Allow attribute get on root#7873
Conversation
earlephilhower
left a comment
earlephilhower
left a comment
There was a problem hiding this comment.
I'm worried that when _valid is false the filename is not guaranteed to be proper. Has this been verified?
Also, the ::format method should be setting these attributes when on-device formatting occurs.
|
The The _valid thing still sets off alarms. You're assuming that dirent is all 0'd (which it is, luckily, in the constructor...but only until something writes to it which can happen on rewind() or next() or other calls). Also, this would return the FS date no matter what the subdir it's in which is not what you're going for. Currently, also, if I read the sprintf logic you're doing a lfs_get_attr("") and not lfs_get_attr("/") |
qistoph
force-pushed
the
pr_root_attributes
branch
from
c384cbf to
9a3f357
Compare
earlephilhower
left a comment
earlephilhower
left a comment
There was a problem hiding this comment.
Good way to add in the call! Just some minor things and we should be good to go!
qistoph
force-pushed
the
pr_root_attributes
branch
from
9a3f357 to
8b6eb00
Compare
qistoph
force-pushed
the
pr_root_attributes
branch
from
8b6eb00 to
869f066
Compare
When `FS::getCreationTime` isn't implemented in the FS, return 0 not -1, same as other getXXXtime() calls in the File object.
Rebuild entire toolchain instead of manually hacking the tools JSON to ensure repeatability. New mklittlefs sets the new FS timestamp added in esp8266#7873
Rebuild entire toolchain instead of manually hacking the tools JSON to ensure repeatability. New mklittlefs sets the new FS timestamp added in #7873
2 participants
This PR enables attributes on the root, such as creation/modification time of volume.
After quite some testing I believe the attached changed is the only one required. It didn't seem to break anything in my tests, though I'm not 100% sure, because the use and value of
_validin undocumented and unclear to me.Test project available: LittleFSTest
Effects of checking 'c' and 't' attributes on a root node, if:
I'm haven't been able to figure out why situation 3. doesn't respond like 2. If this is an undesirable side effect, I'm hoping someone has a suggestion on fixing it or where to look.