Sec 41 - #166
Merged
Merged
Conversation
Encoder: detects objects of uniform objects (>= 2 entries, uniform columns incl. nested field groups), emits keyed headers 'key[N:<delim?>]{fields}:' with entry rows 'entrykey: c1,c2…' in first-entry field order; keyless root form '[N:]{fields}:' (§5); keyed header on the hyphen line for the first field of a list item (§10); entry keys per §7.3, cells quoted per §7.
Decoder: scan-based keyed-header matcher in Headers (balance-counting, quoted keys, optional delimiter after the keyed marker), KeyedObjectDecoder with strict entry-count/duplicate checks and inline-content guard, keyed dispatch in KeyDecoder/ObjectDecoder/ListItemDecoder/ValueDecoder; hyphen-line keyed entries at document depth + 3.
Also drops the single-row tabular exemption: size-1 arrays with uniform columns now use tabular form per §9.3 (updates JToonTest expectation).
support nested field groups in tabular arrays (§9.3)
Decoder: parse header field lists as a tree (FieldNode) and assign
cells depth-first pre-order; match keyed array headers with a
balanced-brace scanner instead of the flat regex.
Encoder: derive a TabularField tree for uniform nested object columns,
format nested field groups in headers, and write row cells depth-first.
read indentSize option from conformance fixtures
The conformance harness mapped only 'indent', but spec 4.1 fixtures use
'indentSize'; the option was silently ignored, so custom-indent tests ran
with the default indent of 2. Rename the model fields accordingly.
enforce header validity, keyed tabular and fatal-scope rules (§5, §6, §7, §9)
- restrict unquoted keyed-array keys to non-whitespace, non-colon tokens (§7.3)
- reject keyless array headers outside document root / list-item plain form (§5, §6)
- reject duplicate field names in strict-mode tabular headers (§9.3)
- fall back to key-value lines on tabular delimiter mismatch in non-strict mode (§6)
- throw fatal on bare token lines and characters after a closing quote in both
strict and non-strict mode (§5.2, §7.4) via new FatalDecodeException
- conformance failures: 56 -> 43
strip full-line comments before structural interpretation (§5.1)
- ValueDecoder.decodeInternal: a line whose first non-space character
is '#' (U+0020 only) is a full-line comment and is discarded before
any structural interpretation (D1).
- Only U+0020 spaces may precede '#': a tab before '#' disqualifies
the line, and '#' anywhere else (quoted values, list items, after a
colon, mid-value) is data, not a comment.
- Comments are exempt from the strict indentation check and do not
count as rows, items, or scope terminators.
- A document of only comments and blank lines decodes to an empty
object.
feat(spec-4.1): enforce scope rules, blank-line and list-item handling (§5, §9, §12, §14)
Decoder:
- ValueDecoder: strip trailing spaces per line (§12); quoted strings
keep their trailing spaces. Reject leftover content after the root
form (D7, §5/§5.2): any unconsumed line errors in strict mode,
bare-token lines error in both modes.
- DecodeHelper: add validateNoTrailingContent() implementing the
root-span rule.
- ArrayDecoder: accept blank lines between header and first item
(§12); classify a bare "-" (no trailing space) as a list item
(D11, §9.2); accept "[]" as an empty array header (§9.1/§9.2);
array length validation now applies in strict mode only, the
declared length never truncates a scope (D13, §14.1).
- TabularArrayDecoder: accept blank lines between header and first
row (§12) and trailing blank lines at end of file (CRLF/§12);
over-indented lines after rows throw in strict mode (§14.2).
- ObjectDecoder/KeyDecoder: inline-primitive fields do not open a
scope - a deeper following line is over-indentation and throws in
strict mode (§14.2); skipped in lenient mode (D6).
- ListItemDecoder: over-indented field lines throw in strict mode
(§14.2).
Tests:
- Update legacy unit tests that asserted the pre-4.1 "silently skip
over-indented lines in strict mode" behavior to run in lenient
mode, where skipping remains the correct behavior.
align quoting rules and number grammar with spec 4.1
- Encoder (StringValidator, spec §7.2/§7.3):
- quote strings equal to '#' or starting with '#' (comment marker)
- treat a leading '+' as numeric-like so '+1' is quoted
- restrict unquoted keys to ASCII ^[A-Za-z_][A-Za-z0-9_.]*$, quoting
non-ASCII keys like 'café' or '名前'
- Decoder (PrimitiveDecoder, spec §4):
- enforce the normative number grammar ^-?[0-9]+(?:\.[0-9]+)?(?:e[+-]?[0-9]+)?$
instead of delegating to Java's wider number parsers; tokens like
'.5', '1.', '+1' or 'NaN' now decode as strings
- Decoder (ValueDecoder, spec §5.1):
- strip a single leading U+FEFF byte-order mark before any processing
- build.gradle: nullaway 0.13.7 -> 0.13.8 - gradle/verification-metadata.xml: regenerate 0.13.8 sha256 checksums, drop 0.13.7 block - ArrayDecoder: extract matcher group index magic number to FIELDS_GROUP_INDEX, literal-first equals - PrimitiveDecoder: group static imports with java.util imports - TabularArrayDecoder: convert FieldNode class to record, varargs cell cursor, avoid assignment-in-operand - TabularArrayEncoder/KeyedObjectEncoder: uniformColumnsOf returns Optional instead of @nullable list - ValueDecoder: replace escaped unicode BOM literal with BOM_CHARACTER constant - KeyDecoder/ObjectDecoder: wrap overlong parseKeyedTabularObject lines - KeyDecoderTest: fix indentation and line length violations
split parseFieldList state machine into focused helpers split parseKeyedTabularObject into per-validity helpers split parseListItem and parseListItemFields into focused helpers share nested-scope parsing between field and key values split parseArrayWithDelimiter and parseDelimitedValues split decodeInternal into focused root-parse helpers
split unicode escape validation and decoding into helpers split scanHeader into phase-scoped scanner helpers
Contributor
Code Coverage
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Linked Issue
Closes #
Description
Type of Change
Changes Made
SPEC Compliance
Testing
Pre-submission Checklist
Breaking Changes
Additional Context