Thank you, Jan. Your answer is work, but my question may not clear.
1. How to find max depth of the graph
LET maxLevel = MAX_DEPTH(table1)
2. For every node visit,
If current node has no children then
LET currentLevel = LENGTH(path.vertices)
LET rowspan = COUNT_CHILDREN(node) == 0 ? maxLevel - currentLevel : 0
If current node has children then
LET colspan = COUNT_CHILDREN(node) > 0 ? COUNT_LEAF_NODE(node) : 0
Example
| A | | | | | |
|---|---|---|---|---|---|
| B | | | | C | |
| D | | E | | F | G |
| H | I | J | K | L | |
LEAF_NODE(A) = [H, I, J, K, L, G] = 6
LEAF_NODE(B) = [H, I, J, K, L] = 4
LEAF_NODE(C) = [L, G] = 2
1. Can AQL Count leaf nodes?
2. Can AQL find max depth of graph?
เมื่อ วันพฤหัสบดีที่ 16 มีนาคม ค.ศ. 2017 15 นาฬิกา 35 นาที 00 วินาที UTC+7,
Jan เขียนว่า:
>
> Hi,
>
> one possible and rather generic solution is to run the actual query in a
> subquery, calculate its number of rows and then return each row from it
> augmented with rowspan and colspan.
>
> Example:
>
> /* this is a template for an empty row */
> /* it needs to contain all the attributes to be returned from the actual
> query, plus default values should an attribute not be present */
> LET empty = { _id: null, _key: null, name: null }
>
> LET keepColumns = ATTRIBUTES(empty)
> LET colspan = LENGTH(keepColumns) + 3 /* LENGTH([ "row", "colspan",
> "rowspan" ]) == 3 */
>
> LET result = (
> /* actual query goes here, e.g. FOR doc IN collection RETURN doc */
> )
>
> LET rowspan = LENGTH(result) /* number of rows in result */
>
> /* iterate over result and emit it */
> FOR i IN 1..rowspan LET row = result[i - 1]
> RETURN MERGE(empty, KEEP(row, keepColumns), { row: i, rowspan, colspan })
>
>
> Best regards
> Jan
>
> Am Donnerstag, 16. März 2017 04:11:53 UTC+1 schrieb Frank Russell:
>>
>> I want to save html table header as graph. My question is
>>
>>
>> 1. Is it possible to calculate rowspan and colspan using only AQL?
>> For example, using edge attribute.
>> 2. How to do numbering for final result? For example, {"index": i + 1}
>>
>>
>> Attachment:
>>
>> - result.json : The final result json should look like.
>> - columns.json: All columns (Nodes)
>> - table1-columns.json: Table1 columns (Edges)
>>
>>
>> table1:
>> Graph({edges: "table1-columns", fromCollections: "columns", toCollections
>> : "columns"})
>>
>>
>>
>> My pseudo query:
>> LET index = 0
>> FOR v,e,p IN 1..10 OUTBOUND "columns/0"
>> GRAPH "table1"
>> /* FILTER p.vertices[2].id != 2 */
>> index = index + 1
>> SORT LENGTH(p.vertices)-1,v.id
>> RETURN {
>> "index": index,
>> "id": v.id,
>> "title": v.title,
>> "row": LENGTH(p.vertices)-1,
>> "root": p.vertices[1].id,
>> "rowspan": rowspan,
>> "colspan": colspan
>> }
>>
>>
>>
>> Thank you
>>
>
--
You received this message because you are subscribed to the Google Groups
"ArangoDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
