Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

History of tests/i18n.test

Simplified the master file of the included test suite, mostly by removing not longer needed cruft. And other minor test suite gardening. file: [3e754eb159] check-in: [fa94139e03] user: rolf branch: trunk, size: 2992
The package name is tDOM, but it always has been requested by [package require tdom] and now the scripted helper commands in tdom.tcl are also in the namespace tdom (not anymore in tDOM). The new pullparser command is now also in this namespace. There are aliases from the old command names to the new one, so there must be nothing done; old scripts will run as they did. It's just, that you in new code don't have to write serveral upcase letters in a row because of tDOM. file: [a204756237] check-in: [afde497780] user: rolf branch: trunk, size: 3028
Removed all that TCL_UTF_MAX cruft, that tries to do the right thing. This is all a dubious hack without proper core support. Until we have that (if ever) it's the responsibility of the user to not mix core and binary extension build with different TUM. file: [085cc18a88] check-in: [549b689298] user: rolf branch: trunk, size: 3028
More tests in this area. file: [f5493417c3] check-in: [0061b52914] user: rolf branch: beyondBMP, size: 3279
Handled this mess even further. New dom method isBMPCharData, to check, if a string has only characters out of the BMP. With TCL_UTF_MAX > 3 even 4 byte utf-8 character will be serialized as character entities only with -escapeNonASCII. With TCL_UTF_MAX = 3 (the default of the core) 4 byte utf-8 character will always be serialized als character entities, even if -escapeNonASCII is not given. If tDOM was build against a core with other TCL_UTF_MAX then the interpreter, which [load]s it, it complains and raises error. New dom featureinfo subcommand TCL_UTF_MAX, which returns the TCL_UTF_MAX value of the tcl core, tDOM was build with. file: [6c74e49ce5] check-in: [ab3fea493c] user: rolf branch: beyondBMP, size: 2872
Some more minor modifications to make tdom handle 4 byte utf-8 characters right (if one gets them into a dom tree, by character entities within the xml or using parsefile). Still, accessing such pcdata beyond BMP with dom methods or xpath queries and using the result at script level will not work correctly with current tcl default build (mangled results without notification by error). file: [edfec94ecd] check-in: [57d6c144cc] user: rolf branch: beyondBMP, size: 1705
Don't panic while serializing 4 bytes utf-8 chars, just write them as character enities. file: [02ee3d545b] check-in: [24d35f6835] user: rolf branch: beyondBMP, size: 1470
Made test suite also running under tcl8.0.5. Added constraint to the tests, which uses i18n tcl features. file: [ad68655284] check-in: [3bafc24e57] user: rolf branch: trunk, size: 1268
Added RCS string to comment file: [ff76ec3dbd] check-in: [e66725042a] user: zoran branch: trunk, size: 1233
Added the inclusion of loadtdom.tcl and adjusted baseURI / load of external files where necessary. file: [e5e3d9b5b0] check-in: [0c74c1fded] user: rolf branch: trunk, size: 1139
Corrected strange constructs [$doc documentElement $doc]. (Problem found bei Zoran.) file: [3e30e68555] check-in: [83d8ed76cc] user: rolf branch: trunk, size: 996
tcltest test suite started. The expat related tests are mostly out of an older tclexpat test suite by Steve Ball (reviewed, bug fixed and adapted to tDOMs SAX interface). Plus a few tDOM specific tests. file: [78d8becc37] check-in: [8ba496bb01] user: rolf branch: trunk, size: 1011 Added