v0.2.1

bump stream-vbyte for performance annotating
fix #43 where output vcf would have Number=, unless number was specified in config.json

v0.2.0

skip non ACGT alts (@migbro #39)
grab description from input vcf and propagate to annotated VCFs (@dmiller15, @migbro #34, #35)

v0.1.9

report missing float as . (thanks @dmiller15 #32)

v0.1.8

better help message indicating how to output bcf/vcf.gz (thanks @dvg-p4)
fix missing value in cadd example (thanks @dvg-p4)
doc types (thanks @enormandeau)
handle big files (#30)

v0.1.7

add wiki page and more information about decomposing variants
clarify warning on non ACGT (#23)
better error messages on missing/bad zip file

v0.1.6

change VCF header (#19)
output fewer warnings for '*' and other alternate alleles (#22)

v0.1.5

fix build on OSX (#18 from @sstadick)
handle empty alts (#16 from @sstadick)
add script for converting dbNSFP -> VCF -> echtvar encode

v0.1.4

fix bug with filter not removing (not filtering) variants correctly.

v0.1.3

exit with error on multi-allelics (previously, only first allele was used) ( #11)
more docs on conf file (thanks @m-pauper)
fix compression error for very dense regions (#12)
fix error when using hts_set_threads more than once and after reading header (#12)

v0.1.2

fix #8. bug introduced for string values.

v0.1.1

support multiple echtvar files to echtvar anno. This makes it possible to annoate with, for example gnomad and dbsnp.
support string fields. This works for fields with a lowish number (max of thousands) of unique values.
support extracting FILTER field from annotation VCF. So now can annotate with, e.g. gnomad FILTER field via:
```
{"field": "FILTER", "alias": "gnomad_filter"},
```
in the json file for echtvar encode. "FILTER" is a special-case, all other fields are from the INFO.