[Datalab] Duplicated nodes on import spreadsheet

[sheymann]

Bug in the Data Lab when loading nodes and edges spreadsheet

Way to reproduce:

1. Open a new Gephi instance (latest version of Gephi 0.8beta up-to-date on 16/12/11).
2. Create a new Project.
3. Go to the DataLab.
4. Import nodes.csv
5. Import edges.csv (check to create the missing nodes)
   => Duplicated nodes appear (i.e. two nodes of ID=4)

Note: edges.csv have exactly the same nodes that in nodes.csv.

nodes.csv

Id ModularityClass
1 0
2 1
4 2
37 1
3 3
7 2
8 2
29 4
5 2
18 5
6 6
10 7
11 8
14 9
9 7
12 7
13 9
15 10
16 10
19 10
20 10
21 10
17 10
31 5
22 11
23 12
24 12
25 11
26 11
27 13
33 14
28 13
36 14
30 15
34 16
38 4
32 14
39 16
35 16

edges.csv

Source Target
1 2
2 4
2 37
3 7
4 7
4 8
4 29
5 7
5 8
5 18
6 7
7 8
7 10
7 11
8 14
8 18
9 10
9 12
10 12
10 13
11 13
13 14
14 15
15 16
15 18
15 19
15 20
15 21
16 17
16 18
16 19
16 20
17 19
17 21
18 20
18 29
18 31
19 20
19 21
20 21
20 22
20 31
22 23
22 24
22 25
22 26
23 24
25 26
25 27
25 31
25 33
26 27
26 28
27 28
27 36
29 30
29 34
29 38
30 31
30 34
30 37
31 32
31 33
31 34
32 33
32 36
32 39
33 34
33 36
34 35
34 38
34 39
35 39
36 39
37 38
38 39

[eduramiba]

Hi,
This was tricky problem with string ids dictionary.
When I built a node with the GraphFactory in data lab, I did it without string id and then set it with graph.setId (instead of creating it already with that id, and this was the problem).

The GraphFactory uses the int id as string id when a string id is not specified so the call to graph.setId was removing from the dictionary the wrong node.
For example, the third node in the nodes.csv has id 4, so when the next node (the fourth) is added, it removed the previous node from the dictionary, which later could not be found by the importer and have its modularity set.