• last updated 2 hours ago
Constraints
Constraints: committers
 
Constraints: files
Constraints: dates
- don't abbreviate subcommands

- modernize tcl

  1. … 8 more files in changeset.
- fix regression test for cookies (was executed only, when tclwebtest is available, for no reason)

- extend regression test for cookies

- added regression test for client properties

- remove obsolete files

  1. … 4 more files in changeset.
- fixing a wierd interaction between a db_foreach and db_transaction, which leads to wierd values in parameters. Moving from db_foreach to db_list_of_lists repairs the test case.

*** empty log message ***

  1. … 5 more files in changeset.
Merged oacs-5-7 to HEAD

  1. … 167 more files in changeset.
Fixed ailing oracle test ...

case fold before checking if column exists

accept pkey, fkey, and key as constraint suffixes since the pg automatic names use them and are good

Merged oacs-5-6 to HEAD. HEAD now installs and passes all core tests. There

may still be some lingering problems due to my screwing up the oacs-5-6

branch earlier, I'll be looking into this later.

  1. … 214 more files in changeset.
- fix test files__tcl_file_common_errors

- report errors as such in aa-test (wrote warnings into error log, this was ignored since ages)

- fixed naming in acs-tcl, acs-mail-lite and acs-subsite

  1. … 5 more files in changeset.
Added a test for global parameter set/get value.

Fixed up the apm parameter stuff to the point where I can at least

register and unregister, and cache values of, global parameters. There's

some really funky code here, I also took the time to have the unregister

code use the unregister API in PL/[pg]SQL rather than roll its own.

  1. … 7 more files in changeset.
TIP #138 - Add relation_tag to acs_data_links

    • -0
    • +149
    ./application-data-link-procs.tcl
  1. … 7 more files in changeset.
Non-info file merge results, with conflicts fixed

  1. … 726 more files in changeset.
Missed a detail when moving bio back to person.

  1. … 4 more files in changeset.
Moved "bio" back to "person", but with type-specific storage. People can

fight over where such things belong on another day, for now keeping it in

"person" has the advantage that the Tcl API is 100% unchanged so moving it

to type-specific storage won't break .LRN.

  1. … 8 more files in changeset.
Moved "bio" from person/generic storage to user/type-specific storage as TIP'd

(PG only, Oracle to follow soon).

  1. … 6 more files in changeset.
Fixed tests to match the tDOM-based tag closing code.

Make the "old style" tag closing style lower case to be compatible with

the "new style" (tDOM) style, for consistency and so the same test cases

can be used with both. Changed test to match.

  1. … 1 more file in changeset.
Dave's fix for this test, copied over from HEAD. Test now passes.

Use a guaranteed non-existant object_id to test data links procedure

remove old debug statement

specify tclwebtest dependency for tests

Ignore HTML comments when convertng html to text (hides MS Word junk)

  1. … 1 more file in changeset.
Merged from 5.4.2

  1. … 283 more files in changeset.
normalize some more expressions

  1. … 5 more files in changeset.
adding missing close statements

  1. … 3 more files in changeset.
Removed the old test case #3 (and renumbered the remaining ones). The test

checked that updating the cache without sync'ing the children didn't change

the results returned by the get procedure. However the lazy caching code by

Malte changed the semantics of get to automatically load the cache with

children if you try to get the node_id of a node with kids. This is a good

thing, actually, since the update cache w/o sync'ing kids exists to make the

cache update quick. But it did break the test checking explicitly for the old

semantics.

It would be nice, however, if people, after writing new code and noticing that

tests break, would actually sit down and determine if the test or code is wrong

and take appropriate action.

Because I just spent two full hours tracking this down and verifying that the

new semantics invalidate the old test code in a reasonable way, and I'm not the

author of the new code.

fix regression test in cases it fails