"Randall S. Becker" <rsbecker@xxxxxxxxxxxxx> writes: > The filter structure provides a mechanism for providing the > working directory's file name path to a filter using a %f > argument. This request is to teach the textconv mechanism to > support the same capability. > > The use case comes from a complex content renderer that needs to > know what the original file name is, so as to be able to find > additional content, by name, that describes the file (base > name+different extension). > > If this is considered a good idea, I would be happy to implement > this but need a pointer or two of where to look in the code to > make it happen. Both in diff, grep and cat-file, textconv eventually triggers diff.c::fill_textconv() and calls run_textconv() unless there is a cached copy of the resut of running textconv earlier on the same contents. This is because for each textconv driver, the output is expected to be purely a function of the input bytestream, and that is why it does not take any other input. So, if we have two identical blobs in a tree object under different pathnames, making the output from textconv different for them because they sit at different pathnames directly goes against the basic design of the system at the philosophical level. Having said that, I _suspect_ (but not verified) that as long as the driver is marked as non-cacheable, it may be acceptable to export a new environment variable, say, GIT_TEXTCONV_PATH, and allow the textconv program to produce different results for the same input. I am not offhand sure if it is OK to allow command line substitutions like the filter scripts, though. It would be nice from the point of view of consistency if we could do so, but those who use an existing textconv program at a pathname with per-cent in it may get negatively affected. Thanks.
