Closed StefanKarpinski closed 3 years ago
Merging #104 (cffb931) into master (82cc7dd) will decrease coverage by
0.10%
. The diff coverage is100.00%
.
@@ Coverage Diff @@
## master #104 +/- ##
==========================================
- Coverage 96.37% 96.26% -0.11%
==========================================
Files 4 4
Lines 634 643 +9
==========================================
+ Hits 611 619 +8
- Misses 23 24 +1
Impacted Files | Coverage Δ | |
---|---|---|
src/Tar.jl | 96.05% <ø> (-1.32%) |
:arrow_down: |
src/create.jl | 97.05% <100.00%> (ø) |
|
src/extract.jl | 96.34% <100.00%> (+0.07%) |
:arrow_up: |
src/header.jl | 94.20% <100.00%> (+0.17%) |
:arrow_up: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update 82cc7dd...cffb931. Read the comment docs.
Allowing the
Tar.list
callback to receive the raw header data (especially in the form of a vector of fields) is highly useful for analyzing wild tarballs. I implemented this when working on adding support for hard links so that I could look at a tarball containing hard links more easily.Adding a
Header
method that takes an existingHeader
object and modifies some of its fields is a pattern that's used in various places in the code base which this method makes cleaner and clearer.This makes all cases where the end of a tarball is encountered before it is expected throw the same
EOFError
whereas previously some places threw that error while others threw a generic error with a premature end of file message.This changes also makes the definition of a
skip
method for processes conditional so that when testing with a Julia version that hasTar
as a stdlib (and thus already has this method defined), we don't break precompilation by redefining that method.