• last updated 10 hours ago
Constraints
Constraints: committers
 
Constraints: files
Constraints: dates
Moved queries from Tcl files to Xql files

    • -3
    • +4
    /openacs-4/packages/cronjob/cronjob.info
    • -98
    • +94
    /openacs-4/packages/cronjob/tcl/cronjob-init.tcl
    • -74
    • +49
    /openacs-4/packages/cronjob/tcl/cronjob-procs.tcl
    • -0
    • +27
    /openacs-4/packages/cronjob/tcl/cronjob-procs.xql
file cronjob-procs.xql was initially added on branch oacs-4-6.

    • -0
    • +0
    /openacs-4/packages/cronjob/tcl/cronjob-procs.xql
Added missing queries.

    • -21
    • +25
    /openacs-4/packages/acs-tcl/tcl/defs-procs.xql
move custom doc index.adp to www directory to ease merging of 4.6 code

    • -0
    • +7
    /openacs.org-dev/www/doc/index.adp
move back to old index file it ease merging. Custom index file is in opeancs.org/www/doc/index.adp

fixed bug in publishing workflow

Made the configure/customize portal page look like it belongs under the Control Panel tab

Added acs-lang to openacs-4-core

OpenACS update 2002-10-08

  1. … 2799 more files in changeset.
Initial revision

    • -0
    • +24
    /openacs.org-dev/packages/acs-content/www/doc/index.html
  1. … 391 more files in changeset.
fix a couple malformed XML bits

incorrect closing tags

Make sure lang::conn::locale can never return an error.

use permissions API and flush cache on grant or revoke

Fixed, yet again, so it works with PG 7.2. Previous fix was wrong.

merge in bug fix from 4.6 branch

merge in bug fixed from sloanspace for handling select boxes that are required

Fix query name bug. Bug, report and fix by Andrew Spencer. closes: #484

Fix some of the broken docbook links. The fact that the docbook.org people can't manage to have consistent URLs is worrying.

small typo

typo

2 things:

1) Excluded CVS Commits forum from recent postings. The package_id 3061 was hardcoded, so I hardcoded this as well figuring this page was slated as a one-off.

2) Removed the limit form the subquery (fm) and added it to the overall query. This was because the subquery would select messages that were later excluded by the where clauses. There was no messages returned from the query when CVS commits was excluded.

A general question - why was the subquery used as opposed to a straightforward join. If it was for general efficiency, moving the limit probably destroyed that. However, is the straightforward join really that expensive?

Peter: Adding a link from the package version page for replacing occurencies of <#message_key message#> in adp templates with #message_key# and inserting the messages into the message catalog

Peter: added proc apm_package_key_from_version_id

Peter: added instructions on internationalizing adp templates, made the parameter -lozalize examples use tables that will actually fit on a page

Peter: added the trn tag for message key lookups on adp pages. Requires a body with a default message.

Peter: replace occurencies of #package_key.message_key# in adp pages with a message key lookup

Peter: for every request we setup ad_conn locale and ad_conn language using lang::conn (based on user preference and site wide default locale and potentially other things)

Peter: adding the localize switch to parameter::get that will replace any #message_key# occurencies in the parameter value with the corresponding messages from the message catalog using the locale of the request

Peter: adding the -localize switch to ad_parameter