All generator output is appended to the current output stream. By default, the stream just goes to a window, whose result will be shown to the user at the end of the generation. It is also possible to direct the output to files (see Section 6.4) or variables (Section 6.5). Temporary streams are also used by some commands to form their arguments, as we will see shortly.

In our example case, we can simply forward the output to a file by using the filename...write...close command:

01 SpecSheet()
02 _translators()
03 filename id%file '.txt' write
04   foreach >Watch
05   {
 .
 .
 .
22   }
23 close

Everything between filename and write goes to make up the name of the file; everything between write and close is written to that file. The name of the file here in line 3 is thus the value of Graph’s identifier property (i.e. ‘WatchModels’), translated to be suitable for a filename with the %file translator, and followed by ‘.txt’. After the write command, all output will be collected to go into this file until the close command at the end of the generator (in line 23), which closes the file stream and actually performs the write to disk.

The example above collects the spec sheets for all Watch models into one file. In order to have a separate file for each model, we can simply move the filename...write...close command inside the main foreach loop:

01 SpecSheet()
02 _translators()
03 foreach >Watch
04 { filename id%file '.txt' write

 .
 .
 .
22   close
23 }

Instead of one ‘WatchModels.txt’ file we will now get ‘Ace.txt’, ‘Delicia.txt, ‘Sporty.txt’ etc. All files are generated into the default directory but we can also prefix the filename with directory information, if needed (the sep command outputs the platform’s directory separator, ‘\’ or ‘/’):

filename 'reports' sep id%file '.txt' write

MetaEdit+ also provides a pre-defined subgenerator that outputs the default relative path for the reports directory created by the installer. Using this subgenerator, the above piece of code would look like this:

filename 
subreport '_default directory' run 
  id%file '.txt' 
write

The filename can thus be constructed using any MERL commands between filename and write: all their output goes to build up the filename. You can also override the existing ‘_default directory’ subgenerator or write a new one that suits your purposes. Alternatively you can store a directory into a variable for further reference:

$myDir = 'reports\'
filename $myDir id%file '.txt' write