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
JSONPointer decodes string tokens in its constructor, then decodes them again during object lookup. For example, /~01 incorrectly selects the / member instead of the literal ~1 member. Builder and token-list pointers also misinterpret literal tokens containing ~0 or ~1.
Use the stored token directly in queryFrom. Add regressions for string and URI-fragment pointers, builders, and token lists with distinct colliding keys. This follows the decoding order in RFC 6901 section 4 and addresses the double-unescaping behavior mentioned in #561.
Validation: the three new tests fail before the fix. The JDK 8 and 21 suites and the JDK 21 strict-mode profile each report 801 tests, zero failures/errors and six existing skips. Package and Javadoc generation pass, as does Java 6 source/target compilation.
@jakezwang Thanks for the PR, good catch and a good fix. Apparently, fixing this issue exposes a pre-existing bug in toURIFragment(), for which I will add a separate issue.
What problem does this code solve?
Fixes an issue in JSONPointer where double-unescaping is performed incorrectly. This bug has been in the code since the initial commit.
Risks
Moderately low, for users who have been working around this issue.
Changes to the Existing Behavior
This fix includes a behavior change: users who have been aware of the problem and worked around it may find that their code no longer works. This is considered an acceptable change. Implementing the fix also exposes another pre-existing bug, which will be fixed elsewhere.
stleary
changed the title
Avoid decoding JSON Pointer tokens twice
Decode JSONPointer tokens once only
Sep 20, 2026
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
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.
stleary
changed the title
JSONPointerdecodes string tokens in its constructor, then decodes them again during object lookup. For example,/~01incorrectly selects the/member instead of the literal~1member. Builder and token-list pointers also misinterpret literal tokens containing~0or~1.Use the stored token directly in
queryFrom. Add regressions for string and URI-fragment pointers, builders, and token lists with distinct colliding keys. This follows the decoding order in RFC 6901 section 4 and addresses the double-unescaping behavior mentioned in #561.Validation: the three new tests fail before the fix. The JDK 8 and 21 suites and the JDK 21 strict-mode profile each report 801 tests, zero failures/errors and six existing skips. Package and Javadoc generation pass, as does Java 6 source/target compilation.