You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This commit was created on GitHub.com and signed with GitHub’s verified signature.
Added
qso_graph_lib.adif: ADIF's ADI format, read and written as ADIF 3.1.7 defines it
(section IV.A): read, iter_records (streaming), read_file and write. One
implementation for lotw-mcp, eqsl-mcp, qrz-mcp, adif-mcp and qsp-client, whose six parsers
disagreed: the same input dropped a field in one, merged the Header into the first QSO in
another, and ran two records together in five of six.
Passes ADIF 3.1.7's own test files: all 6,197 QSOs of ADIF_317_test_QSOs read exactly
as its ADX twin has them, with no problem reported. The files are in tests/adif317/,
checked by SHA-256, and passing them is the floor.
Non-compliant input that real logs contain (a length counted in UTF-8 bytes, a length that
runs past the next tag, a Header starting with <, a last record without <EOR>) is
repaired where it can be and reported in Document.problems, never silently.
read_file reads without newline translation. Text mode turns CR LF into LF and breaks
every multi-line field's declared length: four fields in ADIF's test file.
QRZ's HTML-escaped markers (<call:6>) are read with escaped_markers=True,
without rewriting the values.
write checks field names and always opens a Header with text, as a Header that begins
with < is, by the specification, no Header.
py.typed: the package says it is typed, so mypy checks the packages that use it
against its types instead of treating every import as Any.