Menu
▾
▴
#924 Incremental translate command line.
We use the commmand line a lot to process translation projects. In fact, we continuously process the same projects over and over, but in general maybe 5 or 10% of the projects change between two runs. Processing takes approximately 10 minutes on a high end server.
It would be great when OmegaT command line processing would work similar to Ant compiling Java files: does the target file still exist? Did the source file change after the target file? Did one of the "transformations" (tmx) change after the target file? Then generate target file again. If not: do nothing.
This would save a lot of processing time (and joules of energy).
Comment: the same also holds of course for interactive use, but I do not know whether interactive users would benefit a lot when less target files are written to disk. In general I find OmegaT rather quick when you have only a few tens of files.