release-1.0.2

• Rearrange the codebase: gnparser core does not depend on JSON rendering library
• Fix 463045a: minor issues with multiple year node rendering
• Fix cfd9b1d: JSON snake_case is converted to camelCase
• Fix 86fe367: spray-json is used for rendering, that drops redundant conversion in Akka-based servers of runner
• Fix #381: cover TcpServiceConnection with Spec
• Fix #471: Limit the size of input stream for CLI to 500K
• README is synchronized with the project current state

release-1.0.1

Fix #468, #469: handle out-of-range positions tracking

release-1.0.0

• Fix #463: the bug with positioning in the result JSON
• Fix #438, #439: redirect service information to STDERR to ease the parsing from STDOUT
• Fix #465: introduce json_pretty format flag
• Fix #466: make file completely default command
• minor bug fixes

release-0.4.5

• Fix #460: correctly preparse <i> tag
• Fix #455, #435: correctly preparse of and not
• minor bug fixes

release-0.4.4

• Fix #428: correctly parse nomenyae
• Fix #429: don't parse ex as infraspecies
• Fix #430: parse nvar. rank
• Fix #431: parse "del" as part of authorship
• Fix #432: fix author parsing (e.g. OS-2017)
• Fix #433: revert punctuaiton parsing
• Fix #434: support "emend." and "emend"
• Fix #435: parse "& al." as a variant of "et al."
• Fix #436: parse "apud" ("with") as a possible delimiter between authors
• Fix #437: make filius less ambigous and normalise it to "fil." instead of "f."
• minor bug fixes

release-0.4.3

• Fix #396: don’t parse forbidden strings
• Fix #403: fix viruses patterns
• Fix #404: handle ‘le’ as author prefix
• Fix #402: don’t parse name with ‘of’ word inside (#417)
• Fix #248: handle sp. correctly (#419)
• Fix #405: handle ht and hort
• Fix #415: don’t parse spec. and nov spec (#424)
• Fix #408: fix virus pattern (#423)
• Fix #399: fix handling numbers in beginning of word
• Fix #400: don’t parse epithets with ‘.’
• Fix #401: add more ranks and two-letter abbreviations
• Fix #426: cleanup redundant comparison
• improve speed of Util object
• bugs fixes (#380)

release-0.4.2

• provide EmptyUuid for empty string in GN namespace
• support emend in name (#382)
• correctly parse str. for strain (#384)
• parse -x in species names (#385)
• names with not in uninomial/genus are not parsed (#379)
• restrict abbreviated genera to 2 letters (#245)
• minor optimisations, bug fixes (#380)

release-0.4.1

• support dash in specific epithet (#361)
• migrate to CircleCI 2.0
• minor bug fixes and more test coverage

release-0.4.0

• support bacteria names parsing (#322)
• support hyphenated genus name with capitalized part (#320)
• support Oe unicode character (#321)
• handle lowercase subgenus (#328)
• support input from STDIN and output to STDOUT (#346)
• add el as author's prefix (#305)
• support author with 't (#252, #255)
• pv. (for pathovar.) is parsed as rank (#353)
• socket usage (Ruby example) documentation is improved (#355)
• parse von dem and v. (for von) as author prefix (#249, #329)
• names with "satellite(s)" in the end are not parsed as viruses (#246)
• handle variety for rank (#233)
• all examples are tested during CircleCI build
• parsing bugs fixes (#332, #330, #327, #326)

release-0.3.3

• add optionally showing canonical name UUID
• API change: no more AST node ID
• add year range to ast node
• improve benchmarks
• parse names ending on hybrid sign (#88)
• fix: sometimes warning ids are broken (#147)
• hybrid abbreviation expansions (#310)
• raw hybrid formula parsing (#311)
• move to CircleCI
• minor improvements