tcl/forums-procs.tcl
    • -10
    • +15
    /openacs-4/packages/forums/www/forum-view.adp
    • -7
    • +11
    /openacs-4/packages/forums/www/index.adp
    • -1
    • +4
    /openacs-4/packages/forums/www/index.tcl
    • -6
    • +9
    /openacs-4/packages/forums/www/message-chunk.adp
    • -2
    • +5
    /openacs-4/packages/forums/www/message-chunk.tcl
    • -2
    • +5
    /openacs-4/packages/forums/www/message-email.adp
  1. … 27 more files in changeset.
Fixed bad primary key on host_node_map table. Note: Fixes to Oracle data model have not been tested.

Got rid of references to acs_object_id_seq view in apm-create, which were causing some errors

1. Minor wording change to the static pages admin index template

2. Made ad_form_new_p require a key name since guessing's arcane and

unreliable

The page name wasn't properly quoted before getting sent to the dimensional proc

Append to header_stuff instead of overriding it. Merged in change from 1.9.2.2 to 1.9.2.3. Or tried to.

    • -3
    • +35
    /openacs-4/packages/dotlrn/www/dotlrn-master.tcl
Append to header_stuff instead of overriding it

    • -2
    • +6
    /openacs-4/packages/dotlrn/www/dotlrn-master.tcl
updated the I18n instructions to point to the script for checking catalog files