User`s guide

E-Prime User’s Guide
Chapter 5: Data Handling
Page 173
When a merge operation is undone, the files and icons return to the status maintained prior to the
merge operation. For example, if an unmerged file is included in a merge operation, the file
would then be classified as “already merged,” and the icon for that file would receive a green
checkmark. If the merge operation is subsequently undone, the file status would return to
“unmerged,” and the green checkmark would be removed from the icon. In addition, the “Merge
operation undone” message would be displayed in the Merge Log.
5.2.4 Conflicts
E-Prime data files were designed to be as flexible as possible. For example, a merged data file
can contain data from different experiments, and those experiments can vary greatly in structure.
Unfortunately, in allowing this flexibility, the possibility of merging two data files that are
incompatible becomes a reality and must be addressed. Other issues of concern include merging
duplicate data into the target file and handling file errors (i.e., a file cannot be opened or read)
during the merge process. The following section addresses these issues.
5.2.4.1 Variable/Log-Level Names and Data Types
Two basic rules apply to all data files:
1. The same name may not be used for both a variable and a log-level, or for two different
log-levels within a single file.
2. A single variable cannot have more than one data type in the same file.
If one of these two rules is violated, the two files (i.e., the source file and the target file) are
considered incompatible, and E-Merge cannot merge them. Fortunately, these incompatibilities
should rarely occur.
A violation of the first rule would only occur when merging experiments having different naming
schemes for log-levels, under the condition that E-Merge could not resolve the differences without
creating a conflict. For example, a data file with log-level 2 named “Sections” and log-level 3
named “Questions,” is being merged into a target file with log-level 2 named “Blocks” and log-
level 3 named “Trials.” During the merge operation, E-Merge would try to merge the data by
renaming “Sections” to “Blocks” and “Questions” to “Trials” in the target file. However, if the data
file being merged into the target file happened to also contain a variable called “Trials,” this would
result in a target file that had both a variable and a log-level with the same name. Or, if the
source file had a different log-level named “Trial,” this would result in a target file that had two log-
levels with the same name. E-Merge cannot resolve these conflicts, and therefore cannot merge
these files.