Could it be that there are tiny imprecisions in the start/end vertices of the line segments? Perhaps try inserting a Snapper with end point snapping just before the LineCombiner. If that doesn't work, check if the LineCombiner works without the Group By-attributes -- if that's the case, then something is wrong with the Group By-attributes.
I used ESRI ArcMap tool box, a tool called dissolve, that can do the work without any issue. I don't specify snapping. just dissolve line segments base on same attribute values. it can be multiline while they are not adjacent. I am wondering if there is any transformer dedicated for dissolving line segments as same as ESRI's one?
I used ESRI ArcMap tool box, a tool called dissolve, that can do the work without any issue. I don't specify snapping. just dissolve line segments base on same attribute values. it can be multiline while they are not adjacent. I am wondering if there is any transformer dedicated for dissolving line segments as same as ESRI's one?
Aggregator might suffice. It will create aggregates (groups) of lines
I used ESRI ArcMap tool box, a tool called dissolve, that can do the work without any issue. I don't specify snapping. just dissolve line segments base on same attribute values. it can be multiline while they are not adjacent. I am wondering if there is any transformer dedicated for dissolving line segments as same as ESRI's one?
The ArcGIS toolboxes actually apply a tiny amount of "snapping" to all the features, according to the tolerance settings of the feature class. FME does not, which is why it may sometimes be necessary to use a Snapper with a very small tolerance to make sure all the vertices line up perfectly.
The ArcGIS toolboxes actually apply a tiny amount of "snapping" to all the features, according to the tolerance settings of the feature class. FME does not, which is why it may sometimes be necessary to use a Snapper with a very small tolerance to make sure all the vertices line up perfectly.
I have to use ArcGIS toolboxes. I thought FME has everything ArcGIS toolboxes provided.😏
The ArcGIS toolboxes actually apply a tiny amount of "snapping" to all the features, according to the tolerance settings of the feature class. FME does not, which is why it may sometimes be necessary to use a Snapper with a very small tolerance to make sure all the vertices line up perfectly.
The thing to note about comparing FME and ArcGIS is that on the surface, it looks like it is a one to one relationship (i.e one tool in FME does what one tool in AG does) but it isn't. Another thing to note is that FME operations on a feature by feature level (row by row) where as AG generally operates on a feature class by feature class (table by by table) basis. Two very different ways of processing data
The ArcGIS toolboxes actually apply a tiny amount of "snapping" to all the features, according to the tolerance settings of the feature class. FME does not, which is why it may sometimes be necessary to use a Snapper with a very small tolerance to make sure all the vertices line up perfectly.
Apples and oranges, as @hkingsbury is also alluding to. FME and ArcGIS are complementary, not 100% overlapping (what would be the use of that).
The ArcGIS toolboxes actually apply a tiny amount of "snapping" to all the features, according to the tolerance settings of the feature class. FME does not, which is why it may sometimes be necessary to use a Snapper with a very small tolerance to make sure all the vertices line up perfectly.
The ArcGIS toolboxes actually apply a tiny amount of "snapping" to all the features, according to the tolerance settings of the feature class. FME does not, which is why it may sometimes be necessary to use a Snapper with a very small tolerance to make sure all the vertices line up perfectly.
Apples and oranges, as @hkingsbury is also alluding to. FME and ArcGIS are complementary, not 100% overlapping (what would be the use of that).
The ‘use of that’ would be to use FME to automate processes that require both geoprocessing tools and data interoperability among the ever-growing number of software packages that we have to interact with.