 |
 |
Hello Chris, the python copy API is an interface to the CopyProcess C++ class not `xrdcp`. CopyProcess accepts a list of full (absolute) and possibly unrelated URLs. `xrdcp` on the other hand accepts only two URLs, even in recursive mode. `xrdcp` checks if one or both of them are directories and acts accordingly building and supplying CopyProcess with a list of final URLs (CopyJobs), sometimes these are entire subtrees. Since the copy jobs within the CopyProcess don't have to be related in any way, moving this scanning functionality to CopyProcess would result with at least one extra remote stat per CopyJob. This is too expensive. Cheers, Lukasz --- Reply to this email directly or view it on GitHub: https://github.com/xrootd/xrootd-python/issues/5#issuecomment-46448601 ######################################################################## Use REPLY-ALL to reply to list To unsubscribe from the XROOTD-DEV list, click the following link: https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1