download-as-tar-gz becomes 2.4x faster in docker
segfaults on windows, so don't use it there
does not affect fedora or gentoo,
since zlib-ng is already system-default on those
also adds a global-option to write list of successful
binds to a textfile, for automation / smoketest purposes
since this dumb plugin found an actual usecase,
fix the most glaring issue
when nodes overflow from treeul into treepar, the
eject-button is cloned over as well, but the clone
does nothing (as expected), though this will also
cause a flood of new eject-buttons appearing, and
that's worth fixing
NB: check treeul + treepar explicitly; avoid docul
adds a third possible value for the `replace` property in handshakes:
* absent or False: never overwrite an existing file on the server,
and instead generate a new filename to avoid collision
* True: always overwrite existing files on the server
* "mt": only overwrite if client's last-modified is more recent
(this is the new option)
the new UI button toggles between all three options,
defaulting to never-overwrite
* support globbing/wildcards on windows
* add `osc 9;4` to show upload progress in the taskbar
(currently windows-only; linux is picking it up)
* workaround msys2-terminal not normalizing
absolute paths which contain whitespace
* show a helpful "now hashing..." while the
first file is being hashed, since it kinda
looks like a deadlock on windows otherwise
previously, the biggest file that could be uploaded through
cloudflare was 383 GiB, due to max num chunks being 4096
`--u2sz`, which takes three ints (min-size, target, max-size)
can now be used to enforce a max chunksize; chunks larger
than max-size gets split into smaller subchunks / chunklets
subchunks cannot be stitched/joined, and subchunks of the
same chunk must be uploaded sequentially, one at a time
if a subchunk fails due to bitflips or connection-loss,
then the entire chunk must (and will) be reuploaded
newlines, invalid utf8, and worst of all... %20 (whitespace)
due to up2k protocol limitations,
filenames are normalized when they hit the server,
but folders get to keep their intended jank
previous approach:
* cache 64K on first read
* cache 1M on subsequent intersecting reads
new approach:
* cache 64K on first read
* cache 1M on the next intersecting read
* cache 8M on subsequent intersecting reads
* cache 4M on standalone reads at offsets >1M
improves performance by 50% on windows
and should help on high-latency connections
* wark landed in the wrong registry when moved to another volume
(harmless; upload would succeed on the next handshake)
* dedup did not apply correctly when moved into another volume,
since all the checks were done based on the previous vol;
fix this by recursing the whole thing
also update the reloc example after some real-world experience
Reported-by: @daniiooo
