Difference between revisions of "Bug Reports 2009 - 2013"
Line 4: | Line 4: | ||
Open token taps are still bugged - '''You can neither save them nor provide global tags for them'''. | Open token taps are still bugged - '''You can neither save them nor provide global tags for them'''. | ||
− | We are working at the problem, but | + | We are working at the problem, but until further notice: '''DO NOT WORK WITH TOKEN TAPS''' |
[[[[User:Dorothee|Dorothee]] 14:43, 12 March 2009 (CET)]] | [[[[User:Dorothee|Dorothee]] 14:43, 12 March 2009 (CET)]] |
Revision as of 13:45, 12 March 2009
A creepy old friend ... Open token taps are still bugged - You can neither save them nor provide global tags for them.
We are working at the problem, but until further notice: DO NOT WORK WITH TOKEN TAPS
[[Dorothee 14:43, 12 March 2009 (CET)]]
The Bug List 1. Oh my, long phrases imported from the database into the wiki 'run' out off the browser window. They trail off to the right. :( Is there anything that can be done about that?? You can see what we mean by looking at Documenting Lule Sami. Well ok, it is not a bug really but it looks so incredibly ugly :(
2. In (1) below <:CASE> IS coming before the ILL-SG. Better is: ILL.SG (Kristin) Within the tier annotation of a TCtoken <:CASE> should not appear at all! It is meant as an indication of the grammatical category to which the tag belongs. Information like acc<:CASE should only be available under mousing over tags (Dorothee 12:05, 15 December 2008 (CET))(1)
Hyhto |
hyhto |
cabinGENSG |
N |
sisi |
sisi |
insideILLSG |
Nspat |
manájma | ||
maná | j | ma |
go | PAST | 1PL |
Vitr |
ja |
ja |
and |
CONJC |
jus |
jus |
if |
CONJS |
riekta |
riekta |
right |
ADVm |
de |
de |
then |
CONJS |
oaddát | |
oaddá | t |
sleepINCEP | INF |
Vitr |
galgajma | ||
galga | j | ma |
shall | PAST | 1PL |
AUX |
valla |
valla |
but |
CONJC |
ejma | ||
e | j | ma |
notNEG | PAST | 1PL |
Vtr |
ájn |
ájn |
yet |
ADVtemp |
ájgo |
ájgo |
intendNEG |
PTCP |
The most annoying bug is the following: When one annotates it makes a lot of sense to keep taps of the most recently annotated tokens. However, as soon as one has one token-tap open, the second token one opens cannot be saved anymore. Even if one now starts to close all previously opened token-taps, it remains impossible to save the token that one just had worked on. :( (Dorothee 17:21, 20 December 2008 (CET))
Bugs creeping in with the Lazy Annotation This bug is new and came with the Lazy Annotation Mode: it seems as if saving is not always possible - it is kind of off and on. As was reported to me from one of the TypeCraft classes connected to a course at Dragvoll saving was not possible until the LAM-button was unmarked. Then saving was possible again. Sorry nothing reproducible yet (Dorothee 18:00, 27 November 2008 (CET)) :( I started to work on a long sentence and tried to add construction tags, but could not use that functionality. Pressed button *CHANGE* for construction parameters, but nothing happened, then I also noticed that I cannot save the sentence. That was on the day LAM was introduced. Have not encountered that bug again (Dorothee 18:00, 27 November 2008 (CET))
I guess there was some cook-up having to do with the change of domain that we did around the same time. Yet there is a real bug which arises when one opens a certain number of token tags under annotation. It is only possible to save changes to the last token that was opened. In addition one cannot open the *Change* button that allows one to specify the construction type parameters. (Dorothee 16:55, 15 December 2008 (CET))
Design Suggestions
YOUR SUGGESTION 1. It should be possible to search for glosses and poses. (Kristin)
Fx it be possible to search for, PRES, PAST, IMP, etc. and get to see the different paradigms and the words (from the text) in citation form that belonged into the particular paradigms. (Then it should be possible to put one of the words into the paradigm (show this paradigm for ...(word) - and the paradigm would show the word for all the paradigm instances that is used in the text- without muliplying it when it is used more than once.)
It should also be possible to search for spatial words or temporal words and then get all Nspt and ADVspt - or all Ntmp and ADVtmp - in separated lists - the Ns together with the word they are tied to. All alphabetically with the citation form
Then it should be possible to search for VI and VT and the citation forms should pop up alphabetically.(Kristin)
YOUR SUGGESTION 2. ...............