Description of problem: current version 20100621 udev script only support 11 devices out of about 121 devices supported by usb_modeswitch Version-Release number of selected component (if applicable): usb_modeswitch-data-20100621-1.fc13.noarch How reproducible: always Steps to Reproduce: perl -lwne 'if (/\QATTRS{idVendor}=="\E(\d+)\Q",ATTRS{idProduct}=="\E(\d+)"/) {print "$1:$2"}' /lib/udev/rules.d/40-usb_modeswitch.rules 0421:0610 0471:1210 0471:1237 1004:1000 1033:0035 1266:1000 1410:5010 1410:5020 1410:5030 1410:5031 1410:5041 ls /etc/usb_modeswitch.d/ | wc -l 121 Actual results: few devices Expected results: more devices (about 121) Additional info: update is trivial updated srpm can be found http://www.ojuba.org/downloads/updates/testing/4/SRPMS/usb_modeswitch-data-20100623-1.oj4.src.rpm the upstream developer said to me <<EOQ You must have looked at an older version of the rules file. The devices are all there in version 20100623. Of course I am using a script to generate the rules file, and yes, the idea is that no interaction is required and every configuration should have a corresponding entry in the rules file. EOQ
usb-modeswitch-data-20100707 which is the latest also does not seem to have the fix. The correct way of resolving this is to contact upstream. let me do that :)
Did anyone bother to really LOOK at the original package? http://www.draisberghof.de/usb_modeswitch/usb-modeswitch-data-20100707.tar.bz2 The file "40-usb_modeswitch.rules" has 14256 Bytes and 398 lines. Remove everything except the device lines and you have 369 lines. Including comment and empty line, each entry takes three lines. Divide 369 by 3 and you get 123 devices. I suggest you check that perl script of yours. If I run it I get ZERO output.
I should have seen that at once: the regular expression in the perl line above manages to exclude all values with non-decimal digits, in effect ignoring the hexidecimal digits a-f. I think you REALLY can close this "bug". Josua Dietze Author of usb_modeswitch