Issues

List of pending features and bugs.

Back end

1. (1.0) Parsing of rules need to be improved. IAN and EUGENE were accepting rules with unbalanced parentheses. There is also a problem of an extra comma in the rules. The sensitivity of syntactic check of the Engines should be higher. Eugene and IAN must be sensitive to the following syntactic error:

(%a,A,,B,C):=(%a,+E);

2. (1.0) Eugene and IAN should reject the wrong UTF-8 encoding. From the perspective of the user, the rule was perfect, and the string was clearly and correctly displayed; but the machine was replacing it by empty.

3. (1.1) The rules leading to impossible graphs are working. The example below is generating an impossible graph.

The rule (NB(N,%n;JB(%j;%j2),{and|or},%adjc),%m):= (JB(%j;%j2),rel=%adjc) (NB(N,%n;%j),rel=%m)(NB(N,%n;%j2),rel=%m);

And it's putting the same node %j in two different positions in the node list. This should not be possible. A node cannot be inside two different nodes in a list structure.

4. (1.0) In Eugene, after range, if we try to process another sentence, we get the message “Another process is still running. Please stop it first”. But the process is already over, and there is no button to stop it anyway. We have to reload the whole page in order for the system to work.

5. (1.1) A module for preprocessing is needed in IAN. It will serve for sentence segmentation and morphological preprocessing.

Front end

1. (1.1) To replace the "Welcome" tab in EUGENE and IAN. It is more interesting to have any instructions informed in a help button.

3. (1.0) To create the "text editor" view for dictionary entries and grammar rules (i.e., the user should have the opportunity to treat these files as when they are introduced in the manual input);

4. (1.0) After editing a given dictionary entry or grammar rule, the system should remain in the same page (it currently goes back to the first page);

5. (1.1) To include the possibility of using "drag-and-drop" to reorder dictionary entries and grammar rules (in addition to the current one);

6. (1.0) To create the possibility of having "test sets". The user should have the possibility of flagging a given sentence as correct (i.e., the grammar and dictionary would work for it). After that, the sentence would be automatically included in a "test set", to be defined by the user, and stored as a reference. The user would have the possibility of checking the consistency of the grammar and dictionary at any time by running this test set (but without the need to analyze the individual outputs for the sentences, because the machine would do that automatically, and would highlight all the sentences for which the results would be different than the ones defined as correct).

7. (1.0) The loaded unl sentence issue after logout/loging with different user. Sometimes a wrong unl document/sentence is loaded, when a user is changed within the same session.

8. (1.0) In the dictionary and rule editor. The checkbox at the left should be used to select entries or rules, and the operations (edit, delete, clone, disable/enable) should appear as individual options right after them (and not at the rightmost position in the line), or as multi-entry options at the end of the page. It is recomended to use the same approach and icons used, for instance, in other DBMS such as phpMyAdmin, SQLyog, OracleSQL and others. The thread [1] brings a screenshot of what to be the ideal interface for the editor.

11. (1.1) The trace must be revised to be more intuitive.

12. (1.0) The disambiguation rules should be enabled/disabled in one single place, instead of two. Now they can be turned on/off both in the EUGENE/IAN tabs by a checkbox and loaded/unloaded from the editor.

13. (1.0) The "create" button is confusing. Users are cloning rules instead of saving them. The same for the check box at the left of dictionary and grammar entries. People are using them to select rules for edition rather than for deletion.

14. (1.0) After cloning a rule or dictionary entry, the system goes back to page 1 (instead of remaining in the same page).

15. (1.0) It’s not possible to delete shared dictionaries (although it’s possible to delete shared rule-sets)

16. (1.0) People have been pressing the button “delete dictionary” instead of “delete selected entries”. We do need to put the buttons for files and entries in different places.

17. (1.0) To include a link to EUGENE in IAN's welcome tab, and link to IAN in EUGENE's welcome tab.