Auto-Merge
The preceding screen shot represents the segmentation generated by the applied file type.
Consider the following TM entry.
If this entry is saved into the TM, and the above asset is re-leveraged against the TM, the following will result.
The two segments have been merged together to improve leverage against the match. Merging and translating these two segments together in another asset resulted in this TM entry being created. This merged segment was created manually, and the auto-merge process was able to effectively leverage against the merged TM entry.
The difference in positions of the placeholders, is enough to cause the failure. The auto-merge is in effect an additive exact match merge. The merged pieces exactly match portions of the TM match.
A merge is generally identifiable in the browser workbench by a discontinuity in the segmentation numbers. Before the merge, the segment IDs are in numerical order. (Often you have to select the show markup options to fully see that the IDs increment in order. Refer back to the original asset illustration.) After a merge, all segment IDs between the first segment and the last segment involved in the merge (including the ID of the last segment) are lost. Notice in the most recent screenshot of the Browser Workbench that text segment #4 no longer exists (see the second column).
This example has illustrated two segments being merged together. The auto-merge process can merge an indefinite number of segments to match up to a super TM entry, provided that no complete matches are found along the way.