2014-06-16 17:01 GMT+02:00 Andrew Grieve <[email protected]>: > I think this behaviour has been around for a while, and makes sense in the > majority of cases.
Yep, I did a "git blame" and this fragment of code was there from the beginning. > Best practice is to download to a temporary location, > and then upon success move the file to its final spot. > Yes, this is what I will have to do at the end. The thing is that I want to avoid a double step process. Anyway I'm still thinking that a hard coded and undocumented behaviour is not a good practice neither, mostly because there's the tools for the developer to act as he needs with the success or error callback. It's about flexibility. > > That said, I think it'd be fine to add an option for not delete on error. > It seems like the file transfer download is oriented to use a temporary directory so it's fine for me as is (keep it simple). I think that an explanation in the plugin documentation should be great so everyone can figure how to use the download. About the issue CB-6928, my patch it's not valid for me as I will use the temporary directory now and probably for everyone so I will have to cancel the pull request. The problem is still there but I wonder how can we fix it if the success callback argument is an "Entry" object, the ideal should be to be able to communicate the caching status. I will change the patch to use the error callback to communicate the caching status, it's not an error but I don't see other way. I will add also the documentation for the new caching status code and open a new pull request. > > > On Mon, Jun 16, 2014 at 5:39 AM, Javier Puerto <[email protected]> wrote: > > > Hi Cordova developers, > > > > I'm creating a system to download/update several resources from a server > to > > the device and I've observe a behaviour that breaks my use case. > > > > After fix the issue CB-6928, I'm able to download/update all the > resources > > without problems. My next test was to try to download the resources but > > with no server response (stopped). I've found that the plugin is deleting > > my target file silently because there's was an error. > > > > I think that the developer should be the responsible to delete or leave > the > > "corrupted" file because the file-transfer plugin already communicates > the > > error, I don't think that the hardcoded behaviour is a good solution. > What > > do you think? > > I can open a new issue and provide the patch and test case (Android only) > > > > Best regards. > > >
